American Canyon City Council
February 21, 2023

Item Name E-Mail
Requested
To Speak Comment

11.) Giovannoni Logistics Center Warehouse Project Gary Margadant [email protected] No Do more warehouse projects Environmentally align with the future of American Canyon Families, providing a safe and healthy location for housing, education, community, neighborhoods, jobs and outdoor activities? Â Certainly there are benefits yet they come with a cost against these family goals as discussed in the Environmental Impact Report on today's Agenda. Â A community in Southern California has a tale to tell how a Warehouse boom transformed their Inland Empire. Are jobs worth the environmental degradation? Â https://www.latimes.com/california/story/2023-02-05/warehouses-big-rigs-fill-inland-empire-streets. Â American Canyon families have a similar but different perspective with their proximity to the Napa River, SF Bay and open land for family hiking, wildlife experience and education. Â Napa County designed the warehouse distict around the airport as an industrial zone to concentrate Wine Industry logistics away from the up valley Wineries, communities and vineyards, reducing heavy truck traffic in picturesque locations. Â Does the EIR for this warehouse proposal adequately describe the impacts and tradeoffs the AC families experience upon approval? Should a Wildlife conservation easement become a part of this project as proposed by another comment? Â This would allow AC to include their own family design on this and any future projects.

11.) Giovannoni Logistics Center Warehouse Project Geena Saavedra [email protected] No Please reconsider this plan. We already have too many warehouses in American Canyon and too little open space as it is. The open space we do have is what attracts people to this area because of its natural beauty that is so rare nowadays. Increasing the amount of warehouses in this city would contribute to higher amounts of pollution and drive away the natural habitat. This would harm the ecosystem and destroy what makes American Canyon unique.

11.) Giovannoni Logistics Center Warehouse Project Valerie Zizak-Morais [email protected] No February 21, 2023 The American Canyon Chamber of Commerce strongly endorses the Giovannoni Logistics Center Development and we urge you to approve the applications before you tonight that will move this project forward. There are currently limited vacancies for wine storage. Green Island Road is the Champion of the Wine Industry with millions of cases of wine passing through it every year. We need to support this growing industry, encourage wine supporting businesses, to be in American Canyon, and welcome them as they locate and grow here. If the businesses don't come to American Canyon, they go into South Unincorporated County or Sonoma, meaning no benefits to our City. As the Chamber, we represent our members, and the business community of American Canyon. We believe that the Giovannoni Logistics Center Project needs to be approved because it will provide the following economic development benefits for our community:

1. Promote development that generates net positive tax revenues for the City by generating more in new tax revenues than are consumed by City expenditures on services provided to the development.

2. Create new employment opportunities for residents of Napa County and the surrounding region.

3. Develop compatible land uses near the Napa County Airport in the interests of avoiding interference with aviation operations.

4. Improve American Canyon’s jobs-housing ratio by adding new employment opportunities.

5. Continue the orderly development of the Devlin Road corridor with a well-designed project.

6. Further the goals and policies of the City of American Canyon General Plan by developing land contemplated to support urban development to its highest and best use.

7. Preserve the most biologically sensitive portions of the project site as open space.

8. Install circulation improvements along Green Island Road and Devlin Road that provide efficient ingress and egress to the Proposed Project while also ensuring these facilities operate at acceptable levels.

9. Promote public safety by incorporating security measures into the project design.

10. Mitigate impacts on the environment through implementation of feasible mitigation measures.

We appreciate your consideration and look forward to having the Giovannoni Logistics Center Project welcomed in American Canyon

11.) Giovannoni Logistics Center Warehouse Project Beverly Saxon Leonard [email protected] No Mayor & Council Members:

As you consider the Giovannoni Site Project tonight, I ask that you give full consideration to all impacts the project will engender.

Of particular concern is the detrimental effects to wildlife and specifically, to the raptor population.

The site holds important wetlands and grasslands resources necessary for the continued existence of the Swainson's Hawk, a threatened species that is unique to Southern Napa County.

American Canyon wetlands & grasslands possess the best habitat for these species which also includes Golden Eagles and other raptors The project proposes a small wetlands mitigation however it does nothing to mitigate the loss of important grasslands necessary for the continued existence of raptor populations. The proposed project will significantly contribute if not ensure the loss of a healthy ecosystem.

The plan's mitigation of wetlands & grasslands loss is flawed & inadequate. Please consider the larger picture in your decision tonight.

On a separate note, I would like to address the manner the public has been notified concerning this project. Apparently, notification has been via Vallejo publications & not Napa County publications. Although not illegal, the fact is the majority of people who would be concerned about the impacts of this project would be Napa County not Solano County residents. As such, you may not be receiving important input from all concerned parties. In the future, please consider where and how you provide notice so that it is actually accomplishing its purpose.

Thank you for your attention.

Beverly Saxon Leonard

11.) Giovannoni Logistics Center Warehouse Project Diann Berryhill [email protected] No We already have to much traffic. And the damage that continues by all the building to our local wildlife is heartbreaking . The more that is built then out tax base will need to be higher. Stop the building and keep our tax base lower. Less building means less police and emergency services.

11.) Giovannoni Logistics Center Warehouse Project Jerry Hoffman [email protected] No I think we already have too many warehouses. They don't bring in any revenue other their property tax and the semi's pollute the air, congest traffic, and destroy the roads.

Either leave the land open for the wildlife to live in and people to enjoy or turn the land in a large park and recreation area for all to enjoy. NO MORE WAREHOUSES!

11.) Giovannoni Logistics Center Warehouse Project Sharon Robinson [email protected] No When is enough enough for American Canyon? We should be preserving land, not giving it to conglomerates who have no real interest in our City. The wine industry seems to have some in their back pockets, as their projects are always approved. We should be demanding MORE city improvements before they receive approval. If these Corporations were all forced to invest into a fund, the City could REHAB HIGHWAY 29.

11.) Giovannoni Logistics Center Warehouse Project Jason Kishineff [email protected] No American Canyon has more than enough warehouses. Enough is enough. We can do SO much better than more warehouses. What are we? The storage space for Saint Helena and Calistoga? I say NO MORE! The residents of this city have been demanding, for YEARS, businesses that improve the lives of our residents and all we seem to get are warehouses and gas stations. Why aren't we seeking any other alternatives? We just keep doing the same thing again and again, ignoring the complaining residents and transferring wealth out of our city and to the north valley. Take a stand! Please!

11.) Giovannoni Logistics Center Warehouse Project Jeannette Goyetche [email protected] No I am requesting that the city council members reconsider the building of more warehouses along Green Island Rd, and Devlin Rd. Please consider the negative impact on A.C residents and our precious fragile environment. With more warehouses, there will be more traffic,more air pollution from diesel trucks, increased road deterioration, negative impact on the wildlife. I think that a community group of A.C residents could work with the council to discuss ideas on how this land would benefit both A.C residents and the environment as well. I do not see the positive aspects of the Giovannoni Project for American Canyon.

11.) Giovannoni Logistics Center Warehouse Project Yvonne Baginski [email protected] No The annihilation of 147.5 acres of foraging grasslands for the threatened Swainson Hawk, Golden Eagle and other birds, needs more attention. All the property north of Hwy 29 to Green Island Rd. is zoned commercial/industrial. When developed in the next few years, the loss of habitat will cause the extinction of the Swainson Hawk in south Napa County, the only place in Napa County where it is currently seen. The Giovannoni Project alone will wipe out 7% of the habitat. We need a sober look at this realityi and develop a strategy to lessen the impact of this project. We need to create a biologically diverse conservation easement for the life of the remaining wildlife in south county. This project would be a start. Scientists from the California Dept. of Fish and Wildlife, and citizen activists from the Friends of the Swainson Hawk, the Center for Biological Diversity and Raptors are the Solution, as well as residents of American Canyon, have written letters asking the City of American Canyon to consider setting aside foraging lands for wildlife, as well as appointing a credible manager to monitor compliance. Caring for ecosystems demands far-sightedness, since no one looking for a quick and easy profit is truly interested in their preservation. The cost of damage caused by a selfish lack of concern is much greater than the economic benefits to be obtained. Where certain species are destroyed or seriously harmed, the values involved are incalculable. I implore you to reconsider this effort in terms of doing the least harm to the environment. Please consider the creation of a wildlife conservation easement of foraging grasslands/wetlands, much larger and realistic than what is currently proposed. I also ask that the easement be monitored by a credible manager, the developed property is landscaped with native plants conducive to this region and that poison bait stations to kill small mammals be banned. This first step will establish a strong leadership role for the other cities, and Napa County in demonstrating sustainable possibilities for all future development.

5188-015acp
KEVIN T. CARMICHAEL
CHRISTINA M. CARO
THOMAS A. ENSLOW
KELILAH D. FEDERMAN
RICHARD M. FRANCO
ANDREW J. GRAF
TANYA A. GULESSERIAN
RACHAEL E. KOSS
AIDAN P. MARSHALL
TARA C. RENGIFO
Of Counsel
MARC D. JOSEPH
DANIEL L. CARDOZO

SACRAMENTO OFFICE
520 CAPITOL MALL, SUITE 350
SACRAMENTO, CA 95814-4721
TEL: (916) 444 - 6201
FAX: (916) 444 - 6209

ADAMS BROADWELL JOSEPH & CARDOZO A PROFESSIONAL CORPORATION

ATTORNEYS AT LAW
601 GATEWAY BOULEVARD, SUITE 1000
SOUTH SAN FRANCISCO, CA 94080 - 7037
___________
TEL: (650) 589 - 1660
FAX: (650) 589 - 5062
rfranco @adamsbroadwell.com
printed on recycled paper
February 21, 2023
Via Email and Overnight Mail
Mayor Leon Garcia
Vice Mayor Pierre Washington
Honorable Members of the City Council
City of American Canyon City Hall
4381 Broadway Street
American Canyon, CA 94503
Email: [email protected];
[email protected]
[email protected];
[email protected];
[email protected];
Brent Cooper, AICP
Community Development Director
William He, Associate Planner
Nicole Jones, Administrative
Technician
Community Development
Department
City of American Canyon
4381 Broadway Street, Suite 201
American Canyon, CA 94503
[email protected];
[email protected];
[email protected]
Re: Agenda Item #11 February 21, 2023 City Council meeting-- Giovannoni Logistics Center Project proposed by Buzz Oates Construction, Inc. (PL20-0042, PL20-0043 and SCH #2021010104) Dear Mayor Garcia, Vice Mayor Washington, Honorable City Council Members, Mr.

Cooper, Mr. He and Ms. Jones:

We are writing on behalf of Napa/Solano Residents for Responsible Development (“Napa/Solano Residents”) to provide comments on the Giovannoni Logistics Center Project, PL20-0042, PL20-0043 and SCH #2021010104, (“Project”) The Project appears as Item No. 11 on the agenda for the February 21, 2023 City of American Canyon (“City”) City Council (“Council”) hearing. The Council will consider adopting resolutions to (1) certify the Final Environmental Impact Report (“FEIR), adopt CEQA findings of fact, adopt a Statement of Overriding Considerations and adopt a Mitigation Monitoring and Reporting Program; (2) approve a design permit for Phase 1 of the Project; and (3) approve a Tentative Parcel Map for Phase 1 of the Project.

February 21, 2023

5188-015acp
printed on recycled paper
The City, as lead agency under the California Environmental Quality Act1 (“CEQA”), prepared the Draft Environmental Impact Report (“DEIR”)2 and Final Environmental Impact Report (“FEIR”) for the Project. Napa/Solano Residents’ July 2022 comments on the DEIR (including expert technical comments)3 explained how the DEIR failed to comply with CEQA’s basic requirement to act as an “informational document,” in that it lacked meaningful details in critical areas, such as air quality, health risk, biological resources including wetlands and habitat, and greenhouse gas (“GHG”) and transportation impacts, without which the public and decisionmakers are unable to adequately assess the Project’s significant impacts.

Residents’ comments further explained that, because of the DEIR’s shortcomings, it was deficient as a matter of law because it failed to properly disclose and mitigate the Project’s potentially significant impacts.

The City released its original FEIR in November 2022.4 FEIR Vol. 1 contains responses to Napa/Solano Residents’ DEIR comments and attempts to address some of the issues raised. As discussed in our November 17, 2022 letter to the Planning Commission5 and December 6 letter to the City Council6, however, FEIR Vol. 1 failed to meaningfully respond to many of Napa/Solano Residents’ DEIR comments and failed to resolve or mitigate many of the Project’s potentially significant impacts. In particular, FEIR Vol. 1 failed to analyze or even mention the construction of a new gravel road that runs directly through the Project’s proposed 45-acre open space/wetlands preserve, an area intended to serve as mitigation for the Project’s wetlands impacts. The Planning Commission’s recommendation that the Council certify the FEIR and approve the Project in reliance on FEIR Vol. 1 was not based on substantial evidence.

At the December 6, 2023 City Council hearing to consider approval of the Project’s FEIR and other entitlements, the City Council continued the hearing to February 21, 2023. On February 10, 2023, the City released an update to the FEIR 1 Pub. Resources Code (“PRC”) §§ 21000 et seq. 2 City of American Canyon, Draft Environmental Impact Report-Giovannoni Logistics Center Project (May 20, 2022) (hereinafter, “DEIR”).
3 July 20, 2022 letter from Richard M. Franco to Brent Cooper and Nicolle Jones re Comments on the Draft Environmental Impact Report – Giovannoni Logistics Center Project proposed by Buzz Oates Construction, Inc. (“DEIR Comments”). 4 City of American Canyon Final Environmental Impact Report- Giovannoni Logistics Center Project (October 31, 2022) (“FEIR Vol. 1”).
5 November 17, 2022 letter from Richard M. Franco to City of American Canyon Planning Commission re Agenda Item #3-Comments on the Draft Environmental Impact Report – Giovannoni Logistics Center Project proposed by Buzz Oates Construction, Inc. (“11/17/22 FEIR Comments”). 6 December 6, 2022 letter from Richard M. Franco to City of American Canyon City Council re Agenda Item #17 Comments on the Draft Environmental Impact Report – Giovannoni Logistics Center Project proposed by Buzz Oates Construction, Inc. (“12/6/22 FEIR Comments”).

February 21, 2023

5188-015acp
printed on recycled paper
which contains responses to comments received by the City after release of FEIR Vol. 1, including responses to our 11/17/22 FEIR Comments and 12/6/22 FEIR Comments.7 We reviewed FEIR Vol. 2 and available supporting documents with the assistance of our expert consultants, including wetlands biologist Sarah Corbin and wildlife biologist Scott Cashen, whose comments on the FEIR are attached hereto as Exhibit A8 and Exhibit B9, respectively. As detailed below, the FEIR continues to fail to adequately address the deficiencies under CEQA that were raised in our prior comments, and the Project’s EIR remains legally inadequate.

Specifically, the FEIR fails to resolve issues including (1) impacts of the new gravel access road constructed across the Project site, (2) impacts to biological resources and (3) impacts to wetlands.

An EIR must be prepared with a sufficient degree of analysis to provide decision makers and the public with the information needed to make an intelligent judgment concerning a project’s environmental impacts, and must provide a reasonable, good faith disclosure and analysis of those impacts to ensure that all potentially significant impacts are reduced to the greatest extent feasible.10 The EIR for this Project fails to meet these central requirements of CEQA for all of the reasons set forth herein and in each of our previous comment letters. Accordingly, the City must prepare and circulate a revised EIR.

Napa/Solano Residents respectfully requests that the City Council continue this hearing and remand the Project to staff to prepare a revised DEIR for public comment. The Project should not return to Council for further consideration until the City has fully complied with its obligations under CEQA and all applicable land use laws.
7 City of American Canyon Final Environmental Impact Report—Volume 2- Giovannoni Logistics Center Project (February 7, 2023) (“FEIR Vol.2”). FEIR Vol. 1 and FEIR Vol. 2 are hereinafter collectively sometimes referred to as the “FEIR.”
8 Exhibit A- February 19, 2023 Letter from Sarah Corbin to Richard M. Franco re Wetland Review of Final Environmental Impact Report-Volume 2, Giovannoni Logistics Center, City of American Canyon, California (“Corbin FEIR Vol. 2 Comments”).
9 Exhibit B- February 18, 2023 Letter from Scott Cashen to Richard M. Franco re Comments on Volume 2 of the Final Environmental Impact Report for the Giovannoni Logistics Center Project (“Cashen FEIR Vol. 2 Comments”).
10 14 CCR § 15151; see also, Laurel Heights Improvement Assn. v. Regents of Univ. of California (1988) 47 Cal.3d 376, 392, and Napa Citizens for Honest Gov’t v. Napa County Bd. of Supervisors (2001) 91 Cal.App.4th 342, 356.

February 21, 2023

5188-015acp
printed on recycled paper
I. STATEMENT OF INTEREST
Napa/Solano Residents is an unincorporated association of individuals and labor organizations that may be adversely affected by the potential environmental impacts associated with Project development. Napa/Solano Residents includes American Canyon residents James Aken, Samantha Spangler, Evelyn Hernandez, and Alex Somodio, and members of the International Brotherhood of Electrical Workers Local 180, Plumbers & Steamfitters Local 343, Sheet Metal Workers Local 104, Sprinkler Fitters Local 483, District Council of Ironworkers and their members and their families, and other individuals that live and/or work in the City of American Canyon and Napa/Solano Counties and would be directly affected by the Project’s environmental and health and safety impacts. Napa/Solano Residents and its members also have an interest in enforcing environmental laws that encourage sustainable development and ensure a safe working environment for the members they represent.

II. THE CITY MUST REVISE AND RECIRCULATE THE EIR
We have previously commented on the numerous ways in which the EIR for the Project fails to comply with CEQA.11 Here, our comments will be limited to issues raised in the Project’s FEIR Vol. 2.

A. The EIR Does Not Adequately Address the Impacts of the New Gravel Access Road.

Just prior to the November 17, 2022 Planning Commission hearing for this Project, we learned that that a new gravel road and large dirt berm had been installed on the Project site, and that the road cuts directly through the area designated as the Project’s “wetlands preserve.” Our 11/17/22 FEIR Comments and 12/6/22 FEIR Comments discuss this issue at length and include images depicting the encroachment.

Until now, the City has never disclosed the gravel road, or discussed its impacts, in any of the CEQA documents for this Project. It was not identified in the DEIR’s Project description or environmental setting discussion, was not addressed in the DEIR’s impacts analysis or mitigation discussion, and did not appear in any of the aerial photos or site diagrams in the DEIR. Nor did FEIR Vol. 1 address this issue. Now, in its FEIR Vol. 2, the City finally has provided some information regarding the road. The unpaved, gravel road “allows for access to the Napa Logistics park sewer and recycled water pipelines that cross the Giovannoni site. 11 See, DEIR Comments, 11/17/22 FEIR Comments and 12/6/22 FEIR Comments.

February 21, 2023

5188-015acp
printed on recycled paper
This utility alignment was disclosed and evaluated in the Napa Logistics Park Phase 2 Final EIR, which the City certified in 2015.”12 The FEIR now also discloses a number of documents relating to the road’s purpose and permitting.13 The City makes a number of arguments as to why it was not required to disclose or analyze the new gravel access road and its impacts in the CEQA analysis for this Project. None of these arguments has merit.

First, the City argues that because the road (and associated pipelines) were disclosed and analyzed as part of the EIR for the neighboring Napa Logistics Park project, it can essentially ignore these changes to the Giovannoni Project site for purposes of CEQA: “While the physical environment was altered during construction of the sewer line and waterline installation through excavation, these environmental impacts were fully vetted under CEQA as part of the approved Napa Logistics Park Phase 2 Project, and the road was constructed in compliance with all conditions outlined in the MMRP for the project…[emphasis added]”14 The City argues that CEQA does not require that the gravel road be “analyzed twice.”15 To the contrary, CEQA plainly requires that “where the physical environment was altered” as it was on the Giovannoni site, the EIR must consider the impacts of such alteration on the environmental setting.16 This is especially true where, as here, the alteration is to land designated to be preserved in perpetuity as mitigation for the Project’s impacts to wetlands. Prior CEQA review of the road in connection with a different project is immaterial. The issue is not whether the road was properly approved and entitled before it was constructed.

Even if the new gravel access road was properly approved and permitted in connection with another project, the issue remains that the road was never disclosed or analyzed in connection with this Project. Such disclosure and analysis are essential because the existence of the road has a direct effect on the City’s wetland impact analysis and wetland mitigation measures proposed in the Giovannoni EIR.

Any member of the public reviewing the EIR for this Project would have had no idea that this road had been constructed across the Project site, because the City never disclosed it until now. Both the DEIR and FEIR Vol. 1 therefore misled the public by presenting maps of the Project site which incorrectly depicted the area covered by the gravel access road as undeveloped land capable of wetland preservation.

Regardless of whether the road was legally constructed or analyzed for a different 12 FEIR Vol. 2, pg. 2-10.
13 FEIR Vol. 2, Appendix M. 14 FEIR Vol. 2, pg. 2-11.
15 FEIR, Vol. 2, pg. 2-12.
16 See e.g., 14 CCR §§ 15064, 15088.5, 15125 and 15162.

February 21, 2023

5188-015acp
printed on recycled paper
EIR, the fact remains that the EIR for the Giovannoni Project never disclosed that a gravel access road would be constructed across the proposed wetland preserve.

As we previously explained, the new gravel road represents significant new information requiring that the EIR be revised and recirculated. Lead agencies are required to recirculate an EIR for further public review and comment if significant new information is added to the EIR after notice of public review but before certification.17 Recirculation is required when the addition of new information deprives the public of a meaningful opportunity to comment on a substantial adverse environmental effect of the project or a feasible way to mitigate or avoid such an effect.18 New information can include “changes in the project or environmental setting as well as additional data or other information.”19 The purpose of recirculation is to give the public and other agencies an opportunity to evaluate the new data and the validity of conclusions drawn from it.20 Recirculation is also required where significant new information surfaces prior to certification of an EIR but the lead agency failed to add that information to the EIR.21 Such information must be added to the EIR—and the EIR must be recirculated—if the information would trigger the recirculation requirement under CEQA Guidelines section 15088.5. Failure to include such information thwarts the goals of the EIR process by not disclosing to the public and government agencies critical information necessary to evaluate the significance of the Project’s impacts.22 The new gravel road is a fundamental change in the baseline conditions at the Project site which affects the EIR’s underlying analysis of the Project’s wetland impacts and reduces the feasibility and efficacy of the EIR’s proposed wetland mitigation. The public and other governmental agencies have been denied the opportunity to evaluate the new information and its significance, in violation of CEQA.

Second, while the City ignores the “significant new information” issue, it argues that it wasn’t required to disclose the road because it was not yet in existence on January 12, 2021, when the Notice of Preparation (“NOP”) for the Project was issued.23 The FEIR cites CEQA Guidelines Section 15125(a)(1) which states that the EIR’s description of the environmental setting should “reflect 17 Pub. Resources Code § 21092.1; 14 CCR § 15088.5.
18 14 CCR § 15088.5(a).
19 Id.
20 Spring Valley Lake Ass’n v. City of Victorville (2016) 248 Cal.App.4th 91, 108. 21 Cadiz Land Co. v. Rail Cycle (2000) 83 Cal.App.4th 74, 95. 22 Id. 23 FEIR Vol. 2, pg. 2-11.

February 21, 2023

5188-015acp
printed on recycled paper
physical environmental conditions as they exist at the time the notice of preparation is published[.]”24 The City’s selective quotation of the Guidelines conveniently omits additional terms that are directly relevant here. While it is true that “generally” the baseline description should reflect conditions at the time the NOP is published, “[w]here existing conditions change or fluctuate over time, and where necessary to provide the most accurate picture practically possible of the project’s impacts, a lead agency may define existing conditions by referencing historic conditions, or conditions expected when the project becomes operational, or both, that are supported with substantial evidence.”25 It is central to CEQA that an EIR’s analysis employs “a realistic baseline that will give the public and decision makers the most accurate picture practically possible of the project’s likely impacts.”26 The EIR fails to fulfill this basic purpose.

While technically true that the gravel road did not exist at the time the Project’s NOP was published, the City’s position is disingenuous because it knew that the physical environment was going to be altered by the pipeline and gravel road at the time NOP was published, and throughout the period while the Draft EIR was being prepared. In fact, the same consultant who prepared the Napa Logistics EIR also prepared the EIR for this Project. Despite actual knowledge that the site conditions were going to change in a specific and meaningful way, the City and its consultants chose to deprive the public of that knowledge by completely omitting it from its CEQA analysis until forced to address it at the eleventh hour in response to public comments disclosing the omission. In order to provide the public and decision makers with the most accurate picture possible of the Project’s impacts, the City should have defined the environmental setting to include the pipelines and road it knew were going to be constructed.27 Because the City failed to do so, the EIR cannot fulfill its key role under CEQA as an informational document.28
Finally, the City now makes efforts to downplay the impacts of the new gravel road by offering a number of belated and unsupported claims to support its contention that the road will not have any significant impacts or effects on the Project’s mitigation plan. This “analysis” is too little, too late, and is not supported by substantial evidence. Among other things, the City claims:
24 14 CCR § 15125(a)(1).
25 Id.
26 Neighbors for Smart Rail v. Exposition Metro Line Construction Authority (2013) 57 Cal.4th 439, 449, citing Communities for a Better Environment v. South Coast Air Quality Mgmt. Dist. (2010) 48 Cal.4th 310, 322, 325, 328.
27 Id. at 454.
28 Id.

February 21, 2023

5188-015acp
printed on recycled paper
• The gravel access road “constitutes a negligible change to the physical environment post-construction.”29 The City does not support this assertion with evidence. As wetland biologist Sarah Corbin explains, the conversion of permeable, herbaceous vegetated upland to the impervious graveled surface is not a “negligible change.”30 “Precipitation will not infiltrate, and the upper layer of soil under the road is compacted either purposely (to construct a stable road) or incidentally (heavy equipment traffic), which could also impact groundwater movement under the road footprint and adjacent compacted areas. Also, if the sewer and water lines that were installed under the road were bedded and or backfilled with gravel, this could also serve as a conduit to redirect groundwater flows.”31 The City has provided no evidence that it has considered these issues.
• It is “unlikely that any existing wetland or other water features have been indirectly impacted...as, given the level topography with depressional microrelief, the wetlands on the project site are primarily fed through vertical precipitation.”32 These statements are unsupported by any evidence, let alone substantial evidence. A discussion of the likelihood of impacts to the existing wetlands should have been included in the EIR, supported by substantial evidence, and available for the public and state agencies to review and offer comments on. As Ms. Corbin has previously commented, the EIR lacks a quantifiable existing water budget for the Project site’s wetlands.33 Without such evidence, conclusions addressing indirect hydrologic wetlands impacts are unsubstantiated.
• The stockpiled dirt berm near the road “is small and does not affect the total hydrology of the area.”34 Again, the City offers no support for this statement.

Whether or not the road and associated dirt berm affected the hydrology of the area should have been analyzed in the DEIR, with any conclusions supported by substantial evidence. This was not done.
• The City concedes that the earth disturbance associated with the graveled access road extends into areas proposed for wetland creation.35 It argues, however, that the wetland mitigation plan is only conceptual, and that the locations of proposed wetlands creation may be shifted to avoid impacting the new utility lines.36 Conceptual or not, the City was required to disclose the earth disturbance when describing the mitigation plan in the DEIR, so that 29 FEIR Vol. 2, pg. 2-11
30 Corbin FEIR Vol. 2 Comments, pg. 2.
31 Id.
32 FEIR Vol. 2, pg. 2-13.
33 Corbin FEIR Vol. 2 Comments, pg. 2.
34 FEIR Vol. 2, pg. 2-13.
35 Id., pg. 2-14.
36 Id.

February 21, 2023

5188-015acp
printed on recycled paper
members of the public and decision makers could evaluate the feasibility and efficacy of the proposed mitigation measures. By completely omitting any discussion of the gravel road and associated earth disturbance, the City failed to “provide the public and decision makers the most accurate picture practically possible of the project’s impacts”37 and proposed mitigation, in violation of CEQA.
• The City maintains that “prior earth disturbance, especially if restricted to the upper soil horizons, is not necessarily a barrier to wetland creation.”38 The City cites its consultant Monk & Associates’ experience in creating mitigation wetlands and concludes that “Monk & Associates is confident that the presence of the gravel road associated with the sewer alignment for the Napa Logistics Park Phase 2 Project will not frustrate the creation of the Napa Logistics Wetlands Preserve [sic] as described in the Draft EIR for the proposed project” as set forth in the mitigation measures.39 Assuming this statement was meant to refer to the Giovannoni Wetlands Preserve, the FEIR supplies no analysis or evidence to justify Monk & Associates’ “confidence.”
Each of these issues should have been analyzed and supported by substantial evidence in the DEIR. The City’s belated attempts to explain away potentially significant impacts are not supported by substantial evidence and do not satisfy CEQA. The City must prepare and circulate for public comment a revised DEIR that reevaluates all potentially significant impacts to the Project site’s biological and wetlands resources in light of the new road feature, as well as its impacts on the proposed wetlands preserve’s ability to adequately mitigate the Project’s impacts.

B. The Project’s Wetlands Impact Analysis and Proposed Mitigation Remains Inadequate.

In our previous comments (including technical comments by wetlands biologist Sarah Corbin), we detailed at length the deficiencies in the DEIR’s wetlands impact analysis and the lack of evidence demonstrating that the proposed wetlands mitigation would be effective. The City’s latest attempts to address these shortcomings continue to be inadequate under CEQA.40
37 14 CCR § 15125(a)(1).
38 FEIR Vol. 2, pg. 2-14.
39 Id.
40 See Exhibit A, Corbin FEIR Vol. 2 Comments.

February 21, 2023

5188-015acp
printed on recycled paper
For example, the EIR continues to lack a formal functional assessment of the Project site’s existing wetlands, which precludes a proper analysis of the Project’s impacts on existing wetlands functions.41 As Ms. Corbin explains, the EIR fails to include key elements of a formal functional assessment, including “organized analyses (not simply observational statements from a limited point in time) related to the wetlands’ hydrology (e.g., water sources, hydroperiods, hydrologic connectivity), buffer and landscape (e.g., buffer width and condition), physical structures typical of vernal pool and seasonal depressional wetlands, plant community interspersion, and assessment of anthropogenic stressors.”42 The City now suggests that this formal functional assessment will be performed at a later date in connection with verification of the performance standards for the Project’s proposed wetlands mitigation measures; it states that it will use a “functional assessment method, such as the California Rapid Assessment Method (CRAM).”43 This is improperly deferred analysis.44 The CRAM represents precisely the type of formal functional assessment Ms. Corbin has repeatedly identified as essential for evaluation of the Project’s wetlands impacts. This analysis must be performed and disclosed now, and included in a revised EIR, so that the public and the City’s decision makers have the opportunity to intelligently evaluate the Project’s wetlands impacts and the efficacy of the proposed mitigation.

C. The Project’s Biological Resources Impact Analysis Remains Inadequate.

Our previous comments, including expert technical comments by Scott Cashen, describe in detail the many deficiencies in the EIR’s analysis of impacts to biological resources. Mr. Cashen’s attached comments explain several ways in which the FEIR fails to remedy these issues.45
First, the FEIR now includes a new botanical survey report which identifies certain vernal pool habitat at the Project site as “a sensitive natural community.”46 The presence of a sensitive natural community at the Project site was not disclosed 41 Corbin FEIR Vol. 2 Comments, pgs. 3-4. 42 Id., pg. 4. 43 FEIR Vol. 2, pg. 2-18. 44 CEQA requires that an EIR disclose the severity of a project’s impacts and the probability of their occurrence before a project can be approved. 14 CCR §§ 15143, 15162.2(a); Cal. Build. Indust. Ass’n v. BAAQMD (2015) 62 Cal.4th 369, 388-90; Madera Oversight Coalition v. County of Madera (2011) 199 Cal. App. 4th 48, 82; Berkeley Keep Jets Over the Bay Com. v. Bd. of Port Comrs. (2001) 91 Cal.App.4th 1344, 1370-71; CEQA Guidelines, Appendix G. 45 See Exhibit B, Cashen FEIR Vol. 2 Comments.
46 FEIR Vol. 2, Appendix O, pg. 18.

February 21, 2023

5188-015acp
printed on recycled paper
or analyzed in the DEIR or FEIR. Although the FEIR includes compensatory mitigation for the Project’s vernal pool impacts, this mitigation does not incorporate mitigation for the Project’s impacts to the sensitive natural community.47 Project impacts to the sensitive natural community remain undisclosed, potentially significant and unmitigated. This is significant new information requiring recirculation under CEQA Guidelines section 15088.5, because it is a disclosure of a new potentially significant impact.48
Second, the City continues to maintain that the Project will not cause significant impacts to foraging habitat for Swainson’s hawk and other raptors, because (1) the proposed wetlands preserve will provide suitable foraging habitat, and (2) there is “abundant” other foraging habitat offsite. Swainson’s hawks have been observed foraging over the Project site every time Project biologists visited the site during spring or summer months.49 Northern harriers and a golden eagle have also been observed foraging at the Project site. It is undisputed that the Project will eliminate 163 acres of raptor foraging habitat,50 while “preserving” approximately 45 acres in the wetlands preserve. The City claims that this elimination of nearly 80 percent of the site’s foraging habitat, with the remainder to be surrounded by warehouses and heavily traveled roads, is insignificant. Mr. Cashen explains in detail why these conclusions are unsupported by substantial evidence.

The City argues that the 45-acre wetland preserve will retain its value as raptor foraging habitat in part because it remains “hydrologically connected” to the nearby Napa Logistics Park Preserve.51 However, as Mr. Cashen points out, “the FEIR fails to explain why this hydrologic connectivity would maintain habitat quality for avian and terrestrial species, whose habitat is not dependent on hydrologic connectivity.”52
Third, Mr. Cashen points out the flaws in the City’s unsupported contention that the Project itself would not degrade habitat quality in the wetland preserve.

To the contrary, Mr. Cashen has previously provided substantial evidence that “edge effects,” such as noise, artificial light and human activity including traffic, have significant impacts on sensitive biological resources. While the City admits the Project “will increase noise, light and traffic,” it claims that these impacts “will not result in significant effects on the Wetland Preserve.”53 To support this claim, 47 Cashen FEIR Vol. 2 Comments, pg. 1.
48 14 CCR § 15088.5(a)(1).
49 Id., pg. 3-128.
50 FEIR Vol. 2, pgs. 2-12 and 3-285.
51 FEIR Vol. 2, pg. 2-20 and 3-133.
52 Cashen FEIR Vol. 2 Comments, pg. 2.
53 FEIR Vol. 2, pgs. 3-113 – 3-114.

February 21, 2023

5188-015acp
printed on recycled paper
the City claims that the wetland preserve is already “surrounded on all four sides by commercial development, and bisected by a road and pedestrian trail, causing fauna using the area to be already habituated to effects associated with human presence.”
As an initial matter, this statement is factually inaccurate. The wetland preserve area is not currently surrounded by commercial development. It is a portion of a 208-acre parcel of undeveloped land, and fauna at the Project site can minimize exposure to edge effects by using habitat within the interior portions of the 208-acre site. As Mr. Cashen points out, most of the surveys that were used to establish the environmental baseline for this Project were conducted prior to construction of the Devlin Road extension and Napa Logistics Park Phase 2 projects, so there is no scientific basis for the conclusion that fauna at the site has “already habituated” and would tolerate further development.54 Furthermore, development of the Project will increase the level of disturbances and bring it much closer to a significantly smaller parcel of available habitat, which Mr. Cashen opines is “likely to surpass the disturbance threshold” of species using the site.55 Finally, the City claims that the Project’s impacts on Swainson’s hawk foraging habitat would be less than significant because other nearby areas, including the Newell Open Space, Lynch Canyon Open Space and American Canyon California red-legged frog preserve provide over 2,000 acres of open space that can be used for Swainson’s hawk foraging. Mr. Cashen has previously explained and provided substantial evidence that Swainson’s hawk rarely occur at the Newell and Lynch Canyon Open spaces; the City has yet to rebut this evidence. Mr. Cashen points out that just because an area contains a certain land cover type (e.g., grassland) associated with a species, that does not mean that the area necessarily provides habitat for the species.56 He now provides additional data showing that only about 0.05% of checklists submitted to the eBird database for Newell and Lynch Canyon Open Spaces contain observations of Swainson’s hawk.57 He concludes that the presence of these open spaces does not mitigate the Project’s direct, indirect and cumulative impact on foraging habitat.58 The FEIR also claims that once the project is constructed, 412 acres of suitable foraging habitat would remain within 0.5 miles of the Project site, and that this habitat “would be available to any Swainson’s hawk nesting near the project 54 Cashen FEIR Vol. 2 Comments, pg. 3.
55 Id.
56 Id., pg. 9.
57 Id., pgs. 9-10.
58 Id.

February 21, 2023

5188-015acp
printed on recycled paper
site without traveling further north to find suitable foraging areas.”59 Mr. Cashen examined the areas mapped by the City as comprising this purported 412 acres of “suitable foraging habitat” and found that these areas included (1) vineyards (which generally do not provide foraging habitat for Swainson’s hawk, northern harriers, burrowing owls, short-eared owls and golden eagles, i.e., the species addressed in the EIR); (2) a parcel with a pending application to rezone the property to General Industrial; (3) a parcel on which Napa County plans to develop aircraft storage hangars; and (4) parcels zoned for development under the Napa Valley Business Park Specific Plan.60 Based on this information, Mr. Cashen estimates that only 171 acres of potentially available foraging habitat would remain within the approximately 1,473 acres of land encompassing the Project site and areas within a

0.5 mile radius of its boundaries.61 This constitutes substantial evidence that the Project’s cumulative impact to foraging habitat is significant, and that the loss of 163 acres of habitat from development of the Project is cumulatively considerable.

These impacts require disclosure and mitigation.

The City must prepare a revised EIR that fully discloses, analyzes and mitigates all potentially significant impacts to biological resources, and support each of its conclusions with substantial evidence.

III. CONCLUSION
For the foregoing reasons, and all of the reasons set forth in our prior comments, the Council may not certify the FEIR, nor issue the Project’s other approvals, until the EIR complies with CEQA. The Council should direct that the City revise and recirculate the EIR with a full analysis of the Project’s potentially significant impacts and propose appropriate mitigation. Thank you for your attention to these comments.

Sincerely,
Richard M. Franco
Attachments
RMF:acp
59 FEIR Vol. 2, pg. 3-128 and 3-134 and Exhibit D.
60 Cashen FEIR Vol. 2 Comments, pgs. 10-11.
61 Id.

EXHIBIT A
400 North 34th Street, Suite 100 | PO Box 300303 | Seattle, Washington 98103-8636 | 206-632-8020 www.shannonwilson.com
February 19, 2022
Mr. Richard M. Franco
Adams Broadwell Joseph & Cardozo PC
601 Gateway Blvd., Suite 1000
South San Francisco, CA 94080
RE: WETLAND REVIEW OF FINAL ENVIRONMENTAL IMPACT REPORT – VOLUME 2,

GIOVANNONI LOGISTICS CENTER, CITY OF AMERICAN CANYON, CALIFORNIA Dear Mr. Franco:

This letter outlines our responses to the City’s Master Response B and C, and Responses to Written Comments in the Final Environmental Impact Report (FEIR) – Volume 2 (State Clearinghouse Number 2021010104) for the Giovannoni Logistics Center Project (Project) in City of American Canyon (City), Napa County, California.1 This letter complements our previous review documents, Water Resources Review of Draft Environmental Impact Report (DEIR), Giovannoni Logistics Center,2and letter, Wetland Review of Final Environmental Impact Report, Giovannoni Logistics Center.3 COMMENT RESPONSE This letter uses the same comment numbering system (e.g. S&W-5a) established in the FEIR Volume 2 by the City when responding to our November 16, 2023 review.3 Comment Response to Master Response B – Access Road
In its description of the new access road the City writes, 1 First Carbon Solutions, 2022, Draft environmental impact report, Giovannoni Logistics Center project, City of American Canyon, Napa County, California, State Clearinghouse Number 2021010104: Report prepared by Firstcarbon Solutions, Irvine, Calif., for City of American Canyon, Calif., May. 2 Shannon & Wilson, 2022, Water resources review of draft environmental impact report, Giovannoni Logistics Center, City of American Canyon, California: Report prepared by Shannon & Wilson, Seattle, Wash., job no. 109577-001, for Adams Broadwell Joseph & Cardozo PC, South San Francisco, Calif., July, 18p. 3 Shannon & Wilson, 2022, Wetland review of final environmental impact report, Giovannoni Logistics Center, City of American Canyon, California: Letter prepared by Shannon & Wilson, Seattle, Wash., job no. 109577-002, for Adams Broadwell Joseph & Cardozo PC, South San Francisco, Calif., November, 32 p.

Mr. Richard M. Franco
Adams Broadwell Joseph & Cardozo PC
February 19, 2022
109577-002 Page 2 of 7
109577-002-L2.docx/wp/lkn
“The graveled access “road” was placed over this excavation area at approximately the same grade as the surrounding uplands (i.e., no berm was constructed) and only differs from its surroundings with respect to its graveled substrate and constitutes a negligible change to the physical environment post construction.” The conversion of permeable, herbaceous vegetated upland to the impervious graveled surface is not a negligible change. Precipitation will not infiltrate, and the upper layer of soil under the road is compacted either purposely (to construct a stable road) or incidentally (heavy equipment traffic), which could also impact groundwater movement under the road footprint and adjacent compacted areas. Also, if the sewer and water lines that were installed under the road were bedded and/or backfilled with gravel, this could also serve as a conduit to redirect groundwater flow. The City states the road results in a negligible change but does not provide adequate information to support this. Within the proposed Wetland Preserve, impervious surface now exists where there previously was open meadow and a human-made structure requiring maintenance and upkeep exists where previously there was no human intrusion in an open meadow.

Regarding the access road’s wetland impacts, the City writes, “It is unlikely that any existing wetland or other water features have been indirectly impacted by the utility connection for the Napa Logistics Park Phase 2 Project, as, given the level topography with depressional microrelief, the wetlands on the project site are primarily fed through vertical precipitation.” The City has not studied the indirect impacts of the road and utility lines on the Wetland Preserve, nor do they have a complete understanding of the wetland hydrology upon which they could base their statement that indirect impacts are “unlikely” because the wetlands are “primarily fed through vertical precipitation.” For example, what percentage of the wetland complex’s hydrology is supported by direct precipitation vs shallow ground water flow vs overland flow? The City further confounds its understanding of the site’s wetland hydrology when it writes, “Note that No Name Creek is still the primary hydrologic contributor to the Wetland Preserve and that the creek’s hydrology has not been affected.”4
Is the City saying that the site wetlands are primarily supported through precipitation or No-Name Creek? It’s difficult to tell. Also, not all of the wetlands in this location are connected to No Name Creek. Without a quantifiable existing water budget for the site wetlands, conclusions addressing indirect hydrologic wetland impacts are unsubstantiated.

The City proposes to provide a 20-foot buffer between the new access road easement and the created wetlands in the Wetland Preserve. However, the City fails to identify any buffer 4 FEIR Vol. 2 2-13 Mr. Richard M. Franco
Adams Broadwell Joseph & Cardozo PC
February 19, 2022
109577-002 Page 3 of 7
109577-002-L2.docx/wp/lkn
width between the access road and the existing wetlands in the Wetland Preserve. As described at length in our November 16, 2022 letter Comment Response to S&W-10a, 10b, and 10c – Inadequate Buffer Widths, best available science dictates that adequate buffers are necessary to protect wetland functions and values and provide essential habitat for many species that depend on wetlands5. In the northwest corner of the Wetland Preserve, portions of even the small 20-foot wetland buffers will be impacted by the gravel access road, which will fail to protect the wetland through inability to treat surface and shallow ground water, sediment removal, pollutant sequestration, or any other of the welldocumented ways a vegetated, functioning buffer protects wetlands. Best available science recommends buffer widths between 50 and 300 feet, depending on quality of the wetland habitat (higher quality wetlands require greater buffer width), quality of the buffer (low quality buffer should be wider to compensate for the poor condition), and the surrounding land use (greater protection needed from high intensity land use).5 The City has not demonstrated how the Wetland Preserve wetlands (especially the high value vernal pools) will be adequately protected from the proposed high intensity (industrial) land use by an inadequate 20-foot buffer.

The City asserts that the presence of the new access road only reduces the Wetland Preserve by the square footage of the road. However, this ignores the fragmentation the road creates in the Wetland Preserve by physically disconnecting the eastern corner of the mitigation area from the Preserve. The City states the gravel “should not impede or modify the flow pattern on the project site” but cannot make a more definitive statement because it hasn’t been formally evaluated. To suggest that there is no functional disconnection between the two areas is to ignore the vegetation break, the absence of pervious soil, and the compaction below the road footprint and the introduction of this new avenue for human intrusion.

Comment Response to Master Response C – Wetlands
Functional Assessments
The City reiterated that the rare plant survey and species-specific surveys constitute complete wetland functional assessments. As described in our November 16, 2022 letter Comment Response to S&W-6b – Absent Functional Assessments, we do not disagree that these are important components of a functional assessment, but alone, they are not sufficient to understand a wetland’s complete ecological condition, function, or value. Nor does the completion of the delineation data sheets constitute a functional assessment (the City cites 5 Hruby, 2013, Update on wetland buffers: The state of the science, final report, Ecology publication#13-06-11), Washington State Department of Ecology, available:
https://apps.ecology.wa.gov/publications/documents/1306011.pdf.

Mr. Richard M. Franco
Adams Broadwell Joseph & Cardozo PC
February 19, 2022
109577-002 Page 4 of 7
109577-002-L2.docx/wp/lkn
the standard wetland data sheets completed for the delineation as “detailed functional assessments” of soil and hydrology). Elements of a formal functional assessment that are missing from the DEIR and FEIR include organized analyses (not simply observational statements from a limited point in time) related to the wetlands’ hydrology (e.g., water sources, hydroperiods, hydrologic connectivity), buffer and landscape (e.g., buffer width and condition), physical structures typical of vernal pool and seasonal depressional wetlands, plant community interspersion, and assessment of anthropogenic stressors. The City is relying on a future, yet-to-be-performed formal functional assessment (the CRAM) to meet the current needs under CEQA.

Hydrology and Water Budget Analysis
In our previous reviews we have commented extensively on the need for a quantitative analysis, or water budget, of wetland hydrology sources and hydroperiods to adequately understand wetland impacts and confidently mitigate for lost function. In response, the City writes:
“A pre- and post-project water budget analysis is not required to determine whether there will be sufficient hydrology to create additional wetlands, or if newly created wetlands will impact existing wetlands. Reasonable assumptions can be made based on topography data, soil structure data, hydrology inputs (e.g., direct precipitation, overland flow, point source flows data), and anthropogenic factors (e.g., existing roads, fence lines, proposed structures etc.) to determine the likelihood that a site could support the hydrology required to create wetlands and to guide the location of where newly created wetlands should be positioned within the landscape to maximize hydrology inputs and avoid impacting hydrology of existing wetlands.” As the City acknowledges, the proposed compensatory mitigation relies on assumptions, not quantified hydrologic analyses, to meet the State’s ‘no net loss’ policy and therefore fails to reduce project impacts to insignificant levels. It is unreasonable that the City expects to receive approval under CEQA for conclusions that are based on assumptions, rather than yet-to-be performed analyses.

Inadequate Buffer Widths
The City commits to a minimum 25-foot buffer width around the Wetland Preserve; the buffer will consist of herbaceous vegetation and maintained bioretention basins. As explained above in Comment Response to Master Response B and in our previous reviews, best available science tells us this width provides inadequate protection and will not prevent indirect degradation of wetland function from the surrounding industrial land use.

Further, the City does not incorporate even this inadequate buffer width into a formal Mr. Richard M. Franco
Adams Broadwell Joseph & Cardozo PC
February 19, 2022
109577-002 Page 5 of 7
109577-002-L2.docx/wp/lkn
mitigation measure, meaning that the project will not be legally held to any buffer width and that the buffers could be further diminished in the future. The City repeatedly states that the project site is currently surrounded on all four sides by commercial development without acknowledging that the proposed Wetland Preserve is not currently developed.

The project will effectively reduce an existing 800 to 1,500-foot, undeveloped, vegetated buffer to as little as 25 feet (or 20 feet for created wetlands near the new access road). The City does not account for this clear indirect impact in their impact analysis nor in their proposed mitigation.

Wildlife Corridor
Appendix C of the DEIR states, “The Wetland Preserve would create a contiguous open space area with the adjoining 37-acre Napa Logistics Park Wetland Preserve.”6 However, review of the Napa Logistics Preserve boundaries on the California Conservation Easement Database shows the Napa Preserve is not contiguous with the proposed Wetland Preserve.

The City’s recent Master Response C backs off the assertion that the two preserves adjoin by explaining that No Name Creek provides connectivity. However, the two “preserves” will be separated by a road and bridge, and within the Napa Logistics development, the creek is a narrow corridor flanked by industrial land use. The City’s consideration of this landscape as connected open-space relays their lack of analysis of the project’s cumulative impact of continued fragmentation and habitat degradation.

Comment Response SW.NOV.16-7
The City contends that the remaining 6 square feet of a 195 -square-foot wetland will continue to function as it did prior to the wetland impact and that it will “continue to provide habitat for a wide variety of plants and animals.” This contention suggests that each square foot of a wetland acts in isolation from the wetland as a whole and ignores the inherent connectivity and fluidity of wetland ecology within a single wetland. The permanent impact to 97% of an already-small wetland is a prime example of when indirect wetland impacts to the remaining wetland area should be accounted for and mitigated.7 6 DEIR Appendix C page 5 7 Washington State Department of Ecology, U.S. Army Corps of Engineers Seattle District, and U.S.

Environmental Protection Agency Region 10, 2021, wetland mitigation in Washington State – part 1:

Agency policies and guidance (Version 2). Washington State Department of Ecology Publication #21- 06-003.

Mr. Richard M. Franco
Adams Broadwell Joseph & Cardozo PC
February 19, 2022
109577-002 Page 6 of 7
109577-002-L2.docx/wp/lkn
Comment Response SW.NOV.16-17
The City states that, “A reference wetland is typically not used to prepare performance standards or to demonstrate avoidance of significant environmental effects. A reference wetland is one tool used to assess whether a created/ established wetland is progressing toward the mitigation goals.” However, in contradiction to the City’s statement, the CA State Water Resources Control Board and best available science dictates that reference wetlands are used to ensure performance standards are not only achievable8 but also appropriate for identifying the attributes that define the impacted wetlands.9 The City also contradicts its previous statement in the FEIR Response to S&W-15c, where they write, “The [State Water Board] Procedures require the use of “reference aquatic resources.”

CONCLUSIONS
These comments reflect our responses to the latest City responses and should be used in conjunction with our previously submitted review documents.

The impacts associated with the new graveled access road on the Wetland Preserve have not been adequately assessed.

Without complete functional assessments, the Project’s environmental impacts cannot be fully identified and that therefore the proposed mitigation cannot assume to result in a no-net loss of wetland function or a reduction of project impacts to insignificant levels.

Without adequate buffer widths, there is no basis on which to assume that the wetland preserve will not be degraded by the adjacent development and land use and that significant indirect wetland impacts will not occur. Furthermore, without adequate wetland buffers, wetland functions will be lost in the preserved wetlands and will diminish the potential functions that may be achieved in the proposed created wetlands.

The lack of quantitative hydrologic data to demonstrate that 4.69 acres of additional wetland can be created within the Wetland Preserve does not support the City’s statement that the vernal pools will be fully functioning wetlands within 1 year and consequently, the proposed mitigation ratios are likewise unsupported.

As stated in our November 16, 2022 letter Comment Response to SW-15a – Monitoring Period, the proposed five-year mitigation monitoring period is not compliant with the 8 California State Water Resources Control Board (CSWRCB), 2019a, State policy for water quality control: state wetland definition and procedures for discharges of dredged or fill material to waters of the state (rev.): April 2, 44 p., available:
https://www.waterboards.ca.gov/water_issues/programs/cwa401/docs/2021/proce dures.pdf. 9 Ehrenfeld, 2000, Evaluating wetlands within an urban context: Ecological Engineering, V. 15, p. 253- 265.

Mr. Richard M. Franco
Adams Broadwell Joseph & Cardozo PC
February 19, 2022
109577-002 Page 7 of 7
109577-002-L2.docx/wp/lkn
USACE’s Regional Compensatory Mitigation and Monitoring Guidelines for the South Pacific Division, which states that vernal pool wetlands warrant longer monitoring than the standard five years.

As stated in our November 16, 2022 letter Comment Response to SW-15b – Performance Standards, the hydrology performance standard, as written, would confirm presence of wetland, but would not confirm presence of vernal pools, and the functions specific to vernal pools would be lost even though the performance standard would be met. As described above and in our previous review comments, we maintain that the absence of complete functional assessments for the impacted site wetlands result in an incomplete impacts analysis. Deficiencies including inadequate buffer widths, absence of quantitative hydrologic data to support wetland creation potential, low mitigation ratios, inadequate hydrologic performance standards, and lack of documented mitigation sequencing for Phase 2 indicate that the City has not provided sufficient information to establish that the proposed mitigation can avoid or minimize the Project’s significant environmental impacts.

Sincerely,

SHANNON & WILSON
Sarah Corbin, PWS, MSES
Associate
Senior Biologist
SCC:KLW/scc

1 Sarah C. Corbin, PWS, MSES _ASSOCIATE
Senior Biologist

EDUCATION
Bachelors of Arts, Biology and Environmental Studies, Whitman College, 2003 Masters of Environmental Science, Applied Ecology, Emphasis in Aquatic Ecology, Indiana University, 2008

REGISTRATIONS/CERTIFICATIONS
Certified Professional Wetland Scientist, No. 2375, 2013 Certified Erosion and Sediment Control Lead (CESCL), 70002, Expires 1/12/2025 WSDOT Certified Senior Biological Assessment Writer

ADDITIONAL TRAINING
Using the Revised Washington State Wetland Rating System in Western Washington How to Determine the Ordinary High Water Mark, WA Department of Ecology Oregon Streamflow Assessment Duration Method, US Army Corps of Engineers Designing Compensatory Mitigation and Restoration Projects, WA Department of Ecology Eelgrass Delineation Training, US Army Corps of Engineers Completing habitat assessments for compliance with the 2008 NMFS Biological Opinion, FEMA

PROFESSIONAL SUMMARY
Sarah Corbin is a senior biologist and permit specialist with 14 years’ experience providing wetland and stream delineations, aquatic functional assessments, environmental permitting, biological analyses, habitat assessments, and permit compliance for complex projects. She is a Professional Wetland Scientist with experience delineating and implementing the following regional supplements to the US Army Corps of Engineers’ Delineation Manual: Western Mountains, Valleys, and Coast Region, Arid West Region, the Great Plains Region, and the Alaska Region. Sarah coordinates and provides local, state, and federal environmental permitting for hydropower facilities, restoration, linear transportation, utility, and railway projects; and performs additional technical field activities including Ordinary High Water Mark and High Tide Line determinations, rapid assessment stream surveys, bird nesting surveys, fish exclusion, and erosion control and water quality monitoring. She is experienced in writing biological assessments for Endangered Species Act compliance, providing NEPA documentation, writing wetland and stream delineation reports, identifying aquatic minimization and avoidance measures during project design and construction, and designing planting and mitigation plans.

RELEVANT EXPERIENCE
WSDOT, I-5 Joint Base Lewis McChord Congestion Relief Project, Pierce County, WA. Lead Wetland Biologist and assistant project manager. WSDOT is proposing improvements along I-5 in the vicinity of JBLM to reduce traffic congestion and improve person and freight mobility.

Sarah was instrumental in developing the Environmental Scan for the first phase of the project and authored the wetland, fish, and wildlife sections of the report. Sarah performed wetland and OHWM stream delineations for the project’s 14 wetlands and 3 streams and received WSDOT concurrence on all delineations. She documented existing habitat conditions for federally and state-listed species and worked with the design team to minimize project impacts. She authored the project’s Wetland and Stream Delineation Report and developed the project’s Conceptual Wetland Mitigation Plan. Sarah wrote the wetland section of the project’s NEPA Environmental

2 Assessment. In assessing the project’s impacts to stream and wetland resources, Sarah coordinated extensively with planners and specialists at JBLM, Camp Murray, Pierce County, and the Cities of DuPont and Lakewood.

Puget Sound Energy, Lower Baker Dam Seepage Reduction, Concrete, WA. Lead Wetland Biologist and Permitting Specialist. Sarah is the natural resource task lead for this project to reduce seepage below the 285-foot-high, concrete gravity-arch dam. Sarah coordinated the project fieldwork and prepared the project permit applications and supporting documents. Sarah delineated and rated over 20 slope, depressional, riverine, and lacustrine wetlands on the site, received Corps-concurrence on all delineations, authored the project Critical Areas Report/Delineation Report, and developed the project’s mitigation strategy to use an out-ofservice-area mitigation bank when other options were deemed unviable. She worked closely with PSE environmental staff to develop required permit documents including the JARPA and JARPA drawings, SEPA Checklist, Biological Assessment, Restoration Plan, and Bank Use Plan. Sarah supported PSE environmental staff at environmental regulatory agency meetings and helped to track and organize the project’s local, state, and federal permits. She assisted PSE in revising the Biological Assessment and coordinating with the US Fish and Wildlife Service when the gray wolf was re-listed after the start of construction. Sarah continues to support PSE with environmental permit compliance and contractor submittal review during project construction.

City of Marysville, 71st Avenue NE, Soper Hill Road, and Sunnyside Boulevard Intersection Improvement Project, Marysville, WA. Lead Wetland Biologist. To support the project alternatives analysis, Sarah performed a wetland and stream reconnaissance in the vicinity of the proposed intersection improvements and identified seven potential wetlands. Sarah authored the reconnaissance report and detailed applicable regulatory requirements associated with the wetlands and proposed development.

City of Mukilteo, Harbour Reach Drive Extension Project, Mukilteo, WA. Wetland Biologist.

The project constructed a new urban arterial connecting Beverly Park Road and Harbour Pointe Boulevard SW. Sarah delineated and documented the existing conditions of three streams and five wetlands. Construction was completed in July 2021. Sarah assisted in developing the mitigation strategy, including siting the on-site wetland and buffer restoration and enhancement and assisted in the mitigation as-built documentation.

City of Kent Public Works Department, Mill Creek Reestablishment Project, Kent, WA. Lead Wetland Biologist Review. In the City of Kent, Mill Creek flows through a highly urbanized landscape. The project proposes to remove accumulated sediments in 3.7 river miles of the stream to increase flood storage capacity. Sarah provided a field review of the wetland and ordinary high water mark delineation performed for the project, including over 60 riverine wetlands. She verified the delineated boundaries, dug additional wetland an upland soil pits to support the delineation and reviewed the Cowardin Class designation in each wetland.

City of Edmonds Parks, Recreation, and Cultural Services, Edmonds Marsh Buffer Restoration Project, Edmonds, WA. Lead Wetland Biologist. The City proposed to restore portions of the Edmonds Marsh buffer. Sarah delineated portions of Edmonds Marsh adjacent to the proposed buffer improvements and developed the Edmonds Marsh Buffer Restoration Plan, incorporating feedback from City’s Parks, Recreation, and Cultural Services Department, City maintenance staff, and local environmental groups. The plan included removal of invasive species and planting plans for select areas with varying exposure, hydrology, topography, and proximity to public roadways and public use areas.

3 City of Mill Creek, 35th Avenue Southeast Reconstruction Project, Mill Creek, WA. Wetland Biologist. The road improvement project addressed road flooding issues and included stormwater improvements resulting in minor wetland impacts. Sarah completed the wetland and stream delineation to identify critical areas and support the project’s permitting process. Sarah assisted with preparation of environmental applications and documentation to local, state, and federal agencies. When the agencies questioned the site wetland character, Sarah performed pH monitoring and functional assessment to support the wetland’s bog designation. Sarah also provided guidance and oversight to the biologist performing environmental compliance during construction.

BNSF Railway Company, Box Elder Creek Emergency Stream Crossing, Laurel Subdivision MP 208.5, Cascade County, MT. Lead Wetland Biologist and Permit Specialist. In response to an active landslide affecting the railroad tracks, BNSF installed an emergency access road and stream crossing in order to access and stabilize the slide area. Sarah assisted BNSF in determining which regulations applied to the work, performed a wetland and stream delineation, produced a Corps-approved delineation report, and coordinated the post-project environmental permitting and restoration efforts. After the temporary stream crossing had been removed, Sarah assisted BNSF with permitting a permanent stream crossing in the same location.

Puget Sound Energy, Site Development Feasibility Studies, Centralia and Redmond, WA. Lead Wetland Biologist. Sarah visited potential Puget Sound Energy-owned parcels in Centralia and Redmond to identify critical areas as part of comprehensive site development feasibility studies Shannon & Wilson completed for each PSE site. The review included wetland and stream reconnaissance, identification of floodplains, and potential habitat for federal and state protected species. Sarah co-authored Critical Areas Reports for each site.

Washington Department of Fish and Wildlife, Skagit Wildlife Area Delineation, Skagit County, WA. Lead Wetland Biologist. To support WDFW plans to improve an office trailer at the Skagit Wildlife Area Headquarters, Sarah delineated two wetlands in the vicinity of Wiley Slough, in historic floodplain on the landward side of the Wiley Slough Dike. She characterized and rated the wetlands and authored the delineation to support the environmental permit process.

Washington Department of Fish and Wildlife, Wells Recreation Boat Launch Project, Douglas County, WA. Lead Wetland Biologist. To support WDFW plans to improve the Wells Recreation Area boat launch, Sarah delineated one wetland along the Columbia River (Lake Pateros). She characterized and rated the wetland and authored the delineation report necessary to complete the project’s environmental permit process.

BNSF Railway Company, Culvert Replacement Project, Big Horn Subdivision MP 609.1, Gillette, WY. Wetland Biologist and Permit Specialist. BNSF proposed to replace a culvert beneath the access road adjacent to the BNSF railroad tracks. Sarah delineated the wetland and stream associated with the culvert and produced a delineation report to support the permitting for the project.

Port of Everett, Bay Wood Interim Action Cleanup and Shoreline Restoration Project, Everett, WA. Wetland Biologist. The project combined environmental cleanup, wetland habitat restoration, and buffer enhancements along a shoreline adjacent to an upland site redevelopment.

Sarah documented the project’s as-built condition and co-authored the as-built restoration monitoring report. She also assisted the Port of Everett with a site visit with the Department of

4 Ecology and the Tulalip Tribe and developed a replanting plan for areas disturbed during the upland development construction. Sarah will continue to support the Port in site restoration monitoring.

City of Bellevue, 120th Avenue NE Corridor Project, Bellevue, WA. Wetland Biologist. This road improvement project includes road realignment and widening, stormwater management improvements, and replacing an existing culvert crossing with a new fish-passable culvert. Sarah delineated and characterized three onsite wetlands and one stream and authored the project Critical Areas Report/Delineation Report.

City of Bothell, Horse Creek Improvement and Relocation Project, Bothell, WA. Supporting Biologist. The Horse Creek Improvements Project restored a stream that flowed in an underground pipe through downtown Bothell, Washington. The project realigned the stream into approximately 2,000 linear feet of new aboveground stream channel. To support the local permitting process Sarah wrote an ecological justification for the stream restoration design, which contrasted the pre-project stream water quality, hydrologic, and habitat function with those of the restored condition.

Washington Department of Fish and Wildlife, Wiley Slough Tidegate Replacement, Skagit County, WA. Lead Wetland Biologist. The Wiley Slough tidegate is located on the WDFW – Skagit Wildlife Preserve on Fir Island, near Conway, WA. The replacement designs included two 10-foot-wide by 8-foot-high tidegates attached to a concrete box culvert. Sarah completed the estuarine, palustrine, and riverine wetland and ordinary high water mark delineations on the project site and wrote the project delineation report. The delineations were approved without comment by the regulatory agencies.

Washington Department of Fish and Wildlife, Wiley Dike Repair Project, Skagit County, WA.

Lead Wetland Biologist and Permit specialist. The Project will repair 5,800 feet of the existing Wiley Slough Dike to prevent dike overtopping and meet dike stability requirements. The project includes replacing the Wiley Slough tidegate, raising the dike, incorporating riprap within and along the waterward face of the dike, and re-establishing estuarine wetland in an area of dike retraction. Sarah delineated the estuarine, freshwater tidal, riverine, and palustrine wetlands on the project site and authored the project’s Delineation Report, Bank Use Plan, and Restoration Plan. To facilitate permitting of the project explorations, Sarah worked with the design team to avoid impacts to wetlands, completed the shoreline exemption application, and wrote the No Effect letter to document Endangered Species Act compliance. Sarah continues to support WDFW through the main project’s permit process.

Skagit County Public Works, Freestad Lake Barrier Lagoon Restoration, Samish Island, WA.

Lead Wetland Biologist and Permit Specialist. The project proposed to restore tidal inundation and estuary habitat to a 20-acre farm field. Sarah delineated and characterized the existing onsite palustrine and estuarine wetland systems and assisted the County in developing a permitting approach for the project.

Reserve Silica Corporation, Reserve Silica Cement Kiln Dust Landfill, Ravensdale, WA.

Lead Wetland Biologist. Shannon & Wilson assisted Reserve Silica to better understand the extent and impacts of cement kiln dust present on the property. Sarah completed a wetland delineation in an area downgradient of the cement kiln dust burial. During the delineation she took pH measurements in surface water and exposed ground water. Sarah documented the findings and summarized applicable natural resource related local, state, and federal permits to

5 assist Reserve Silica in understanding requirements associated with remediating a high pH pond within the delineated wetland complex.

Puget Sound Energy, Upper Baker Dam Spillway Stability Improvements and Tailrace Channel Rock Debris Removal Project, Whatcom County, WA Lead Wetland Biologist and Permitting Specialist. Sarah is the natural resource and permit task lead for this project to improve spillway slope stability and remove accumulated rockfall debris in the tailrace channel downstream of the dam. Sarah is supporting PSE environmental staff by preparing local, state, and federal permit applications and attending environmental regulatory agency meetings with PSE to help track and coordinate permit process, permit requirements, and answer agency questions. Sarah wrote the project Critical Areas Report/Delineation Report, the JARPA, SEPA Checklist, and Restoration Plan, contributed to the Biological Assessment, and coordinated production of the JARPA drawings. She assisted PSE in revising the Biological Assessment and coordinating with the US Fish and Wildlife Service when the gray wolf was relisted during the consultation process. Sarah continues to assist PSE through the environmental permit process.

City of Everett, Transmission Line 5 Replacement Crossing Pilchuck River Design-Build Project, Snohomish, WA. Lead Wetland Biologist, Environmental Compliance, and Permit Specialist. Project replaced City of Everett Transmission Line 5, a 51-inch steel pipe carrying potable water from Lake Chaplain reservoir to the City of Everett, where it crosses the Pilchuck River. Sarah performed the wetland and ordinary high water mark delineations and produced the delineation report and supporting permit documents, including the Joint Aquatic Resource Application and the restoration planting plan to facilitate the project’s environmental permit process. Sarah also coordinated with the Army Corps and the Snoqualmie Tribe to resolve permitting issues and expedite the permit process. During in-water work construction, Sarah performed spot-check environmental compliance in accordance with project permit conditions and provided onsite guidance to the contractor and their subcontractors. Sarah designed the project planting plans and provided guidance during the restoration phase. Sarah also produced the monitoring reports required by USFWS and NMFS to successfully close out the environmental permits following construction.

WSDOT, SR 532 Corridor Improvements: Camano Island to Interstate 5, WA. Assistant Environmental Compliance Manager. As assistant Environmental Compliance Manager for the SR532 design build project, Sarah performed wetland delineations and stream ordinary high water mark determinations and provided on-site design change guidance to avoid and minimize sensitive area impacts. Sarah is the primary author of the SR532 Environmental Compliance Plan, as well as the Water Quality Monitoring Plan for the Clean Water Act (CWA) Section 401 Water Quality Certification. She managed the SR532 environmental commitments database for the project and reviewed construction daily to ensure compliance with environmental permit commitments and regulations. Sarah performed fish exclusion in fresh water and tidally influenced work areas, allowing construction to continue without significant delay. As the Lead CESCL Field Review for the project, Sarah performed weekly site inspections of Best Management Practices (BMPs) and monitored all construction stormwater discharge points throughout the 10 mile project corridor. She also performed water quality monitoring for in-water and over-water work to meet CWA Section 401 Water Quality Certification conditions, and performed TESC design review and aided in the environmental compliance training of construction personnel. Following construction, Sarah oversaw the planting of wetland mitigation areas, performed mitigation monitoring, and obtained permit closure from the Corps after one year of monitoring, in areas where the contractor was responsible for mitigation.

6 Southwest Suburban Sewer District, Lift Stations No. 14 and 15 Design Retrofit, Burien, WA.

Lead Wetland Biologist and Permit Specialist. The SWSSD proposed to retrofit and upgrade an existing below grade sewer line located within the intertidal zone of public and private beaches along Puget Sound. Sarah performed a wetland reconnaissance and forage fish assessment for the project, prepared a Joint Aquatic Resource Permit Application and coordinated with the Washington Department of Fish and Wildlife to obtain a Hydraulic Project Approval to cover the project in-water work. Sarah also provided technical review of the SEPA Checklist prepared by the sewer district prior to submittal to the City.

City of Edmonds, Harbor Square Outfall Stormwater Improvements, Willow Creek Daylighting Project, Edmonds, WA. Wetland Biologist and Permit Specialist. As part of the Willow Creek Daylighting project, the City is assessing potential improvements to existing outfalls within Edmonds Marsh. Sarah delineated the wetland and ordinary high water mark of Edmonds Marsh and Stella’s Marsh and authored the project’s delineation report, providing an assessment of functional characterizations of the wetlands and outlining the applicable local, state, and federal permits for the project. Sarah also completed water quality monitoring and macroinvertebrate sampling and contributed to the project’s quarterly monitoring reports.

Washington Department of Fish and Wildlife, Fir Island Farm Restoration Feasibility Study, Skagit County, WA. Wetland Biologist. The project evaluated multiple alternatives to restore tidal marsh and coastal shoreline habitat on a 200-acre site. Sarah performed surface water elevation monitoring and data synthesis to help describe existing hydrologic conditions on the site. Sarah also helped obtain local, state, and federal permitting for the project and reviewed the marsh restoration planting and seeding plan.

Skagit County Public Works, North Fork Skagit Levee Setback Feasibility Study, Skagit County, WA. Lead Wetland biologist. The project studied the feasibility of a levee setback project to restore salmon habitat on the Skagit River. Sarah performed a wetland reconnaissance and authored a wetland characterization report to establish baseline wetland and vegetation conditions on the site. Sarah evaluated the biological merit of several tidal channel design alternatives for the Conceptual Design Alternatives report and designed a proposed planting and seeding mix for the marsh restoration habitats.

U.S. Coast Guard, Storis Water Line Replacement, Kodiak, AK. Lead Biologist. The USCG proposes to upgrade existing water line and stormwater infrastructure on the USCG Base Kodiak.

The project includes constructing new stormwater outfalls in Womens Bay. Sarah wrote the Biological Assessment to evaluate the project’s impact to fish and wildlife species or critical habitat listed under the federal Endangered Species Act and essential fish habitat protected under the Magnuson-Stevens Fishery Conservation and Management Act.

U.S. Department of the Navy, Navy Culvert Tunnel Construction Monitoring Project, Bremerton, WA. Biologist, Environmental Compliance and Permit Specialist. Sarah coordinated the environmental tasks for the construction of a fish passage barrier correction along Union River through an active Navy railroad embankment. To restore fish passage upstream in Union River, Shannon & Wilson assisted Navy Facilitates Engineering Systems Command with the construction of a new tunnel, measuring 20-feet wide and 280-feet long and within a 70-foot-tall Navy railroad embankment which connects Naval Base Kitsap Bremerton and NAVBASE Kitsap Bangor. Sarah performed a bird nest survey for compliance with the Migratory Bird Treaty Act, managed fish removal and exclusion downstream of the project site and performed routine monitoring of the block nets to ensure proper function throughout the project. Sarah authored the project’s Environmental Protection Plan, prepared the Stormwater Pollution and Prevention Plan,

7 applied for the construction stormwater NPDES permit, and served as one of the project’s Certified Erosion and Sediment Control Leads (CESCL), conducting TESC site inspections and reporting for compliance throughout construction and site restoration.

BNSF Railway Company, Norfolk Embankment Repair Project, Larimer County, Colorado.

Wetland biologist, environmental compliance and permit specialist. The project involved bank stabilization along 0.4 miles of BNSF Railway embankment in Larimer County, Colorado. Sarah performed a site wetland delineation, wrote the project’s Stormwater Management Plan to meet Colorado State guidelines and designed a native erosion control seed mix for use during the restoration phase based on the area’s climate, topography, and elevation.

International and C Street LLC, International Airport Road and C Street, Anchorage, AK.

Wetland Biologist. The project involved constructing the Anchorage Neighborhood Health Center and associated office buildings, and the Hyatt House hotel and associated parking on a series of parcels located in Anchorage, Alaska. Shannon & Wilson obtained a US Army Corps of Engineers Section 404 Individual Permit to fill wetlands at the development site and designed and implemented a mitigation plan for the project. Sarah performed vegetation and wetland inundation mitigation monitoring for the project, confirmed and mapped wetland boundaries, and coordinated with the Corps to assess whether the mitigation plan’s performance standards had been met.

Whatcom County, Deming Levee Upstream Improvements, Deming, WA. Wetland Biologist.

The Deming Levee project was an engineering and environmental design and permitting project that extended the Deming Levee 1,000 ft to tie into and prevent overtopping of a substandard structure. Sarah delineated the existing wetlands in the project area, wrote the project delineation report, contributed to the conceptual mitigation plan, and monitored the surface and groundwater dataloggers for the project.

BNSF Railway Company, Kelso to Martin Bluff, Task 6, Kelso, WA. Wetland Biologist. The project included constructing an additional rail, as well as new bridge across the Coweeman River. Sarah delineated ten wetlands and the ordinary high water mark along the Coweeman River and Owl Creek and provided functional assessments to document existing conditions in the project area. Sarah authored the wetland discipline report as part of the project’s Environmental Assessment.

City of Redmond, West Lake Sammamish Sewer Manhole #2, Redmond, WA.

Wetland Biologist, and Permit Specialist. The project constructed a manhole in an existing sewer line along the shoreline of Lake Sammamish to facilitate City maintenance. Sarah delineated site wetlands and delineated the Lake Sammamish Ordinary High Water Mark to assist the design team in minimizing and avoiding impacts. She also authored the project’s Biological Assessment, contributed to the Wetland Delineation Report, and applied for and obtained local, state, and federal permitting, including coordination and site visits with agency representatives. Sarah also contributed to the project’s Mitigation Plan and performed the mitigation baseline monitoring.

Lake Washington School District, Juanita High School Replacement, Kirkland, WA.

Project manager, wetland biologist, and permit specialist. The school district intends to replace Juanita High School on the existing school property, which is constrained by a wetland and stream system on 3 of the 4 parcel boundaries. Sarah delineated site wetlands and the ordinary high water mark of the site stream to assist the design team in minimizing and avoiding impacts to critical areas and their buffers and authored the project Wetland and Stream Delineation Report/Critical Areas Report. During the project’s design process the City of Kirkland updated

8 their critical areas code in several iterations and Sarah coordinated with the City to identify the best permitting pathway for the project. Sarah designed a buffer mitigation plan to compensate for temporary buffer impacts. Sarah has also coordinated with Shannon & Wilson hydraulic engineers to provide permitting and hydraulic analyses for a pedestrian footbridge that crossed the project site’s stream. Sarah will continue to assist the project following construction completion through mitigation construction oversight and mitigation monitoring.

Lake Washington School District, Peter Kirk Elementary School Replacement, Kirkland, WA.

Project manager, wetland biologist, and permit specialist. The school district is to replacing Peter Kirk Elementary School on the existing school property. Sarah delineated site wetlands and the ordinary high water mark of the site stream to assist the design team in minimizing and avoiding impacts to critical areas and their buffers. During the project’s design process, the City of Kirkland updated their critical areas code in several iterations and Sarah coordinated with the City to identify the best permitting pathway for the project. She authored the project’s Wetland and Stream Delineation Report/Critical Areas Report and designed and wrote the Buffer Averaging and Enhancement Mitigation Plan. Sarah successfully advocated for the School District when the City of Kirkland attempted to impose critical areas code requirements that were not applicable to the project. Sarah helped the project avoid the unnecessary code constraints by developing a casespecific ecological assessment, detailed code analysis and hydraulic analysis; by assisting the School District at City meetings; and by successfully presenting the School District’s case at the project’s public hearing. Sarah provided buffer mitigation construction compliance and is performing the City-required buffer mitigation monitoring and reporting.

City of Edmonds School District, New Madrona K-8 School Replacement Project, Edmonds, WA. Wetland Biologist and Permitting Specialist. The Edmonds School District replaced the Madrona K-8 School on the school property. Sarah performed wetland delineations to identify critical area constraints on the school property and prepared technical documentation to support design selection and permit applications. She also provided local critical areas code interpretation to minimize impacts to critical areas and reduce the permitting effort. Sarah designed and wrote a buffer averaging and enhancement plan to facilitate approval of local permits. Sarah also successfully presented the mitigation strategy and code compliance summary on behalf of the School District at the project’s public hearing examination. Sarah will continue to provide mitigation monitoring to complete the City’s critical area code requirements.

City of Sammamish, Issaquah-Fall City Road Improvements, Sammamish, WA. Wetland biologist. The project improved SE Issaquah-Fall City Road between 242nd Avenue SE to just beyond the intersection with Klahanie Drive SE. Sarah delineated and documented the existing conditions of North Fork Issaquah Creek and three wetlands, characterized, and rated the onsite wetlands and contributed to the delineation report. Sarah also assisted in the onsite mitigation asbuilt monitoring documentation.

City of Mill Creek, Delineation Services for Cook, Dobson, and Remillard Properties, Mill Creek, WA. Wetland Biologist. Performed wetland delineations and delineation reports to support the City’s planning and development decisions for three City properties. Reduced the City’s site restrictions by identifying and correcting delineation inconsistencies performed previously by other consultants on the sites.

City of Snohomish, Stocker Property, Snohomish, WA. Wetland Biologist. The City of Snohomish purchased 20 acres of farmland along the Snohomish River in order to build a public boat launch and associated parking area. Sarah performed a wetland delineation on the 20-acre property, delineated the Ordinary High Water Mark on the Snohomish River to help the City better understand site constraints. Sarah also assisted the City in understanding the federal, state,

9 and local permitting implications for the project, which was located within both City of Snohomish and Snohomish County jurisdictions.

Clallam County, Dungeness River Levee Setback and Floodplain Restoration, Clallam County, WA. Wetland Biologist and Permit Support. Clallam County is proposing a 2,500-foot-long, river levee setback, roadway relocation, and 120-acre river floodplain restoration project along the Dungeness River. Shannon & Wilson is performing fluvial geomorphology analysis, multidimensional (2D) hydraulic modeling, erosion and scour protection design, groundwater studies, sea level rise and climate change analysis, drainage studies, geotechnical stability analyses, seismic analyses, wetland and stream delineations, wetland functional assessments, and environmental permitting coordination. Sarah delineated nine wetlands and the ordinary high water mark of site streams within the 125-acre site, rated the wetland function using the 2014 Ecology wetland rating system, and analyzed project impacts and mitigation credits made available from the floodplain restoration.

City of Kent, Green River Levee Improvements, Kent, WA. Wetland Biologist. Sarah performed wetland reconnaissance of nine project sites, which included proposed sheet pile walls, proposed road realignment and proposed earthen berms. Sarah assisted in wetland delineations of wetlands that would be impacted by the proposed projects and categorized the wetlands using the Washington State Wetland Rating System for Western Washington. She also authored a wetland delineation report and biological assessment for the proposed projects and authored a biological assessment addendum when the project design changed.

Raft Island Improvement Association, Raft Island Community Bridge Replacement Project, Pierce County, WA. Lead Biologist and Permit Specialist. Sarah facilitated the permitting for the Raft Island Geotechnical Explorations, which included seven over-water borings. The project received funding from the US Department of Agriculture, necessitating coordination between federal and local agencies. Sarah prepared an addendum to the Raft Island Biological Evaluation, the Joint Aquatic Resource Permit Application, wrote a Shoreline Exemption letter and coordinated with Pierce County to obtain permitting for the project.

King County Library System, Algona Pacific Library Expansion, Algona Pacific, WA. Wetland Biologist. Sarah performed a wetland reconnaissance and wetland delineation for the project, rated observed wetlands using the Revised Washington State Wetland Rating System in Western Washington, and prepared a wetland delineation report.

Port of Seattle, Consolidated Rental Car Facility, SeaTac, WA. Biologist. Sarah conducted a subgrade and soil conditions evaluation for the landscaped areas associated with the Consolidated Rental Car Facility in SeaTac, WA. She used detailed field documentation and soil fertility analytical tests to compile a soil evaluation report used by Port of Seattle and Turner Construction in a case against a sub-contractor.

Kitsap County, Central Kitsap Collection & Conveyance System Upgrades Project, Kitsap County, WA. Wetland Biologist. Kitsap County Public Works plans to upgrade three lift stations.

As a biologist, Sarah performed a wetland and stream reconnaissance at the lift station locations and along two proposed force main replacement alignments. Sarah rated the observed wetlands using the Revised Washington State Wetland Rating System in Western Washington, and produced a wetland and stream reconnaissance report.

City of Gig Harbor, Cushman Trail Project Phases 3 and 4, Gig Harbor, WA. Wetland Biologist. The City of Gig Harbor plans to build a non-motorized trail through areas with existing wetlands and streams. Sarah performed a wetland and stream delineation, rated observed wetlands, produced a wetland and stream delineation report, and completed a WSDOT

10 Environmental Classification Summary (ECS) checklist and Joint Aquatic Resource Permit Application to assist in project design and permitting.

WSDOT, SR 520 Bridge Replacements and High Occupancy Vehicle (HOV) Program, I-5 to Medina, Seattle, WA. Biologist. Sarah provided drilling observation erosion control monitoring for over-water and land borings in the I-5 to Medina portion of the SR520 Bridge Replacement Program. As a Shannon & Wilson (SW) field representative, Sarah visually classified soil samples, compiled a field log of each boring, and returned the samples to the SW laboratory for further analysis and testing. She provided TESC Quality Control through her drilling observations and documented those BMPs implemented by WSDOT personnel. Sarah also performed field reconnaissance for boring locations, assessing factors such as impact to critical areas, and BMPs and site restoration requirements.

BNSF Railway Company, BNSF Seattle to Everett Slope Stability Study, Seattle, WA. Project research and data synthesis. The study identified areas of high landslide hazard, prioritized the areas based on landslide hazard, and provided conceptual-level landslide mitigation measures and conceptual-level order-of-magnitude costs to improve slope stability and reduce the landslide hazard to railroad operations. The study included approximately 900 landslide events recorded between 1914 and 2011. Sarah compiled, organized, evaluated, and presented geologic and landslide hazard information for the report.

U.S. Department of the Navy, P-834 Pier Extension Geotechnical Explorations at Naval Base Kitsap (NBK) – Bangor, Kitsap County, WA. Biologist. The project included in-water geotechnical borings in Hood Canal, as well as upland borings and test pits along the Bangor shoreline. Sarah assessed the potential of the project to affect species listed under the Endangered Species Act. Sarah also wrote the No Effect letter for geotechnical explorations on the NBK Bangor naval base, which documented that the project will have no effect on listed species. Sarah also completed the Joint Aquatic Resource Permit Application for the project.

Youngsman Property, Twin Bridge Marine Park Biological Evaluation Addendum, Swinomish Channel, Skagit County, WA. Biologist. The Twin Bridges Marina was originally permitted in 2002 to install in-water piling, and over-water access ramps and floats for 16 boat houses. In 2012, Twin Bridge Marina proposed to modify 13 existing boat slips by moving and re-driving in-water piling. Sarah assessed the potential of the revised project to affect species listed under the Endangered Species Act and authored an addendum to the Twin Bridge Biological Evaluation documenting that the project would not likely adversely affect listed species.

The Nature Conservancy, Fisher Slough Levee Setback Project, Skagit County, WA. Biologist.

The project involved the construction of approximately 3,300 feet of new levee, demolition of existing levee, and habitat restoration of 120 acres of marsh habitat. Sarah completed the Notice of Intent and implemented the NPDES permit during the last phase of the project, which involved bank stabilization in a flood damaged stream channel. Sarah also performed fish exclusion in the stream prior to the bank stabilization work.

Port of Friday Harbor, Bulkhead Replacement Project, Friday Harbor, WA. Biologist. Project replaced an aging rockery bulkhead with a 175-foot long geogrid and concrete wall bulkhead in Friday Harbor on San Juan Island, WA. The project also included the removal of approximately 17, 12-inch derelict timber creosote-treated piles. Sarah performed a qualitative beach survey to assess riparian vegetation and habitat features, aquatic substrate and macroalgae. Sarah also contributed to the project Biological Evaluation assessing potential impacts to listed species under the federal Endangered Species Act and essential fish habitat under the Magnuson-Stevens Fishery Conservation and Management Act, which resulted in a May Affect, not Likely to Adversely Affect effects determination for the project.

11 Kitsap County, Central Kitsap Pump Stations 1, 6, 8, 18, and 38 Conveyance System Upgrades, Kitsap County, WA. Wetland Biologist and Permit Secialist. Kitsap County Public Works proposes to upgrade five pump stations and associated gravity sewers in Central Kitsap. Sarah delineated wetlands and streams along the project alignment and produced the wetland and stream delineation report needed to permit the project. Sarah also prepared the SEPA Checklist for the project on behalf of the County.

BNSF Railway Company, Lake Oahe MP 811.0 Landslide Repair, Mobridge, Corson County, SD. Wetland Biologist and Permit Specialist. Sarah assisted BNSF in permitting slope stabilization and repair measures near Lake Oahe, where an active, deep-seated landslide threatened the safety of train passage on the tracks above the landslide. Sarah successfully proposed and received Corps concurrence that onsite wetlands were hydrologically isolated from the lake and were therefore not regulated under the Clean Water Act, reducing project permitting effort and costs.

Stillaguamish Tribe of Indians, Harvey Creek Fish Ladder Replacement Project, Arlington, WA. Biologist and permit specialist. Project upgraded the Stillaguamish Tribe’s HarveyArmstrong Creek Hatchery, including replacing the fishway, fish ladder, and water intake to meet current fish passage criteria. Sarah wrote the project JARPA, completed the SEPA Checklist, and produced the project Biological Evaluation, documenting that the project was not likely to adversely affect ESA-listed species.

Alaska Department of Transportation and Public Facilities, Integrated Vegetation Management Plan, Alaska. Biologist. The Alaska Department of Transportation and Public Facilities (ADOT & PF) oversees 255 airports, 11 ferries, 5,619 miles of highway and 720 public facilities throughout the state of Alaska. Shannon & Wilson prepared an integrated vegetation management plan to assist ADOT & PF in using herbicide and non-chemical vegetation maintenance and control measures to improve public safety. Sarah conducted interviews with over a dozen ADOT & PF maintenance superintendents and completed site assessments of highway vegetative management throughout Alaska’s ecoregions to evaluate maintenance challenges and prioritize effective management strategies. Sarah also completed factsheets on common herbicides to assist ADOT & PF maintenance operators and the public in understanding the potential risks associated with herbicide as described in the vegetation management plan.

Genesee & Wyoming, Inc., Seeding Specification for Milepost 522.3, Roseburg, OR. Lead Biologist. Genesee & Wyoming, Inc. planned slope stabilization efforts on the slopes above Milepost 522.3, where recent fires and surficial erosion resulted in vegetation loss on the slopes.

To assist Genessee & Wyoming’s slope stabilization efforts in the area, Sarah designing a seed mix of native, erosion control species appropriate for the area’s climate and elevation and wrote the seeding specifications for the projects RFP.

Miller Hayashi Architects, Miller Hiyashi Wetland Reconnaissance, Seattle, WA.

Wetland Biologist. Sarah performed a wetland reconnaissance evaluate existing conditions on a parcel within the City of Seattle. The reconnaissance and Sarah’s coordination with the City of Seattle Department of Construction and Inspections was used to assist the client in the decision to purchase the subject parcel.

Lake Washington School District, Clara Barton (New North Redmond) Elementary School, Redmond, WA. Wetland Biologist, and Permit Specialist. The school district intends to build a new elementary school on a vacant parcel located in Redmond. Sarah performed a wetland and stream reconnaissance of the parcel, determined that none existed within the parcel, and provided

12 a letter report to the District documenting the findings. Sarah also assisted the School District during the SEPA comment review process.

Snohomish County, North Meander Stillaguamish River, Snohomish County, WA. Wetland Biologist. Performed wetland delineation and Ordinary High Water Mark delineation on a backwater area associated with the North Fork of the Stillaguamish River to support future geotechnical explorations and project design.

City of Renton, Ron Regis Park Restroom/Storage Facility, Renton, WA. Wetland Biologist and Permitting Specialist. The City of Renton is planning to construct a restroom and storage facility on Ron Regis Park. The park property is flanked by the Cedar River, Madsen Creek, and several associated wetlands, and lies within the 100-year floodplain of the Cedar River. Sarah delineated wetlands and streams on the park property and prepared the wetland delineation report to assist the City in understanding the extent of onsite wetland and stream buffers and building constraints.

Sarah also wrote the Biological Assessment to assess project impacts on threatened and endangered species.

Kitsap County, Central Kitsap Treatment Plant, Kitsap County, WA. Wetland Biologist and Permitting Specialist. As a result of system improvements on the Central Kitsap Treatment Plant, the County constructed a wetland mitigation site on the treatment plant property. Sarah performed baseline and Years 1 through 5 of monitoring of the mitigation site and coordinated with the County on plant replacements, surface water monitoring, and maintenance requirements. Sarah also coordinated with the US Army Corps of Engineers in evaluating whether the project meets the mitigation success criteria. The mitigation site has successfully completed mitigation monitoring.

Kitsap County, South Central Force Main Project, Kitsap County, WA. Wetland Biologist and Permitting Specialist. The South Central Force Main project consisted of replacing approximately 6,500 feet of 30-inch-diameter techite force main from Paulson Road to the Central Kitsap Wastewater Treatment Plant (CKTP) with 34-inch high-density polyethylene pipe force main.

The project included open trenching, installing the force main, and backfilling the trench with native soil. Mitigation for this project’s impacts to wetlands and buffers generally consisted of restoring temporarily impacted areas to pre-construction conditions. Sarah performed mitigation monitoring over a period of 6 years and coordinated with the County on plant replacements and maintenance requirements. Sarah also coordinated with the US Army Corps of Engineers in evaluating whether the project meets the mitigation success criteria. The mitigation site has successfully completed mitigation monitoring.

Lake Washington School District, 2018 Bond Program School Reconnaissance, Cities of Kirkland, Redmond, and Sammamish, and Unincorporated King County, WA. Wetland Biologist, and Permitting Specialist. Sarah assisted the School District in evaluating several school locations that were previously identified for the 2018 Bond program. She provided wetland and stream reconnaissance, wildlife observations, and regulatory review at several proposed school sites to assist the School District in determining potential site constraints associated with critical areas. The schools under evaluation included Lake Washington High School, New Eastside School, Kamiakin Middle School, and Louisa May Alcott Elementary School.

University of Washington, University of Washington ICA Drainage Ditch Violation, City of Seattle, WA. Wetland biologist and Permitting Specialist. The University of Washington received a violation from the US Army Corps of Engineers and the WA State Department of Ecology for illegal placement of fill in a wetland during construction of the Husky Ballpark Stadium. Shannon

13 & Wilson assisted the UW in responding to the violation by assessing preconstruction condition and wetland function; coordinating with local, state, and federal agencies; and designing an agency-approved mitigation plan. Sarah performed oversight during mitigation site construction and will write the mitigation as-built or baseline monitoring report.

City of Edmonds Department of Parks, Recreation, & Cultural Services, Edmonds Marsh Critical Areas Report, City of Edmonds, WA. Wetland biologist. The City of Edmonds Department of Parks, Recreation, and Cultural Services is proposing to perform maintenance activities, including invasive species removal and native plantings within and in the vicinity of Edmonds Marsh. Sarah was the lead biologist in both field and aerial delineations of Edmonds Marsh, including performing a functional assessment of the marsh and identifying areas of salt and freshwater vegetation communities. Sarah wrote a Critical Areas Report to assist the City in obtaining approval to complete the proposed maintenance activities.

South Puget Sound Salmon Enhancement Group, Titlow Shoreline Restoration Project, City of Tacoma, WA. Permitting Specialist. To support the Titlow Shoreline Restoration Project, Shannon & Wilson performed geotechnical borings within existing BNSF right-of-way. The borings were located within shoreline jurisdiction and wetland buffer. Sarah obtained a City of Tacoma Shoreline Exemption to allow the borings to proceed.

City of Edmonds, Dayton Street Pump Station, City of Edmonds, WA. Permitting Specialist.

The City of Edmonds proposes to upgrade the storm sewer system by constructing a stormwater pump station that connects to existing storm sewer in order to prevent flooding in the Dayton Street area. Sarah contributed to the SEPA Checklist and Shoreline Master Program Compliance Letter to facilitate local permitting for the project.

DATE JOINED FIRM
April 2009

EXPERIENCE WITH OTHER FIRMS
Washington State Department of Ecology. As a biologist for the WA State Department of Ecology, Sarah led a stream habitat monitoring project in the Wenatchee sub-basin. She performed rapid assessment stream surveys through stream channel characterization, macroinvertebrate collection and water chemistry analysis in wadeable and non-wadeable streams. The project required coordination with federal and state agencies as well as local land owners. Sarah also assisted in electro-fishing for the National Rivers and Streams Assessment (NRSA). Sarah implemented the Environmental Protection Agency’s Environmental Monitoring and Assessment Program (EMAP), and NOAA Fisheries’ Integrated Status and Effectiveness Monitoring Program (ISEMP).

PROFESSIONAL ASSOCIATIONS
Society of Wetland Scientists (SWS)
Women in Transportation Services (WTS)

EXHIBIT B
Scott Cashen, M.S.—Independent Biological Resources Consultant 1 February 20, 2023
Mr. Richard M. Franco
Adams Broadwell Joseph & Cardozo
601 Gateway Boulevard, Suite 1000
South San Francisco, CA 94080
Subject: Comments on Volume 2 of the Final Environmental Impact Report for the Giovannoni Logistics Center Project
Dear Mr. Franco:

This letter contains my comments on Volume 2 of the Final Environmental Impact Report (“FEIR”) prepared by the City of American Canyon (“City”) for the Giovannoni Logistics Center Project (“Project”). Buzz Oates LLC (“Applicant”) is proposing to develop a 2.4-millionsquare-foot logistics center on approximately 163 acres of an approximately 208-acre site located in the northwestern portion of the City of American Canyon. The remaining 44.8 acres of the site (“Wetland Preserve”) would serve as mitigation for the Project’s impacts to wetlands and wildlife habitat. I previously submitted comments on the Project’s Draft Environmental Impact Report (“DEIR”) and Volume 1 of the FEIR.

Impacts to Sensitive Natural Communities
Volume 2 of the FEIR includes a new botanical survey report. According to the report:
“The equivalent to vernal pool habitat onsite, would be Sawyer et al. (2009) Lasthenia glaberrima-Pleuropogon californicus association under the Lasthenia glaberrima alliance, which has a global and state rarity ranking of 2 (G2 and S2) and therefore is considered a sensitive natural community regardless of its wetland status.”1
The presence of a sensitive natural community at the Project site constitutes new information that was not disclosed or analyzed in the DEIR or FEIR.2 Although the FEIR incorporates compensatory mitigation for the Project’s impacts to vernal pools, it does not incorporate mitigation for the Project’s impacts to the sensitive natural community (i.e., MM BIO-2 does not require the Applicant to establish the Lasthenia glaberrima-Pleuropogon californicus association within the Wetland Preserve). As a result, Project impacts to the sensitive natural community remain potentially significant.

Connectivity with the Napa Logistics Park Wetland Preserve The FEIR argues that the Wetland Preserve would not function as a habitat “island” with limited value for mitigation purposes because the Wetland Preserve “is hydrologically connected to the 1 FEIR, Vol 2, Appendix O, p. 18.
2 Whereas the DEIR identified presence of vernal pools at the site, not all vernal pools contain sensitive natural communities. The DEIR’s analysis of impacts to sensitive natural communities (Impact BIO-2) pertains to jurisdictional wetlands—not sensitive natural communities. 2 Napa Logistics Park Wetland Preserve approximately 60 linear feet to the north via an existing clear span bridge.”3 This statement is inconsistent with information in the California Conservation Easement Database (“CCED”), which depicts an approximately 700-foot gap between the proposed Wetland Preserve and the Napa Logistics Park Wetland Preserve (Figure 1, below).4 Figure 2 in DEIR Appendix C.6. depicts this gap area as being “slated for development.” In addition, according to the CCED, the Napa Logistics Park Wetland Preserve is

32.1 acres in size, not 37 acres, as stipulated in the FEIR. Nowhere does the EIR provide evidence that 37 acres of permanently protected land occur 60 linear feet north of the proposed Wetland Preserve.

The wetlands and uplands in the proposed Wetland Preserve provide habitat for avian and terrestrial species (i.e., not fish or other aquatic species).5 Even if one assumes the proposed Wetland Preserve would remain hydrologically connected to the Napa Logistics Park Preserve, the FEIR fails to explain why this hydrologic connectivity would maintain habitat quality for avian and terrestrial species, whose habitat is not dependent on hydrologic connectivity. For example, hydrologic connectivity between the two preserves does nothing to maintain highquality foraging habitat for Swainson’s hawks and other raptors, especially given the Projectrelated habitat loss and numerous Project-related factors (e.g., edge effects) that would degrade habitat at the Wetland Preserve.

Figure 1. Conservation easement (in green) associated with the Napa Logistics Park Preserve.

Blue arrow points to bridge north of the proposed Wetland Preserve. 3 FEIR, Vol 2, pp. 2-20 and 3-133. 4 See <https://www.calands.org/>. 5 FEIR, Vol 2, p. 3-114. 3 Edge Effects
The FEIR continues to argue—without scientific justification—that the Project would not degrade habitat quality in the Wetland Preserve. There is substantial evidence in the administrative record that edge effects (e.g., noise, artificial light, human activity, and invasive species, among others) have significant impacts on sensitive biological resources, including those addressed in the FEIR.

The FEIR acknowledges the Project would increase the Wetland Preserve’s exposure to adverse edge effects:
“The commenter is correct that the development for Phase 1 will bring noise, light, and traffic to the southern boundary of the Wetland Preserve, thereby reducing the buffer from 800 to 1,500 to in some areas less than 100 feet, and the City assumes the Phase 2 development would also bring noise, light, and traffic a similar distance to the Wetland Preserve.”6
The FEIR then states:
“The assessment in the Draft EIR and Final EIR, although not specifically stated, considered the location of the Phase 1 and 2 development and the potential effect noise, light and traffic could have on the Wetland Preserve. As addressed in the Response to Comment S&W-11, and further addressed below, the project will increase noise, light, and traffic, but these impacts will not result in significant effects on the Wetland Preserve. This assessment is based on the Wetland Preserve are already being surrounded an all four sides by commercial development, and bisected by a road and pedestrian trail, causing fauna using the area to be already habituated to effects associated with human presence. In addition, the Phase 1 development includes bioretention basins between the Phase 1 development and the Wetland Preserve, providing a noise, visual and sound buffer between the hardscaped areas (i.e., parking, traffic, buildings etc.) and the Wetland Preserve; and the City assumes that Phase 2 development would have similar bioretention basins positions between the Wetland Preserve and hardscaped areas of Phase 2.”7
The FEIR’s reasoning is not supported by evidence. First, most of the surveys that were used to establish the environmental baseline were conducted prior to construction of the Devlin Road/Vine Trail Extension and Napa Logistics Park Phase II Projects. Therefore, there is no scientific basis for the conclusion that fauna at the site had “already habituated” and would tolerate additional development. Furthermore, the FEIR’s rationale is invalid because it ignores the facts that: (a) currently, fauna at the Project site can minimize exposure to edge effects by using habitat within interior portions of the 208-acre site; and (b) although many species can tolerate some levels of anthropogenic disturbance, development of the Project, which increases the levels of disturbances and generates disturbance much closer to available habitat, is likely to surpass the disturbance threshold of those species.
6 FEIR, Vol 2, pp. 3-113 and -114. 7 Ibid.
4 Second, the bioretention basins would not reduce potentially significant edge effects to less than significant levels. Livezey et al. (2016) analyzed published literature to generate a database of minimum approach distances (“MADs”) for birds during the nesting and non-nesting seasons.8 MADs were defined as the distances (buffers) needed to prevent adverse effects to birds from human disturbance. In addition to the value of the Project site’s grasslands as habitat for raptors, one of the primary ecological functions of the wetlands at the site is wintering habitat for waterfowl and shorebirds. The MADs needed to prevent adverse effects on raptors (Order Falconiformes) range from 392 feet (for motorized vehicle disturbance during the non-nesting season) to 2,343 feet (for pedestrian disturbance during the nesting season). The MADs needed to prevent adverse effects on wintering waterfowl (Order Anseriformes) are 233 feet (for pedestrian disturbance) and 404 feet (for motorized vehicle disturbance). The MADs needed to prevent adverse effects on wintering shorebirds (Order Charadriiformes) are 138 feet (for pedestrian disturbance) and 110 feet (for motorized vehicle disturbance).

The buffers provided by the Project’s bioretention basins would range from 25 feet to 150 feet.9 Even with these buffers, most (or all) of the habitat within the Wetland Preserve would be within the MADs described above, and thus, the buffers would not be sufficient to prevent significant impacts on habitat for raptors, waterfowl, and shorebirds.

Third, the FEIR continues to ignore the fact that MM LU-3 requires measures to reduce attraction of birds to the Wetland Preserve. The Applicant’s Wildlife Hazard Assessment predicted the overall wildlife activity at the property would remain at or below current levels due to the disturbances caused by the lighting, human presence, and noise associated with the new development site. Therefore, it is illogical for the FEIR to conclude that the value of the Wetland Preserve as raptor foraging habitat “would be nearly equivalent to that of the project site in its current condition.”10
Invasive Plants
Invasive plants pose a significant threat to both the Wetland Preserve and nearby wildlands (e.g., Napa River marshes). Although the FEIR (Vol 2) incorporates revisions to MM BIO-2, those revisions do not ensure potentially significant impacts associated with the colonization and spread of invasive plants would be reduced to less than significant levels. MM BIO-2 establishes the following performance standard:
“To meet the wetland performance standard invasive plant species within the wetlands shall not comprise greater than 5 percent of the total absolute cover. To meet the upland performance standard, invasive plant species, which currently are not supported in the uplands, within the Wetland Preserve shall not comprise greater than 5 percent of the total absolute cover. Invasive plants shall be defined 8 Livezey KB, Fernandez-Juricic E, Blumstein DT. 2016. Database of bird flight initiation distances to assist in estimating effects from human disturbance and delineating buffer areas. Journal of Fish and Wildlife Management 7(1):181-191. 9 FEIR, Vol 2, p. 2-20. 10 FEIR, Vol 2, p. 3-129. 5 as species rated as “High” by California Invasive Plant Council (Cal-IPC) and species on the “Alert” or “Watch” list with a High rating.” The Cal-IPC does not assign a “High” rating to “Alert” or “Watch” list species. 11 Therefore, the
FEIR’s revision to include these nonexistent categories does not make MM BIO-2 more robust.

Many invasive plants, especially those rated as “High” by the Cal-IPC, possess attributes that enable them to rapidly reproduce and colonize surrounding areas. As a result, a performance standard that allows up to 5 percent absolute cover of invasive plant species does not reduce impacts to less than significant levels.

As stated in my previous comments, all of the plants listed in the Cal-IPC Inventory have the potential to cause significant impacts on the environment, not just those rated as “High.” For example, stinkwort (Dittrichia graveolens) and Pacific bentgrass (Agrostis avenacea) are major threats to ecosystems in the North Bay (including the San Pablo Bay IBA).12 Sinkwort has a Cal-IPC rating of “Moderate” and Pacific bentgrass has a rating of “Limited.” The FEIR states: “[p]lants rated as ‘Moderate’ or ‘Limited’ do not pose ‘severe ecological impacts’ but do pose a certain level of risk.”13 This statement is misleading and does not comport with information provided by the Cal-IPC. As a whole, plants rated as “Moderate” generally do not cause severe ecological impacts (which means some species do indeed cause severe impacts).14 For example, stinkwort forms dense stands with few plant competitors; the “Moderate” rating appears to be due to its recent discovery in California, and the fact that impacts to California wildlife, natural ecosystems and working landscapes have yet to be broadly characterized.
15 Despite its “Limited” rating, Pacific bentgrass “easily outcompetes native vegetation” and “is especially invasive in vernal pool habitat.”16 The FEIR states the following in response to my comment regarding the inadequacy of a 5-year monitoring period: “[o]nce the Wetland Preserve meets all the performance standards, a longterm management plan will be prepared. Monitoring of invasive species would be included in this long-term management plan.”17 This statement is not reflected in the MMRP, which indicates mitigation compliance verification would occur “prior to issuance of grading permits.” Furthermore, the MMRP does not incorporate enforcement mechanisms, reporting requirements, or performance standards for long-term management of the Wetland Preserve, nor do any of the 11 See <https://www.cal-ipc.org/plants/inventory/>. 12 Goals Project. 2015. The Baylands and Climate Change: What We Can Do. Baylands Ecosystem Habitat Goals Science Update 2015 prepared by the San Francisco Bay Area Wetlands Ecosystem Goals Project. California State Coastal Conservancy, Oakland, CA. Available at: <https://www.sfei.org/documents/baylandsgoalsreport>. See also Audubon. 2023. Important Bird Areas: San Pablo Bay Wetlands [website]. Available at:
<https://www.audubon.org/important-bird-areas/san-pablo-bay-wetlands>. 13 FEIR, Vol 2, p. 3-132. 14 See Cal-IPC definitions on FEIR, Vol 2, p. 3-132. 15 Brownsey R, Kyser G, DiTomaso J. 2013. Stinkwort is rapidly expanding its range in California. Calif Agr 67(2):110-115. Available at: <https://calag.ucanr.edu/Archive/?article=ca.v067n02p110>. 16 See <https://www.cal-ipc.org/plants/inventory/>. 17 FEIR, Vol 2, p. 3-132. 6 FEIR’s mitigation measures require invasive species monitoring as part of any long-term management plan.

Impacts to Raptor Foraging Habitat
Swainson’s hawks were observed foraging over the Project site every time project Biologists visited the site during the spring or summer months.18 In addition, northern harriers were observed foraging over the Project site during the winter and spring (breeding) season surveys, and a golden eagle was observed foraging at the Project site during the 2021 breeding season. 19 These data demonstrate that the Project site provides important (core) foraging habitat for special-status raptors.

The FEIR acknowledges the Project would remove 163 acres of raptor habitat.20 However, it then argues that the impact would not be significant, and that the Project would not result in an indirect or cumulative loss of raptor habitat.
21 The crux of the FEIR’s argument is that the
impacts would not be significant because there is abundant habitat elsewhere, including: (a) at 339 acres of privately-owned farmland within 0.5 miles of the Project site, and (b) at “many other privately-owned and vacant parcels between the project site and the City of Napa.”22 The FEIR’s argument is not supported by evidence. As discussed further below, under the cumulative impacts scenario there would be approximately 171 acres of potential habitat remaining (as small, isolated habitat fragments) within 0.5 miles of the Project site.23 Expanding the analysis to the area between the Project site and the City of Napa does not alleviate the significance of the impact. The Project site is located within the Napa Valley Business Park (“NVBP”) Specific Plan area, which extends north to Highway 12. Most of the land north of Highway 12 and south of the City of Napa is comprised of vineyards, urban development, and marshes, which generally do not provide foraging habitat (although marshes provide habitat for the northern harrier and short-eared owl).24 The EIR for the NVBP provided the following analysis:
“The grassland and pasture areas found within the study area serve as an important raptor feeding area. Conversion of these areas from their present agricultural uses to urban development would significantly reduce their value to the various predatory bird species found in the area. Complete development of these areas would eliminate their value to all species while incremental development would slowly reduce their habitat values. Species less tolerant of man's activities such as prairie falcons and rough-legged hawks would be the first to be displaced. Species such as the red-tailed hawk, American kestrel, and turkey vulture would continue to use the larger, undeveloped parcels until they too were 18 FEIR, Vol 2, p. 3-128. 19 DEIR, p. 3.3-25 and Appendix C.1, Biological Resources Report, p. 30. 20 FEIR, Vol 2, pp. 2-12 and 3-285. 21 FEIR, Vol 2, Master Response A and p. 3-285. 22 FEIR, Vol 2, Master Response A. 23 Assumes 45 acres at the proposed Wetland Preserve, 37 acres at the Napa Logistics Park Preserve, 3 acres at the Biaggi Bros wetlands, and the areas depicted on FEIR (Vol 2) Exhibit D (after removing Areas 1 and 3 [vineyard],10 acres from Area 5 [planned development], and Areas 6 and 7 [planned development]). 24 See Napa County, PBES Map Explorer [interactive platform with data layers for current projects, vegetation, and vineyards]. Available at: <https://experience.arcgis.com/experience/406a6b9e95da4a5dac57dc2dba0f4ded/>. 7 developed. Small animals and birds that use the grassland habitat type would be displaced when development occurred. These species would experience a decline in their local populations proportional to the loss of habitat.”25 Approximately 98% of all agriculture and open space land within the NVBP planning area has been, or will be, developed.26 Consequently, the NVBP EIR appropriately concluded that development of the project would have unavoidable, significant impacts on “naturally occurring grassland and vernal pool habitats and associated vegetation and wildlife values.”27 The FEIR also points to habitat at the 78,000-acre San Pablo Bay Wetlands Important Bird Area (“IBA”), the 308-acre American Canyon California red-legged frog preserve, the 620-acre Newell Open Space preserve, and the 1,039-acre Lynch Canyon Open Space Preserve. 28 None
of these areas lie within the geographic scope of the City’s cumulative impacts analysis. The FEIR’s use of one scale to analyze development projects (i.e., 0.5 miles), but a much larger scale to analyze available habitat (17+ miles for the San Pablo Bay Wetlands IBA), fatally distorts the FEIR’s cumulative impacts analysis. In addition, none of the areas provide high quality foraging habitat for Swainson’s hawks. For example, the FEIR acknowledges: “Swainson’s hawk will generally avoid permanently-inundated emergent wetlands as foraging areas.”29 This also applies to the golden eagle, which is associated with open stages of terrestrial habitats (not emergent wetlands).
30 The undeveloped portions of the San Pablo Bay Wetlands IBA primarily consist of permanently-inundated emergent wetlands.
31 The City’s assertion that the proposed Wetland Preserve would provide habitat for the Swainson’s hawk and other raptors is not supported by evidence, nor do any of the FEIR’s mitigation measures incorporate occupancy (habitat use) as a performance standard. Monk & Associates provided the following analysis of the Safe Harbor 3.0 Project (located approximately

0.8 miles north of the Project site):
“The project site is regarded as low quality Swainson’s hawk foraging habitat as it is located in an industrial area, adjacent to the tall walls of the adjacent warehouse which inhibit Swainson’s hawk’s sight and maneuverability with respect to hunting and “stooping” (diving) for prey on this very small site (Estep 1989, Smallwood 1995 as referenced in CDFW’s 2016 Status Review). Boral Stone is located just south of the project site which is characterized by continual high noise levels and human activity.”32
25 Napa County. 1986 (as amended thru 10/22/13). Napa Valley Business Park Specific Plan and EIR. p. 250. [emphasis added].
26 California Department of Fish and Wildlife. 2020 Dec 9. Comments on the Safe Harbor 3.0 Project, Mitigated Negative Declaration, SCH No. 2020110227, Napa County. 27 Napa County. 1986 (as amended thru 10/22/13). Napa Valley Business Park Specific Plan and EIR. p. 279. 28 FEIR, Vol 2, Master Response A. 29 Ibid. 30 See FEIR, Vol 2, p. 2-8. 31 See <https://www.audubon.org/important-bird-areas/san-pablo-bay-wetlands>. 32 Monk & Associates. 2020 Dec 15. Response to Comments Regarding the Mitigation Negative Declaration Safe Harbor 3.0 Project (SCH No. 2020110227) Napa County, California. p. 4. 8 Habitat at the proposed Wetland Preserve would not be appreciably different from habitat at the Safe Harbor 3.0 Project site: both habitat patches are linear in shape, are located in industrial areas, and are (or would be) exposed to high noise levels and human activity. Therefore, if these variables render habitat at the Safe Harbor 3.0 Project site “low quality,” the habitat at the proposed Wetland Preserve would also be low quality.

The FEIR (Vol 2) provided the following response to the issue of warehouse walls inhibiting a Swainson’s hawk’s sight and maneuverability:
“Mr. Cashen cites that M&A statement regarding the fact that Swainson’s hawk avoid foraging in areas with pre-existing buildings and suggests that the Swainson’s hawk will not use the 45-acre Wetland Preserve as a foraging area. It should be pointed out that the project site itself is surrounded by development, including the buildings for the Napa Logistics project to the north and a wine warehouse to the south, and Swainson’s hawks were observed foraging over the project site every time project Biologists visited the project site during the spring or summer months. It is expected that Swainson’s hawk would use the proposed Wetland Preserve in addition to the adjacent Napa Logistics Park Preserve.”33 The FEIR’s response ignores the current size and shape of the Project site (i.e., roughly square), and the fact that surrounding development does not preclude Swainson’s hawks from being able to forage in the site’s interior. The proposed Wetland Preserve would not maintain this interior habitat due to its linear shape (maximum width approximately 700 feet).

Swainson’s Hawk
I previously commented that there is evidence of at least six active Swainson’s hawk nesting territories near the Project site,
34 including an active territory located approximately 0.8 miles to the north. The FEIR provided the following response.
“Based on the information available from the CNDDB, it appears as though there may be as many as six Swainson’s hawk nesting territories north of the project, but there is nothing in the CNDDB data or ebird data or otherwise that suggests that all six would be active during the same nesting season.”35 Swaim Biological Inc. detected six active nest sites north of the Project site during surveys conducted in 2021; additional active nest sites were detected in 2022.36 Data pertaining to these nest sites are available from the CNDDB, are mapped in CDFW’s Biogeographic Information and Observation System, and were cited in my previous comment letter. 33 FEIR, Vol 2, p. 3-128. 34 As reported in Master Response 1, Swainson’s hawks may change nesting locations from year to year. However, the number of active nests detected in a given year represents the minimum number of nesting territories. The CNDDB has records of six active nests within 10-miles of the Project site in 2021. 35 FEIR, Vol 2, p. 3-128. 36 Swaim Biological Inc. 2022 June. Swainson’s Hawk (Buteo swainsoni) 2022 Pre-Construction Protocol Level Survey Report: State Route 29/State Route 221 Soscol Junction Improvement Project. Figures A-1 and 3-1. 9 The FEIR repeatedly make the argument that Project impacts on Swainson’s hawk foraging habitat would be less than significant because the Newell Open Space, Lynch Canyon Open Space, and American Canyon California red-legged frog preserve provide over 2,000 acres of open space that can be used for foraging by Swainson’s hawks. I commented that eBird data do not support the argument that those areas provide foraging habitat for Swainson’s hawks (in addition, neither the CNDDB nor the iNaturalist database37 have records of Swainson’s hawks at Newell Open Space, Lynch Canyon Open Space, or the American Canyon California red-legged frog preserve). The FEIR provided the following response:
“habitat within the Lynch Canyon Open Space consists of rolling hills vegetated with annual grasslands and riparian canyons suitable habitat for foraging by Swainson’s hawks. The Newell Open Space contains some forested area, but the presence of open grasslands within this Open Space area is also capable of supporting foraging by the species. The lack of ebird sightings in either location does not mean that use of these areas for this purpose does not occur.”38 The Swainson’s hawk is a conspicuous species. The Newell Open Space and Lynch Canyon Open Space are publicly-accessible birding “hotspots” with 822 “complete checklists” in the eBird database. As explained further below, this constitutes substantial evidence that Swainson’s hawks rarely occur at those open space areas. Although the FEIR suggests the eBird data are uninformative, it provides no alternative evidence that Swainson’s hawks do indeed forage at the open space areas.

Habitat is dictated by numerous biotic, abiotic, and spatial variables. Therefore, just because an area contains a land cover type (e.g., grassland) associated with a species, does not mean that area provides habitat for the species. As explained in my previous comments, there is a negative relationship between Swainson’s hawk survival and travel distance to foraging habitat.

Therefore, one of the reasons Swainson’s hawks do not regularly forage in the open space areas east of American Canyon may be because those areas are located too far from the nest sites (all of which are located north of the Project site).

The eBird database reports the number of “complete checklists” for each “hotspot.” A complete checklist is defined as “any eBird list where birding was your primary purpose, and every species you could identify to the best of your ability, by sight and/or sound, is reported.” A hotspot is defined as a “public birding location where checklists are aggregated and you can view data summaries.”39 Out of the 575 complete checklists that have been submitted for the Lynch Canyon Open Space Hotspot, only 3 (0.05%) contain observations of a Swainson’s hawk. This is consistent with data from the Newell Open Space Hotspot, which has 247 complete checklists, but only one (0.04%) with an observation of Swainson’s hawks (two individuals).40 By comparison, 28 of the 86 (33%) complete checklists for the Butler Bridge Hotspot contain Swainson’s hawk observations, while 17 of the 120 (14%) complete checklists for the Sheehy Creek Hotspot contain Swainson’s hawk observations (both of these hotspots are located in the 37 See <https://www.inaturalist.org/>. 38 FEIR, Vol 2, p. 3-128. 39 eBird. 2021. eBird Glossary. Available at: <https://support.ebird.org/en/support/solutions/articles/48000948655- ebird-glossary#anchorF>. 40 See <https://ebird.org/hotspots>. 10 vicinity of the Project site). Collectively, these data provide substantial evidence that Lynch Canyon Open Space and Newell Open Space are rarely used by Swainson’s hawks, and thus, that presence of those open space areas does not mitigate the Project’s direct, indirect, and cumulative impacts on foraging habitat.

Cumulative Impacts
The FEIR states: “the Final EIR Volume 1 Response to Cashen-13 regarding cumulative impacts provides a quantification of the amount of foraging habitat available to Swainson’s hawk within

0.5 mile of the project site.”41 The FEIR’s statement is false: neither the DEIR nor FEIR (Vol 1) quantify the amount of foraging habitat that would be available to Swainson’s hawks when cumulative impacts are considered.

My FEIR comment letter included a map of cumulative impacts within 0.5 miles of the Project site.42 The FEIR uses this map to make the determination that 412 acres of suitable foraging habitat would remain within 0.5 miles of the Project site, and that this foraging habitat “would be available to any Swainson’s hawk nesting near the project site without traveling further north to find suitable foraging areas.”43
Because the DEIR did not include a map of land cover types within 0.5 miles of the Project site, I relied on Google Earth imagery and data available from the City (e.g., regarding future projects) to generate the map. I have subsequently discovered additional data available from Napa County.44 The data reveal the following:
1) Areas #1 and #3 (depicted on Exhibit D in the FEIR) are vineyards (Figure 2, below).

Due to their canopy and dense cover, vineyards generally do not provide foraging habitat for Swainson’s hawks, golden eagles, northern harriers, burrowing owls, and short-eared owls (i.e., the species addressed in the EIR).
45 2) Area #1 has a pending application for a General Plan Amendment, which would rezone the property to General Industrial (GI) Land Use and General Industrial:Airport Compatibility (GI:AC).
46 3) Napa County plans to develop aircraft storage hangers or fixed base operator leaseholds within an approximately 10-acre portion of Area #5.47
41 FEIR, Vol 2, p. 3-128. 42 Cashen FEIR comments, Figure 6. 43 FEIR, Vol 2, pp. 3-128, -134, and Exhibit D. 44 Napa County, PBES Map Explorer [interactive platform with data layers for current projects and vineyards].

Available at: <https://experience.arcgis.com/experience/406a6b9e95da4a5dac57dc2dba0f4ded/>. 45 See California Department of Fish and Wildlife. 2016. Status Review: Swainson’s Hawk (Buteo swainsoni) in California. See also Shuford WD, Gardali T (editors). 2008. California Bird Species of Special Concern: A ranked assessment of species, subspecies, and distinct populations of birds of immediate conservation concern in California.

Studies of Western Birds 1. Western Field Ornithologists, Camarillo, California, and California Department of Fish and Game, Sacramento. Available at: <https://wildlife.ca.gov/Conservation/SSC/Birds>. 46 <https://www.countyofnapa.org/2876/Current-Projects-Explorer>. 47 Napa County. 2016. Approved Airport Layout Plan (ALP). Available at:
<https://www.countyofnapa.org/DocumentCenter/View/17999/Approved-Airport-Layout-Plan-ALP-May-2016- PDF>.
11 4) Areas #6 and #7 are zoned for development under the Napa Valley Business Park Specific Plan.

Based on this new information, only 171 acres of potentially available foraging habitat would remain within the approximately 1,473 acres of land that encompass the Project site and areas within a 0.5-mile radius of its boundaries. This constitutes substantial evidence that the cumulative impact to foraging habitat is significant, and that the Project’s 163-acre contribution to that impact is cumulatively considerable.

Figure 2. Vineyards (shaded green) in the vicinity of the Project site (blue star). Light green areas were fallow (between plantings) when the map was last updated by Napa County (in 2021).

Sincerely,
Scott Cashen, M.S.

Senior Biologist
Cashen, Curriculum Vitae 1
Scott Cashen, M.S.

Senior Wildlife Biologist
Scott Cashen has 28 years of professional experience in natural resources management. During that time he has worked as a field biologist, forester, environmental consultant, and instructor of Wildlife Management. Mr. Cashen focuses on CEQA/NEPA compliance issues, endangered species, scientific field studies, and other topics that require a high level of scientific expertise.

Mr. Cashen has knowledge and experience with numerous taxa, ecoregions, biological resource issues, and environmental regulations. As a biological resources expert, Mr.

Cashen is knowledgeable of the various agency-promulgated guidelines for field surveys, impact assessments, and mitigation. Mr. Cashen has led field investigations on several special-status species, including ones focusing on the yellow-legged frog, red-legged frog, desert tortoise, steelhead, burrowing owl, California spotted owl, northern goshawk, willow flycatcher, Peninsular bighorn sheep, red panda, and various forest carnivores.

Mr. Cashen is a recognized expert on the environmental impacts of renewable energy development. He has been involved in the environmental review process of over 100 solar, wind, biomass, and geothermal energy projects. Mr. Cashen’s role in this capacity has encompassed all stages of the environmental review process, from initial document review through litigation support. Mr. Cashen provided expert witness testimony on several of the Department of the Interior’s “fast-tracked” renewable energy projects. His testimony on those projects helped lead agencies develop project alternatives and mitigation measures to reduce environmental impacts associated with the projects.

Mr. Cashen was a member of the independent scientific review panel for the Quincy Library Group project, the largest community forestry project in the United States. As a member of the panel, Mr. Cashen was responsible for advising the U.S. Forest Service on its scientific monitoring program, and for preparing a final report to Congress describing the effectiveness of the Herger-Feinstein Forest Recovery Act of 1998.

AREAS OF EXPERTISE
• CEQA, NEPA, and Endangered Species Act compliance issues • Comprehensive biological resource assessments
• Endangered species management
• Renewable energy development
• Scientific field studies, grant writing and technical editing

EDUCATION
M.S. Wildlife and Fisheries Science - The Pennsylvania State University (1998) Thesis: Avian Use of Restored Wetlands in Pennsylvania
B.S. Resource Management - The University of California, Berkeley (1992) Cashen, Curriculum Vitae 2

PROFESSIONAL EXPERIENCE
Litigation Support / Expert Witness
Mr. Cashen has served as a biological resources expert for over 125 projects subject to environmental review under the California Environmental Quality Act (CEQA) and/or the National Environmental Policy Act (NEPA). As a biological resources expert, Mr.

Cashen reviews CEQA/NEPA documents and provides his clients with an assessment of biological resource issues. He then submits formal comments on the scientific and legal adequacy of the project’s environmental documents (e.g., Environmental Impact Report).

If needed, Mr. Cashen conducts field studies to generate evidence for legal testimony, or he can obtain supplemental testimony from his deep network of species-specific experts.

Mr. Cashen has provided written and oral testimony to the California Energy Commission, California Public Utilities Commission, and U.S. district courts. His clients have included law firms, non-profit organizations, and citizen groups.

REPRESENTATIVE EXPERIENCE
Solar Energy Geothermal Energy
• Abengoa Mojave Solar Project • Casa Diablo IV Geothermal • Avenal Energy Power Plant • Project East Brawley Geothermal • Development • Beacon Solar Energy Project • Mammoth Pacific 1 Replacement • Blythe Solar Power Project • Facility Orni 21 Geothermal Project • ff
• Steamfield
• Calico Solar Project • Western GeoPower Plant
• California Flats Solar Project Wind Energy
• Calipatria Solar Farm II • Catalina Renewable Energy
• Carrizo Energy Solar Farm • Project Ocotillo Wind Energy Project • Catalina Renewable Energy
Project
• SD County Wind Energy
• Fink Road Solar Farm • Ordinance Searchlight Wind Project • Genesis Solar Energy Project • Shu’luuk Wind Project
• Heber Solar Energy Facility • Tres Vaqueros Repowering Project • Imperial Valley Solar Project • Tule Wind Project
• Ivanpah Solar Electric Generating
System
• Vasco Winds Relicensing Project
• Maricopa Sun Solar Complex Biomass Facilities
• McCoy Solar Project • CA Ethanol Project
• • Mt. Signal and Calexico Solar
Projects
• Colusa Biomass Project
• Panoche Valley Solar • Tracy Green Energy Project
• • San Joaquin Solar I & II Other Development Projects • San Luis Solar Project • Cal-Am Desalination Project
• Stateline Solar Project • Carnegie SVRA Expansion Project • Solar Gen II Projects • Lakeview Substation Project
• SR Solis Oro Loma • Monterey Bay Shores Ecoresort
• Vestal Solar Facilities • Phillips 66 Rail Spur
•
•
• Victorville 2 Power Project • Valero Benecia Crude By Rail • Willow Springs Solar • World Logistics Center
Cashen, Curriculum Vitae 3
Project Management
Mr. Cashen has managed several large-scale wildlife, forestry, and natural resource management projects. Many of the projects have required hiring and training field crews, coordinating with other professionals, and communicating with project stakeholders. Mr.

Cashen’s experience in study design, data collection, and scientific writing make him an effective project manager, and his background in several different natural resource disciplines enable him to address the many facets of contemporary land management in a cost-effective manner.

REPRESENTATIVE EXPERIENCE
Wildlife Studies
• Peninsular Bighorn Sheep Resource Use and Behavior Study: (CA State Parks) • “KV” Spotted Owl and Northern Goshawk Inventory: (USFS, Plumas NF) • Amphibian Inventory Project: (USFS, Plumas NF)
• San Mateo Creek Steelhead Restoration Project: (Trout Unlimited and CA Coastal Conservancy, Orange County)
• Delta Meadows State Park Special-Status Species Inventory: (CA State Parks, Locke)
Natural Resources Management
• Mather Lake Resource Management Study and Plan – (Sacramento County) • Placer County Vernal Pool Study – (Placer County)
• Weidemann Ranch Mitigation Project – (Toll Brothers, Inc., San Ramon) • Ion Communities Biological Resource Assessments – (Ion Communities, Riverside and San Bernardino Counties)
• Del Rio Hills Biological Resource Assessment – (The Wyro Company, Rio Vista) Forestry
• Forest Health Improvement Projects – (CalFire, SD and Riverside Counties) • San Diego Bark Beetle Tree Removal Project – (SDG&E, San Diego Co.) • San Diego Bark Beetle Tree Removal Project – (San Diego County/NRCS) • Hillslope Monitoring Project – (CalFire, throughout California) Cashen, Curriculum Vitae 4
Biological Resources
Mr. Cashen has a diverse background with biological resources. He has conducted comprehensive biological resource assessments, habitat evaluations, species inventories, and scientific peer review. Mr. Cashen has led investigations on several special-status species, including ones focusing on the foothill yellow-legged frog, mountain yellowlegged frog, desert tortoise, steelhead, burrowing owl, California spotted owl, northern goshawk, willow flycatcher, Peninsular bighorn sheep, red panda, and forest carnivores.

REPRESENTATIVE EXPERIENCE
Biological Assessments/Biological Evaluations (“BA/BE”) • Aquatic Species BA/BE – Reliable Power Project (SFPUC) • Terrestrial Species BA/BE – Reliable Power Project (SFPUC) • Management Indicator Species Report – Reliable Power Project (SFPUC) • Migratory Bird Report – Reliable Power Project (SFPUC) • Terrestrial and Aquatic Species BA – Lower Cherry Aqueduct (SFPUC) • Terrestrial and Aquatic Species BE – Lower Cherry Aqueduct (SFPUC) • Terrestrial and Aquatic Species BA/BE – Public Lands Lease Application (Society for the Conservation of Bighorn Sheep)
• Terrestrial and Aquatic Species BA/BE – Simon Newman Ranch (The Nature Conservancy)
• Draft EIR (Vegetation and Special-Status Plants) - Wildland Fire Resiliency Program (Midpeninsula Regional Open Space District)
Avian
• Study design and Lead Investigator - Delta Meadows State Park Special-Status Species Inventory (CA State Parks: Locke)
• Study design and lead bird surveyor - Placer County Vernal Pool Study (Placer County: throughout Placer County)
• Surveyor - Willow flycatcher habitat mapping (USFS: Plumas NF) • Surveyor - Tolay Creek, Cullinan Ranch, and Guadacanal Village restoration projects (Ducks Unlimited/USGS: San Pablo Bay)
• Study design and Lead Investigator - Bird use of restored wetlands research (Pennsylvania Game Commission: throughout Pennsylvania) • Study design and surveyor - Baseline inventory of bird species at a 400-acre site in Napa County (HCV Associates: Napa)
• Surveyor - Baseline inventory of bird abundance following diesel spill (LFR Levine-Fricke: Suisun Bay)
Cashen, Curriculum Vitae 5
• Study design and lead bird surveyor - Green Valley Creek Riparian Restoration Site (City of Fairfield: Fairfield, CA)
• Surveyor - Burrowing owl relocation and monitoring (US Navy: Dixon, CA) • Surveyor - Pre-construction burrowing owl surveys (various clients: Livermore, San Ramon, Rio Vista, Napa, Victorville, Imperial County, San Diego County) • Surveyor - Backcountry bird inventory (National Park Service: Eagle, Alaska) • Lead surveyor - Tidal salt marsh bird surveys (Point Reyes Bird Observatory:
throughout Bay Area)
• Surveyor – Pre-construction surveys for nesting birds (various clients and locations)
Amphibian
• Crew Leader - Red-legged frog, foothill yellow-legged frog, and mountain yellow-legged frog surveys (USFS: Plumas NF)
• Surveyor - Foothill yellow-legged frog surveys (PG&E: North Fork Feather River)
• Surveyor - Mountain yellow-legged frog surveys (El Dorado Irrigation District:

Desolation Wilderness)
• Crew Leader - Bullfrog eradication (Trout Unlimited: Cleveland NF) Fish and Aquatic Resources
• Surveyor - Hardhead minnow and other fish surveys (USFS: Plumas NF) • Surveyor - Weber Creek aquatic habitat mapping (El Dorado Irrigation District:

Placerville, CA)
• Surveyor - Green Valley Creek aquatic habitat mapping (City of Fairfield:

Fairfield, CA)
• GPS Specialist - Salmonid spawning habitat mapping (CDFG: Sacramento River) • Surveyor - Fish composition and abundance study (PG&E: Upper North Fork Feather River and Lake Almanor)
• Crew Leader - Surveys of steelhead abundance and habitat use (CA Coastal Conservancy: Gualala River estuary)
• Crew Leader - Exotic species identification and eradication (Trout Unlimited:

Cleveland NF)
Mammals
• Principal Investigator – Peninsular bighorn sheep resource use and behavior study (California State Parks: Freeman Properties)
Cashen, Curriculum Vitae 6
• Scientific Advisor –Study on red panda occupancy and abundance in eastern Nepal (The Red Panda Network: CA and Nepal)
• Surveyor - Forest carnivore surveys (University of CA: Tahoe NF) • Surveyor - Relocation and monitoring of salt marsh harvest mice and other small mammals (US Navy: Skagg’s Island, CA)
• Surveyor – Surveys for Monterey dusky-footed woodrat. Relocation of woodrat houses (Touré Associates: Prunedale)
Natural Resource Investigations / Multiple Species Studies • Scientific Review Team Member – Member of the scientific review team assessing the effectiveness of the US Forest Service’s implementation of the Herger-Feinstein Quincy Library Group Act.
• Lead Consultant - Baseline biological resource assessments and habitat mapping for CDF management units (CDF: San Diego, San Bernardino, and Riverside Counties)
• Biological Resources Expert – Peer review of CEQA/NEPA documents (various law firms, non-profit organizations, and citizen groups) • Lead Consultant - Pre- and post-harvest biological resource assessments of tree removal sites (SDG&E: San Diego County)
• Crew Leader - T&E species habitat evaluations for Biological Assessment in support of a steelhead restoration plan (Trout Unlimited: Cleveland NF) • Lead Investigator - Resource Management Study and Plan for Mather Lake Regional Park (County of Sacramento: Sacramento, CA)
• Lead Investigator - Biological Resources Assessment for 1,070-acre Alfaro Ranch property (Yuba County, CA)
• Lead Investigator - Wildlife Strike Hazard Management Plan (HCV Associates:

Napa)
• Lead Investigator - Del Rio Hills Biological Resource Assessment (The Wyro Company: Rio Vista, CA)
• Lead Investigator – Ion Communities project sites (Ion Communities: Riverside and San Bernardino Counties)
• Surveyor – Tahoe Pilot Project: Validation of California’s Wildlife Habitat Relationships (CWHR) Model (University of California: Tahoe NF) Cashen, Curriculum Vitae 7
Forestry
Mr. Cashen has five years of experience working as a consulting forester on projects throughout California. Mr. Cashen has consulted with landowners and timber operators on forest management practices; and he has worked on a variety of forestry tasks including selective tree marking, forest inventory, harvest layout, erosion control, and supervision of logging operations. Mr. Cashen’s experience with many different natural resources enable him to provide a holistic approach to forest management, rather than just management of timber resources.

REPRESENTATIVE EXPERIENCE
• Lead Consultant - CalFire fuels treatment projects (SD and Riverside Counties) • Lead Consultant and supervisor of harvest activities – San Diego Gas and Electric Bark Beetle Tree Removal Project (San Diego)
• Crew Leader - Hillslope Monitoring Program (CalFire: throughout California) • Consulting Forester – Forest inventories and timber harvest projects (various clients throughout California)
Grant Writing and Technical Editing
Mr. Cashen has prepared and submitted over 50 proposals and grant applications.

Many of the projects listed herein were acquired through proposals he wrote. Mr.

Cashen’s clients and colleagues have recognized his strong scientific writing skills and ability to generate technically superior proposal packages. Consequently, he routinely prepares funding applications and conducts technical editing for various clients.

PERMITS
U.S. Fish and Wildlife Service Section 10(a)(1)(A) Recovery Permit for the Peninsular bighorn sheep

PROFESSIONAL ORGANIZATIONS / ASSOCIATIONS
The Wildlife Society
Cal Alumni Foresters
Mt. Diablo Audubon Society

OTHER AFFILIATIONS
Scientific Advisor and Grant Writer – The Red Panda Network Scientific Advisor – Mt. Diablo Audubon Society
Grant Writer – American Conservation Experience
Cashen, Curriculum Vitae 8

TEACHING EXPERIENCE
Instructor: Wildlife Management - The Pennsylvania State University, 1998 Teaching Assistant: Ornithology - The Pennsylvania State University, 1996-1997

PUBLICATIONS
Gutiérrez RJ, AS Cheng, DR Becker, S Cashen, et al. 2015. Legislated collaboration in a conservation conflict: a case study of the Quincy Library group in California, USA.

Chapter 19 in: Redpath SR, et al. (eds). Conflicts in Conservation: Navigating Towards Solutions. Cambridge Univ. Press, Cambridge, UK.

Cheng AS, RJ Gutiérrez RJ, S Cashen, et al. 2016. Is There a Place for Legislating PlaceBased Collaborative Forestry Proposals?: Examining the Herger-Feinstein Quincy Library Group Forest Recovery Act Pilot Project. Journal of Forestry.

ERRATA
Table 2. Number of pairs and breeding densities of four raptor species in the Central Valley Swainson's Hawk study areas, 1987.

Gal t
Woodland Wilton Thornton Vernalis Total
Swainson's ZO(0.21) ll(0.12) 16(0.17) 13(0.14) 60(0.16) Hawk (0.56) (0.31) (0.44) (0.36) (0.42)
Red-tailed 4(0.04) 14(0.15) lS(0.16) 13(0.14) 46(0.12)
Hawk (0.11) (0.39) (0.42) (0.36) (0.32)
Red-shouldered O(O.00) 7(0.07) g(0.10) S(0.02) 21 (0.08) Hawk (0.00) (0.19) (0.25) (0.14) (0.15)
Black-shouldered 12(0.13) 6(0.06) 9(0.10) 2(0.02) 29(0.08) Kite (0.33) (0.17) (0.25) (0.06) (0.20)
Total #(x#/sq km) 36(0.38) 38(0.41) 49(0.52) 33(0.35) 156(0.42) (Z#/sqmi) (1.00) (1 .06) (1.36) (0.91 ) (1.08)
Note: The corrections in Table 2 contained in this Errata are not reflected in the original text of the document.
14 See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/254855117 Wetland Mitigation in the United States: Assessing the Success of Mitigation Policies
Article  in  Wetlands Australia · January 2000
DOI: 10.31646/wa.242

CITATIONS
34 READS
941
1 author:

Some of the authors of this publication are also working on these related projects:

University of California Stormwater Initiative View project Ecotoxicology of Oil Spills View project
Richard F Ambrose
University of California, Los Angeles
162 PUBLICATIONS   6,559 CITATIONS

SEE PROFILE
All content following this page was uploaded by Richard F Ambrose on 10 August 2014.

The user has requested enhancement of the downloaded file.

View publication stats
An Evaluation of Compensatory Mitigation Projects Permitted Under Clean Water Act Section 401 by the California State Water Quality Control Board, 1991-2002.

Richard F. Ambrose1
John C. Callaway2
Steven F. Lee1
1 University of California, Los Angeles
2 University of San Francisco
Prepared for:

California State Water Resources Control Board
August 2006
An Evaluation of Compensatory Mitigation Projects Permitted Under Clean Water Act Section 401 by the California State Water Quality Control Board, 1991-2002.

Final Report
(Review Copy)
Richard F. Ambrose1,2
John C. Callaway3
Steven F. Lee2
With technical assistance from:

UCLA: Elizabeth Bernier, Leanna Heffner, Marjorie Lundgren, Lisa Max, Natalie (Wenner) Diaz
1 Environmental Science and Engineering Program
2 Department of Environmental Health Sciences
University of California, Los Angeles
Los Angeles, CA 90095-1772
USF: Robert Genova, Robert Goldstein, Sandee Hufana, Laura Wainer 3 Department of Environmental Science
University of San Francisco
2130 Fulton St.

San Francisco, CA 94117-1080
Prepared for:

State of California
California Environmental Protection Agency
California State Water Resources Control Board
Los Angeles Region (Region 4)
320 W. 4th Street, Suite 200
Los Angeles, CA 90013
Contract number: 03-259-250-0
August 2006
Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) i
1 Note to reader:
2 This report is the Final Report under California State Water Resources Control Board 3 Contract 03-259-250 to the University of California, Los Angeles. Because the 4 recommendations in this report may result in changes to State and Regional Water 5 Quality Control Board regulatory practice, the report is being made available for public 6 review; thus, it is designated the Final Report (Review Copy). Following receipt of 7 comments on the report (due no later than November 15, 2006), this version of the report 8 will be revised to produce a Final Report.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) ii
9 Abstract
10 The purpose of this project, which was funded by the California State Water 11 Resources Control Board (SWRCB), was to evaluate the compliance and wetland 12 condition of compensatory wetland mitigation projects associated with Clean Water Act 13 Section 401 Water Quality Certifications throughout California. This was done by 14 selecting, reviewing and performing field evaluations for 143 permit files distributed 15 across the 12 Water Board regions and sub-regions of the State. For each permit file we 16 assessed the extent to which permittees complied with their mitigation conditions, 17 including acreage requirements, whether the corresponding mitigation efforts resulted in 18 optimal wetland condition, and if the habitat acreages gained through compensatory 19 mitigation adequately replaced those lost through the permitted impacts. We found that 20 permittees are largely following their permit conditions (although one-quarter to one21 third of the time these are not met), but the permit conditions that are being met are not 22 resulting in compensatory mitigation projects that are similar to natural wetlands. 23 Methods
24 Our goal was to evaluate the mitigation actions associated with at least 100 25 randomly chosen Section 401 permit files issued in California between 1991 and 2002. 26 The permit files were selected using the SWRCB’s permit tracking database, and 27 reviewed through multiple visits to the SWRCB, each of the three Army Corps of 28 Engineers district offices (Los Angeles, San Francisco, and Sacramento), and various 29 Regional Boards. Mitigation projects from 129 permit files were visited for assessment 30 of compliance with permit conditions (including acreage) and wetland condition, and 14 31 additional files were evaluated for compliance only. 32 Our determinations of Section 401 compliance included consideration of all 33 mitigation conditions specifically outlined in the 401 permit letter, plus any additional 34 conditions found in other agency permits when the 401 permit included explicit or 35 implicit statements requiring that those documents be followed. In addition to the 36 regulatory permits, the mitigation plan, if present, was carefully read to extract the 37 essential compliance elements. Compliance with these conditions was scored using 38 categorical scores, on a scale from 0% (no attempt to comply) to 100% (condition fully 39 met).
40 To evaluate existing wetland condition, we performed the California Rapid 41 Assessment Method (CRAM) at all assessable mitigation sites associated with our permit 42 files. CRAM includes evaluations of the following attributes: buffer and landscape 43 context, hydrology, physical structure and biotic structure. To provide a sound 44 foundation for evaluating mitigation sites in this study, our mitigation site results were 45 compared to the results from CRAM evaluations performed at 47 reference sites 46 distributed throughout the state.
47 At each mitigation site we also mapped the border of the mitigation sites using 48 GPS to evaluate acreages, and performed qualitative delineations of the sites to determine 49 the approximate proportions of jurisdictional and non-jurisdictional habitat types that Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) iii
50 were present. These proportions, along with the overall site acreages, were used to 51 calculate the component acreages of “waters of the US” versus non-“waters” habitats, 52 wetlands versus non-wetland “waters,” and subsets of these habitat types. These were 53 compared to the impact acreage values in the permits to evaluate “no net loss” from the 54 standpoint of habitat acreages.
55 Results
56 The mitigation sites were well distributed across the state, although some regions 57 had issued relatively few 401 permits and, thus, had correspondingly few site evaluations 58 (Figure AB-1). Of the 129 projects that we evaluated in the field, 62% had onsite 59 mitigation with the rest offsite. Of these projects, 75% had file-specific mitigation, while 60 25% had mitigation that was competed at mitigation banks, were part of other larger 61 projects, or were completed through in-lieu fee payments. 62 In terms of permit compliance, the average compliance score for 401 conditions 63 was 84% (based on 124 files with assessable 401 conditions); 46% of the files fully 64 complied with all permit conditions and the average percent-met score was 73% (Table 65 AB-1). The average compliance score based on mitigation plan requirements (a proxy 66 for all agency requirements) was slightly lower than the 401 compliance scores (81% vs. 67 84%). Only 16% of the files fully complied with all mitigation plan conditions; however, 68 42% had scores of 90% or greater. Compliance with 401 permit conditions showed no 69 trend over time, and there was no significant difference in 401 compliance or mitigation 70 plan compliance among regions. Permittees usually complied with acreage requirements 71 and third party acreage credit purchases, but there was much lower compliance with 72 monitoring and submission requirements (Table AB-2). In general, most 401 permits 73 contained relatively few compensatory mitigation-related permit conditions (often a 74 single acreage-related requirement was specified); conditions regarding success and 75 performance standards were notably infrequent.
76 The 129 files that were evaluated in the field had 204 discrete mitigation sites due 77 to multiple mitigation actions (e.g., wetland creation plus riparian enhancement) that 78 needed to be evaluated separately. Fifty three of these mitigation sites were sub-sampled 79 because they were too large or complex for a single CRAM evaluation. Thus, a total of 80 321 separate CRAM evaluations were completed for this study. 81 Despite relatively high permit compliance, most mitigation sites were not 82 optimally functioning wetlands. As measured by CRAM scores, mitigation sites scored 83 lower than reference wetlands, with an overall mean score of 59% compared to 79% for 84 reference sites (Figure AB-2). On average, sites scored better for biotic structure (e.g., 85 plant community metrics) than for hydrology metrics (Figure AB-3). In comparison to 86 reference sites, only 19% of the mitigation files were classified as optimal, with just over 87 half sub-optimal and approximately one-quarter marginal to poor. There was some 88 variation in CRAM scores among the SWRCB regions, with Region 2 exhibiting a 89 slightly lower mean CRAM score than other regions (Figure AB-4). 90 The 143 Section 401 orders that were evaluated authorized approximately 217 91 acres of impacts (including temporary impacts) and required that 445 acres of mitigation Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) iv
92 be provided. Our analyses indicate that 417 acres of actual mitigation acreage was 93 obtained; 72% of files met or exceeded their acreage requirements, resulting in an overall 94 mitigation ratio of 1.9:1. When considering permanent impacts (true losses) to “creation” 95 mitigation, our results showed that “no net loss” of acreage is being achieved (1) overall, 96 (2) for jurisdictional “waters of the US” acreage, and (3) for wetlands themselves (Table 97 AB-3). However, 39% of individual files resulted in net acreage losses overall, 47% 98 resulted in a net loss of jurisdictional “waters” acreage, and 28% had net wetland losses 99 (Table AB-4).
100 A simple reporting of acreage losses and gains does not provide the full picture of 101 “no net loss.” This approach assumes no existing wetland acreage was present at the 102 mitigation site prior to any mitigation activity (not always the case), it does not address 103 whether the habitat types mitigated were appropriate given the corresponding impacts, 104 and it assumes that the mitigation site exhibited no wetland function prior to the 105 mitigation activities and impacts resulted in a compete loss of function. Assessing this 106 latter issue is challenging in an after-the-fact investigation such as the present study. 107 However, we were able to investigate habitat correspondence in this study. Within most 108 regions, the habitat types mitigated were appropriate given the impacts (Figure AB-5); 109 however, approximately 50% of the mitigation acreage within Regions 4 and 5S 110 consisted of drier riparian and upland habitats that were outside jurisdictional “waters of 111 the US.” Overall, 27% of mitigation acreage was non-jurisdictional. Vague regulatory 112 language and a lack of clear accounting have contributed to this result; in the reporting of 113 regulated impacts, the term “riparian” refers only to habitats within “waters of the US” 114 while in mitigation planning, a more ecological definition of riparian has often been 115 applied that includes the entire zone of transition to fully terrestrial habitats. 116 In comparing results from permit compliance, acreage requirements and wetland 117 condition, we found little relationship between these different aspects of mitigation. 118 Meeting acreage requirements did not ensure overall permit compliance (r²=0.002), nor 119 was there any relationship between percent acreage met and CRAM score for wetland 120 condition (r²=0.015). General compliance with permit conditions was statistically 121 correlated with CRAM scores; however, low r² values indicate the relationships between 122 the variables were not very strong (mean 401 compliance score and CRAM score, 123 r²=0.126 (Figure AB-6); mean percent of 401 conditions met and CRAM score, r²=0.207; 124 and mitigation plan compliance and CRAM score, r²=0.150). 125 Taken together, the findings of this study suggest that permittees are, for the most 126 part, meeting their mitigation obligations, but the ecological condition of the resulting 127 mitigation projects is not optimal (Figure AB-7). The functional deficiencies and the 128 likely failure of many projects to meet the “no net loss” goal of the Clean Water Act are 129 largely due to shortcomings in mitigation planning and in the development of the permit 130 conditions. The root of these shortcomings lies with a lack of explicit consideration of 131 the full suite of functions, values, and services that will be lost through proposed impacts 132 and might be gained through proposed mitigation sites and activities. In short, this is at 133 least partly due to regulatory agencies approving mitigation projects that are too heavily 134 focused on the vegetation component of wetland function, with inadequate emphasis on Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) v
135 compensating for impacts to hydrological and biogeochemical functions and their 136 associated services (e.g., flood attenuation, water quality improvement). 137 Recommendations
138 The results of this study have informed a large number of recommendations 139 (Table AB5). The recommendations are separated into five main categories. 140 First, we present recommendations aimed at improving mitigation requirements. 141 These recommendations concern mainly permit conditions, but also issues of the location 142 of mitigation projects and how gains and losses associated with a project are tracked by 143 habitat. The success of compensatory mitigation depends fundamentally on the 144 mitigation requirements specified by the regulatory agencies. Our study found relatively 145 high levels of compliance with mitigation permit conditions. In addition, there was no 146 relationship between compliance with permit conditions and the condition of wetland 147 mitigation sites. It appears that compliance with permit conditions yields no guarantee 148 that a mitigation wetland will have high condition or function. Perhaps the most effective 149 way to improve the success of compensatory mitigation would be to include permit 150 conditions that lead to better mitigation projects. 151 Second, we present recommendations under the general heading of Information 152 Management. The performance of this study revealed the difficulty of retrieving specific 153 permit files. Of the 429 files we sought, we could locate only 257. The difficulty in 154 locating files had a variety of causes, ranging from limitations in the database to the 155 physical management of hardcopy permit files. These recommendations concern 156 improvements to the database (either the existing database, or a modified database), 157 improvements to permit archiving, and improvements to tracking the progress of 158 mitigation projects.
159 Third, we present recommendations to improve the clarity of permits. Permit 160 conditions should be written as clearly assessable criteria, with individual conditions for 161 each specific criterion to be evaluated. Permit conditions should be written with a clear 162 and direct method of assessment in mind. Our results suggest that more clearly written 163 conditions would improve the chance of compliance. Presently, some conditions are too 164 vague or may be presented in a way that it is not possible to assess them. 165 Fourth, we recommend that the goal of “no net loss” be assessed in a more 166 effective manner. Although we were able to assess whether there has been a net loss of 167 wetland acreage, studies of the functions of wetlands before and after construction at both 168 impact and mitigation sites are required to evaluate the net change in wetland functions. 169 Finally, we present recommendations concerning coordination with other 170 agencies. Although the Water Board has responsibility for 401 permits, the entire 171 process of regulating impacts to wetlands and waters of the United States is closely 172 coordinated with other agencies, especially the U.S. Army Corps of Engineers and the 173 California Department of Fish and Game. Improved information management might 174 improve this coordination.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) vi
175 Compliance Monitoring
176 The results of this study clearly indicate the need to evaluate the compliance of 177 mitigation projects with their permits. Thirteen of the 257 permits we located had to be 178 excluded because of potential compliance issues. This indicates that up to 5% of the files 179 we reviewed may have significant compliance problems (such as the impact occurring 180 but no mitigation being undertaken). Our analysis of discrepancies between 401 permits 181 and information in the permit files identified additional compliance issues. For example, 182 8% of the 143 files we evaluated had information indicating that the actual impacts were 183 greater than authorized in the 401 permit; overall, there appeared to be compliance issues 184 with 42% of the files we evaluated. We found relatively high compliance with third185 party mitigation requirements, but substantial lack of compliance with nearly every other 186 category of permit conditions we assessed. Moreover, many of the categories we 187 assessed had a high fraction of permits for which the conditions could not be assessed; 188 for example, we could not assess monitoring and submission conditions for more than 189 half of the permits.
190 These results indicate a definite need for compliance monitoring. Without a 191 significant compliance effort, permittees are failing to comply with a wide range of 192 permit conditions without the Water Board staff knowing about it. 193 Our data allow us to identify some areas that seem most likely to have low 194 compliance. However, in our view it does not provide a very sharp focus. Compliance 195 issues are spread quite broadly across all aspects of the 401 program, so compliance 196 monitoring will also need to be spread quite broadly. The areas identified as having 197 lower compliance might warrant a particular emphasis during compliance monitoring, but 198 compliance was not so high for most other areas (with the possible exception of third199 party mitigation conditions) that it would be safe to assume high compliance with them. 200 Although monitoring requirements were regularly included as 401 permit 201 conditions, and evaluated for compliance when appropriate, the relative scarcity of 202 monitoring reports in the permit files we reviewed suggest that compliance with the 203 monitoring requirement is checked infrequently. Our compliance assessment indicated 204 that conditions requiring mitigation monitoring were met only about 53% of the time; it 205 was unclear whether any enforcement actions were undertaken in response to the absence 206 of monitoring reports. While we were conducting our study for the Los Angeles 207 Regional Board (Ambrose and Lee 2004), that region was compiling lists of permit files 208 without monitoring reports and contacting permittees to obtain the reports. This seems 209 like a relatively cost-effective area on which to focus compliance monitoring efforts. 210 We make two specific recommendations concerning compliance monitoring. 211 First, we recommend that mitigation monitoring reports should be streamlined and 212 focused around demonstrating compliance with an established list of permit conditions. 213 Second, we recommend that regulatory agencies establish a multi-agency cooperative to 214 monitor compliance and track wetland losses and mitigation success across the State. 215
Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) xiv
Table AB 1. Summary of compliance scores based on 401 and mitigation plan evaluations including average scores and scores for the percentage of conditions met to 100% satisfaction. Successful included files with compliance scores greater than 75%, partially successful included files with scores between 25% and 75%, and failure included files with scores less than 25%.

N Score Successful Partially Successful Failure
Average 401 84.3% 76% 20% 4%
Average 401 percent-met 124 73.3% 57% 40% 13%
Average mitigation-plan 80.7% 68% 32% 0%
Average mitigation plan percent-met 81 67.6% 48% 35% 6% Table AB 2. Compliance breakdowns for 401 compliance grouped by compliance condition category (N=143 files). Condition scores that were not able to be determined were labeled ND, or Not Determinable.
401
Condition
Code Condition Category Total #
Conditions
Average #
Conditions
Average
# ND
Average
Score
1 Third Party 58 1.5 0.1 99.3
2 Acreage 158 1.8 0.2 81.5
3 Site Implementation 411 6.0 2.7 84.8
4 Site Maintenance 49 1.6 0.8 76.0
5 Site Protection 66 1.5 0.6 81.3
6 Success & Performance Standards 199 3.9 1.5 76.4
7 Monitoring & Submission 254 3.6 2.0 59.5
8 Invocation of Other Agency Permits 126 1.7 1.1 N/A
9 Other 35 1.3 0.6 96.1
3 - 6
Site Implementation,
Maintenance, Protection,
Success/Performance Standards
725 3.2 1.4 79.6
Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) xv
Table AB 3. Permanent impacts and created mitigation acreage, including waters of U.S. and non waters of U.S., and wetland, non wetland waters.

Permanent
Impact
Created
Acreage
Proportion
Obtained
Net Acreage
Gain
Gained/Loss
Ratio
Overall Acreage 165.8 270.9 NA 105.1 1.6

Waters of U.S. 162.7 223.1 82.4 60.4 1.4
Non Waters of U.S. 3 47.8 17.6 44.8 NA

Waters of U.S.:

Wetlands 106.3 146.7 66.4 40.4 1.4
Non Wetland Waters 54.9 74.2 33.6 19.3 1.4
Table AB 4. Permanent impacts and created mitigation acreage, including waters of U.S. and non waters of U.S., and wetland, non wetland waters.
% Files w/Gains % Files Gained=Lost % Files w/Loss
Overall Acreage 41 20 39

Waters of U.S. 36 17 47
Non Waters of U.S. 24 76 1

Waters of U.S.:

Wetlands 40 32 28
Non Wetland Waters 17 37 46
Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) xvi
Table AB 5. Summary of administrative and regulatory recommendations. Improving mitigation requirements Information management Improve permit clarity Assessment of “no net loss” Coordination with other agencies Permit conditions should ensure complete compensation for the full suite of wetland functions and services lost
X
Ensure that mitigation projects compensate for losses in water quality (pollution) improvement services
X
There should be a better accounting of the habitat types lost and gained
X
Mitigation projects should have appropriate landscape context X Offsite mitigation should be within the same catchment, or at least the same watershed
X
Improvements to Database X
Improve permit archiving X
Improve tracking the progress of mitigation projects X
Important permit information should be clearly delineated in tables
X
Permit conditions should be written so that the extent of efforts must match the intent of the condition to be in compliance X
Every mitigation plan and permit should include a table of requirements upon which compliance will be judged
X
Permits should be clear about the meaning of enhancement, restoration and creation
X
Performance standards should be clear about the goal of invasive species control
X
Proof of inundation or saturation appropriate for wetland development should be required for mitigation wetlands
X
Pre- and post-construction functional assessments of impact and mitigation sites should be required
X
Improve incorporation of final permit information into Water Board files
X
Consider developing an integrated permit X
Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) xvii
Figure AB 1. Statewide distribution of the assessed mitigation sites associated with the 143 permit files.

Several of these sites, especially those in the central valley (Region 5) involved a collection of shared mitigation banks which resulted in fewer than 143 mitigation sites. Points represent each assessed mitigation site rather than multiple sites per file.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) xviii
Percentage Score
0 10 20 30 40 50 60 70 80 90 100
Percentage of Files / Reference Sites
0 10
20 30
40 50
Reference Site Data
Filewide Data
Poor Marginal Sub-Optimal Optimal
Figure AB 2. All CRAM data combined into a single overall wetland condition success score for each of the 129 files and 47 reference sites evaluated using CRAM.

Overall
Landscape Context
Hydrology
Physical Structure
Biotic Structure
Percentage of Files
0 20
40 60
80 100
Poor/Marginal
Sub-optimal
Optimal
Figure AB 3. Percentage of files in CRAM success categories for overall CRAM scores and the four main attributes. For overall CRAM scores, optimal was considered 70 to 100 percent, sub-optimal was 49 to 70 percent (lower and upper bounds not inclusive), and marginal to poor was 28 percent and below. For buffer and landscape context, optimal was considered 74 to 100 percent, sub-optimal at 52 to 74 percent and marginal to poor 52 percent and below. For hydrology, optimal was considered 76 to 100 percent, suboptimal at 53 to 76 percent and marginal to poor 53 percent and below. For physical structure, optimal was considered 53 to 100 percent, sub-optimal at 38 to 53 percent and marginal to poor 38 percent and below.

For biotic structure, optimal was considered 47 to 100 percent, sub-optimal at 34 to 47 percent and marginal to poor 34 percent and below.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) xix
Region
123456789
Mean Overall CRAM Score
0 20
40 60
80 100
N= 9 25 13 15 34 4 2 13 14
Figure AB 4. Filewide mean Total-CRAM percentage scores by State Board region (N=129 files).

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) xiv
Figure AB 5. Total acreage impacted and obtained proportioned into Wetland, Non-Wetland Waters, Riparian and Upland jurisdictional habitats by state board region. Total required acreage per region is also displayed. N displayed = number of files assessed per region for both impacted and obtained. Total N=138 files (There are five files for which wetland acreage was not specified for waters of the US).

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) xiv
Average 401 Compliance Score
0 20 40 60 80 100
Overall CRAM Scores
0 20
40 60
80 100
Figure AB 6. Correlation analysis between average 401 permit compliance score and overall filewide CRAM score (N= 110 files).

Acreage Requirement
401 Conditions
Mitigation Plan Conditions
CRAM Evaluation
Percent of Files
0 20
40 60
80 100
Percent of Failure Files
Percent of Partially Successful Files
Percent of Successful Files
Figure AB 7. Mitigation success by permit file for each evaluation category: acreage requirement, 401 conditions, mitigation plan conditions, and wetland condition. Data shown for acreage and compliance are percentages out of a total number of 143 permit files. Wetland condition data are % of a total number of 129 files. For the acreage requirements, success was considered 100%, partial success was considered 75- 100% (lower and upper bounds not inclusive), and failure was 75% and below. For the 401 and MP compliance evaluation, success was considered 75-100%, partial success was considered 25-75% (lower and upper bounds not inclusive), and failure was 25% and below. For the CRAM evaluation of wetland condition, success was considered 70-100%, partial success was 49-70% (lower and upper bounds not inclusive), and failure was 28% and below.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) xiv
Table of Contents

1. Introduction................................................................................................................. 1

1.1. Scope and Objectives.......................................................................................... 1

1.2. Previous Studies.................................................................................................. 3

2. Background................................................................................................................. 4

2.1. Definitions and Characteristics........................................................................... 4

2.2. Functions and Services ....................................................................................... 5

2.2.1. Hydrologic Functions.................................................................................. 6

2.2.2. Biogeochemical Functions.......................................................................... 6

2.2.3. Ecological Functions................................................................................... 7

2.3. The Protection of Wetlands ................................................................................ 7

2.4. Clean Water Act Section 404.............................................................................. 8

2.5. Assessing mitigation success .............................................................................. 9

3. Methods..................................................................................................................... 10

3.1. Project Management ......................................................................................... 10

3.2. Permit File Selection and Review..................................................................... 11

3.3. Office Review and Assessment ........................................................................ 12

3.4. Site Visits.......................................................................................................... 12

3.5. Acreage Determinations using GPS.................................................................. 13

3.6. Compliance Evaluations ................................................................................... 15

3.7. Evaluations of Wetland Condition.................................................................... 18

3.7.1. California Rapid Assessment Method (CRAM)....................................... 18

3.7.2. Reference Sites.......................................................................................... 23

3.7.3. Wetland Ecological Assessment............................................................... 24

3.8. Mitigation Habitats Analysis ............................................................................ 25

3.8.1. Jurisdictional Habitat Assessment ............................................................ 25

3.8.2. Habitat Acreage Analysis ......................................................................... 26

3.9. Digital Photographs .......................................................................................... 27

3.10. Data Management and Analysis ................................................................... 27

3.11. Quality Assurance/Quality Control............................................................... 28

4. Results....................................................................................................................... 31

4.1. Permit Review................................................................................................... 31

4.1.1. Discrepancies between file information and SWRCB database............... 35

4.2. Status of Regulatory Compliance of Compensatory Mitigation Sites.............. 38

4.3. Function and Condition of Compensatory Mitigation Sites ............................. 44

4.3.1. Total-CRAM Scores ................................................................................. 45

4.3.2. CRAM Attribute Scores............................................................................ 46

4.3.3. Individual CRAM Metrics ........................................................................ 48

4.3.4. Wetland Class ........................................................................................... 49

4.4. Habitat Acreage Analysis ................................................................................. 49

4.4.1. Riparian Jurisdictional Issues ................................................................... 50

4.4.2. Permanent vs. Temporary Impacts ........................................................... 52

4.4.3. Regional Comparisons.............................................................................. 53

4.5. Combined Acreage, Compliance and CRAM Results...................................... 54 Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) xv

5. Conclusions............................................................................................................... 56

5.1. Permit Compliance............................................................................................ 57

5.2. Wetland condition............................................................................................. 58

5.2.1. Permit conditions ...................................................................................... 59

5.3. Changes in habitat types and acreage ............................................................... 61

5.4. Differences among regions ............................................................................... 62

5.5. Mitigation banks ............................................................................................... 63

5.6. Evaluating “no net loss”.................................................................................... 64

6. Recommended Administrative and Regulatory Changes ......................................... 66

6.1. Improving Mitigation Requirements................................................................. 66

6.1.1. Permit conditions should ensure complete compensation for the full suite of wetland functions and services lost. ..................................................................... 67

6.1.2. Ensure that mitigation projects compensate for losses in water quality (pollution) improvement services ............................................................................. 68

6.1.3. There should be a better accounting of the habitat types lost and gained. 70

6.1.4. Mitigation projects should have appropriate landscape context............... 71

6.1.5. Offsite mitigation should be within the same catchment, or at least the same watershed. ........................................................................................................ 72

6.2. Information Management Recommendations................................................... 73

6.2.1. Improvements to Database........................................................................ 73

6.2.2. Improve permit archiving ......................................................................... 77

6.2.3. Improve tracking the progress of mitigation projects............................... 78

6.3. Improve permit clarity ...................................................................................... 78

6.3.1. Important permit information, including impact and mitigation acreage and permit conditions, should be clearly delineated in tables and not buried within the permit text. .......................................................................................................... 79

6.3.2. Permit conditions should be written so that efforts made in a small portion of the site cannot satisfy the verbatim text of the condition when the intention of the condition was that the efforts would be made throughout the site. .......................... 79

6.3.3. Mitigation plans (and perhaps all permits) should include a table listing the requirements upon which compliance will be judged......................................... 79

6.3.4. Permits should be clear about the meaning of enhancement, restoration and creation............................................................................................................... 80

6.3.5. When invasive species removal is required, performance standards should be clear about the goal of invasive species control................................................... 82

6.3.6. If a wetland is planned as part of a mitigation project, proof of inundation or saturation appropriate for wetland development should be required. .................. 82

6.4. Improve the assessment of “no net loss” .......................................................... 82

6.4.1. Pre- and post-construction functional assessments of impact and mitigation sites should be required to ensure no net loss of wetland functions........ 82

6.5. Coordination with other agencies ..................................................................... 83

6.5.1. Improve incorporation of final permit information into Water Board files ...................................................................................................... 83

6.5.2. Consider developing an integrated permit ................................................ 84

7. Recommended Compliance Monitoring Program .................................................... 84

7.1. The need for compliance monitoring................................................................ 84 Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) xvi

7.2. How should compliance monitoring efforts be focused? ................................. 85

7.2.1. Frequency of compliance monitoring ....................................................... 86

7.3. Specific monitoring recommendations ............................................................. 88

7.3.1. Mitigation monitoring reports should be streamlined and focused around demonstrating compliance with an established list of permit conditions. ................ 88

7.3.2. Form a multi-agency cooperative for compliance monitoring and project tracking. ................................................................................................................... 88

8. Literature Cited ......................................................................................................... 90

9. Tables........................................................................................................................ 94

10. Figures................................................................................................................. 110 Appendix 1. Detailed permit file selection methodology.............................................154 Appendix 2. Lists of assessed files by file identification number ................................169 Appendix 3. Detailed discrepancy analysis Results .....................................................179 Appendix 4. GPS information ......................................................................................197 Appendix 5. Distribution of sites within regions..........................................................202 Appendix 6. Detailed permit compliance assessment methodology ............................214 Appendix 7. Supplemental CRAM results ...................................................................226 Appendix 8. CRAM by wetland class Results and Discussion ....................................252 Appendix 9. Mitigation bank analysis..........................................................................258 Appendix 10. Wetland ecological assessment (WEA) analysis .....................................266 Appendix 11. Detailed habitat acreage analysis results..................................................276 Appendix 12. Site narratives...........................................................................................286 Appendix 13. Digital images of sites..............................................................................390 Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) xvii
List of Tables
Table 1. Reference Site information ................................................................................ 95 Table 2. Jurisdictional habitat hierarchy.......................................................................... 96 Table 3. Overall summary of the permit file selection results by region......................... 96 Table 4. Number of onsite and offsite mitigation sites for file specific mitigation actions, formal mitigation banks, informal mitigation banks, and in lieu fees. ..................... 97 Table 5. Summary of the discrepancies between the impact and required mitigation acreage values obtained through our detailed permit reviews and the corresponding values in the State Board’s permit tracking database. Multiple discrepancy categories may apply to a particular file................................................................... 97 Table 6. Summary of compliance scores based on 401 and mitigation plan evaluations including average scores and scores for the percentage of conditions met to 100% satisfaction. ............................................................................................................... 98 Table 7. Compliance breakdowns for 401 and Mitigation Plan compliance grouped by compliance condition category (N=143 files). ......................................................... 99 Table 8. Summary statistics of mitigation CRAM scores (N=129) and reference site CRAM scores (N=47) for Total-CRAM scores and the four attributes, along with the percentage of files within each success category.................................................... 100 Table 9. Summary statistics and success breakdowns of Total-CRAM scores by SWRCB region (N=129 files)................................................................................................ 100 Table 10. Summary statistics and success breakdowns of landscape context metrics CRAM scores by SWRCB region (N=129 files).................................................... 101 Table 11. Summary statistics and success breakdowns of hydrology metrics CRAM scores by SWRCB region (N=129 files)................................................................. 101 Table 12. Summary statistics and success breakdowns of physical structure metrics CRAM scores by SWRCB region (N=129 files).................................................... 102 Table 13. Summary statistics and success breakdowns of biotic structure metrics CRAM scores by SWRCB region (N=129 files)................................................................. 102 Table 14. Summary statistics and success breakdowns of CRAM scores by individual CRAM metric (N=204 mitigation sites). ................................................................ 103 Table 15. Total impacted and obtained acreage for all files (overall), waters of U.S. and Non waters of U.S., wetland, and non wetland waters. .......................................... 104 Table 16. Permanent impacts and created mitigation acreage, waters of U.S. and non waters of U.S., and wetland, non wetland waters. .................................................. 104 Table 17. Total impacted and obtained acreage for all files (overall), waters of U.S. and Non waters of U.S., wetland, and non wetland waters. .......................................... 105 Table 18. Permanent impacts and created mitigation acreage, waters of U.S. and non waters of U.S., and wetland, non wetland waters. .................................................. 105 Table 19. Mitigation success by permit file for each evaluation category: acreage requirement, 401 conditions, mitigation plan conditions, and wetland condition.. 106 Table 20. Acreage, compliance, and CRAM summaries by permittee type. These permittee type categories were taken directly from the 401 Permit Files............... 107 Table 21. Summary of administrative and regulatory recommendations...................... 108 Table 22. Suggested jurisdictional and non-jurisdictional habitat hierarchy, with structure for tracking losses and gains................................................................................... 109 Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) xviii
List of Figures
Figure 1. Map of California state board regions with breakdown of number of permit files.......................................................................................................................... 111 Figure 2. Statewide distribution of reference sites.......................................................... 112 Figure 3. Statewide distribution of the assessed mitigation sites associated with the 143 permit files. ............................................................................................................. 113 Figure 4. Files assessed fully and for compliance only by state board region............... 114 Figure 5. Statewide distribution of the impact and mitigation sites associated with the 143 permit files assessed......................................................................................... 115 Figure 6. Percentage of applications per certification year listed in the SWRCB database from 1991 to 2002 compared with the percentage of files per year in our sample of files assessed fully and for compliance only (N for files assessed=143, N for SWRCB database=9924). ....................................................................................... 116 Figure 7. Percentage of files assessed by permittee type (N=143 files). ....................... 116 Figure 8. Breakdown of the 143 assessed files by habitat type impacted as reflected by the SWRCB database, and by our detailed permit reviews. ................................... 117 Figure 9. Acres impacted and acres of mitigation required displayed by acreage-size categories using data from project analyses for files assessed (N=143)................. 117 Figure 10. Breakdown of the 143 assessed permit files by permanent and temporary impacts as reflected by the SWRCB database, and by our detailed permit reviews. ................................................................................................................................. 118 Figure 11. Acres impacted and acres of mitigation required displayed by certification year from the project analyses for files assessed (N=143). .................................... 119 Figure 12. Average mitigation ratios required by certification year as determined from our detailed permit file review (N=143). ................................................................ 119 Figure 13. Acres impacted and acres of mitigation required displayed by state board region from the project analyses for files assessed (N=143). ................................. 120 Figure 14. Mitigation ratios required by region (N=143).............................................. 120 Figure 15. Plot of the differences between the impacted and required acreage values obtained through our detailed file review, and the corresponding values recorded in the SWRCB database.............................................................................................. 121 Figure 16. Distribution of files according to the average 401 permit compliance score and 401 percent-met score (N=124 files with assessable 401 permit conditions).. 122 Figure 17. Distribution of files according to the average mitigation plan compliance score and mitigation plan percent-met score (N=81 files with assessable mitigation plan conditions)....................................................................................................... 122 Figure 18. Distribution of files according to the percentage of 401 permit compliance conditions that could not be determined (N=124 files with assessable 401 permit conditions)............................................................................................................... 123 Figure 19. Distribution of files according to the percentage of mitigation plan compliance conditions that could not be determined (N=81 files with assessable mitigation plan conditions). .................................................................................... 123 Figure 20. Relationship between 401 certification year and average 401 permit compliance score (N= 124 files with assessable 401 permit conditions; p=0.845, r²=0.000). ................................................................................................................ 124 Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) xix
Figure 21. Relationship between 401 certification year and average mitigation plan compliance score (N= 81 files with assessable mitigation plan conditions; p=0.119, r²=0.030). ................................................................................................................ 124 Figure 22. Average percentage score for 401 permit compliance by state board region (N=124 files with assessable 401 permit conditions). ............................................ 125 Figure 23. Average percentage score for mitigation plan compliance by state board region (N=81 files with assessable mitigation plan conditions). ............................ 125 Figure 24. Average 401 score by certification type (N=143 files). ............................... 126 Figure 25. Average scores for 401 permit compliance and average percentage of conditions that could not be determined grouped by the type of permit condition (N=124 files with assessable 401 permit conditions). ............................................ 127 Figure 26. Average scores for mitigation plan compliance and average percentage of conditions that could not be determined grouped by the type of permit condition. (N = 81 files with assessable mitigation plan conditions). .......................................... 127 Figure 27. Breakdown of the number of mitigation-related permit requirements (conditions) in each 401 permit order (N=143). ..................................................... 128 Figure 28. Mean number of mitigation-related 401 conditions per order within each SWRCB Region, including standard error bars (N=132). Eleven files for which no 401 permit was obtained were excluded................................................................. 128 Figure 29. Breakdown of all mitigation-related 401 permit conditions by condition category (N=132).................................................................................................... 129 Figure 30. Mean number of mitigation-related 401 permit conditions per permit order (N=132)................................................................................................................... 130 Figure 31. Frequency of occurrence for the eight permit condition categories when the 401 order includes just a single mitigation-related condition (N=36, 23, 18, 14, respectively)............................................................................................................ 130 Figure 32. Percentage of mitigation-related conditions found in 401 permit orders that were unique to the 401, redundant with equivalent conditions required by other regulatory agencies, or invoking those other agency permits or the common mitigation plan (i.e., “must follow the 404”) (N=115). .......................................... 131 Figure 33. Percentage of redundant and invoking 401 conditions by Region (N=115). 131 Figure 34. Distribution of files according to the average 401 permit compliance score including only those mitigation conditions explicitly specified in the 401 permit order (N=143). ........................................................................................................ 132 Figure 35. Breakdown of wetland hydrogeomorphic classes as defined and assessed by the CRAM evaluations for all 204 mitigation sites representing 129 files evaluated using CRAM. .......................................................................................................... 132 Figure 36. Distribution of assessed mitigation sites by wetland class across the state.. 133 Figure 37. All CRAM data combined into a single overall wetland condition success score for each of the 129 files and 47 reference sites evaluated using CRAM. ..... 134 Figure 38. Map of California showing location of mitigation sites color coded by condition score........................................................................................................ 135 Figure 39. Relationship between 401 certification year and filewide mean overall CRAM percentage scores grouped by certification year (N=129 files, r2 =0.005, p=0.415).
................................................................................................................................. 136 Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) xx
Figure 40. Filewide mean Total-CRAM percentage scores by SB region (N=129 files). ................................................................................................................................. 136 Figure 41. Percentage of files in CRAM success categories by state board region (N=129 files). ....................................................................................................................... 137 Figure 42. Landscape context metric CRAM scores compared to reference-site data.. 138 Figure 43. Hydrology metric CRAM scores compared to reference-site data. ............. 138 Figure 44. Physical structure metric CRAM scores compared to reference-site data. .. 139 Figure 45. Biotic structure metric CRAM scores compared to reference-site data....... 139 Figure 46. Mean percentage scores for each CRAM metric for mitigation sites (N=204) and reference sites (N=47)...................................................................................... 140 Figure 47. Mean percentage scores for each CRAM metric by state board region. (N=204 mitigation sites) ......................................................................................... 141 Figure 48. Overall CRAM percentage scores by wetland class (N=204 mitigation sites). ................................................................................................................................. 142 Figure 49. Overall acreage obtained compared to required and impacted (N=143 files). ................................................................................................................................. 143 Figure 50. Acreage required and obtained by year (N=143 files). ................................ 143 Figure 51. Average mitigation ratios of required and obtained acreage by certification year as determined from our detailed permit file review........................................ 144 Figure 52. Acreage required and obtained by region..................................................... 144 Figure 53. Total acreage impacted, required and obtained for 143 files assessed.

Acreage also grouped by jurisdictional habitat classifications: “Waters of the US” and non-jurisdictional waters (“Non-Waters”). ...................................................... 145 Figure 54. Total acreage impacted and obtained, with jurisdictional habitats data for “Waters of the US” proportioned into wetland and non-wetland waters habitats, and data for “Non-Waters” proportioned into riparian and upland habitats.................. 145 Figure 55. Total acreage impacted proportioned into permanent and temporary impacts, and obtained acreage proportioned into created, enhanced and preserved, each proportioned further into Waters of the US and Non-Waters of the US (N=143 files). ................................................................................................................................. 146 Figure 56. Total acreage for Wetland and Non-Wetland Waters, each displaying impacted and obtained acreage. Impacted acreage is proportioned into permanent and temporary impacts, while obtained acreage is proportioned into created, enhanced and preserved. ......................................................................................... 146 Figure 57. Total acreage impacted and obtained proportioned into Waters of the US and Non-Waters of the US by state board region (N=143 files). .................................. 147 Figure 58. Total acreage impacted and obtained proportioned into Wetland, NonWetland Waters, Riparian and Upland jurisdictional habitats by state board region.

Total required acreage per region is also displayed................................................ 148 Figure 59. Correlation analysis between percentage of acreage requirement met and average 401 permit compliance score (N=123 files). ............................................. 149 Figure 60. Correlation analysis between percentage of acreage requirement met and overall filewide CRAM score (N=128 files). ......................................................... 149 Figure 61. Correlation analysis between average 401 permit compliance score and overall filewide CRAM score (N= 110 files). ........................................................ 150 Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) xxi
Figure 62. Correlation analysis between percentage of 401 permit conditions met and overall filewide CRAM score (N=110 files). ......................................................... 150 Figure 63. Correlation analysis between average mitigation plan compliance score and overall filewide CRAM score (N=77 files). ........................................................... 151 Figure 64. Correlation analysis between overall filewide CRAM percentage score and average 401 permit compliance score for four of the permit condition categories. 152 Figure 65. Total impacted acreage and obtained acreage weighted by condition score (N=129 files)........................................................................................................... 153 Figure 66. Mean 401 compliance score for different wetland types.............................. 153 Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 1 1 1. Introduction
2 For about the last quarter century, the principle regulatory mechanism for the 3 protection of wetland habitats has been Section 404 of the Clean Water Act (CWA). 4 Every applicant for a 404 permit must also obtain state CWA Section 401 certification 5 that the proposed discharge will not violate state water quality standards. In California 6 the State Water Resources Control Board issues certifications for multi-Regional 7 projects, and Regional Water Quality Control Boards issue certifications for projects 8 entirely within their jurisdiction. In addition, if the work will involve impacts to a 9 streambed, a Streambed Alteration Agreement must be obtained from the State 10 Department of Fish and Game (DFG), and if there are threatened or endangered species 11 issues, the US Fish and Wildlife Service and /or DFG may issue permits under the federal 12 or State endangered species acts. Since about 1990, these regulatory agencies have 13 pursued a State and National goal of “no net loss” of wetland acreage and function. 14 Given this goal, any wetland losses that do occur must be offset through compensatory mitigation actions.1 15 Within the regulatory framework, a strong emphasis has been placed 16 on the avoidance and minimization of proposed impacts. However, the majority of CWA 17 Section 404 proposals are ultimately approved (NRC 2001), making mitigation for 18 permitted wetland impacts essential for the protection of wetland function. 19 1.1. Scope and Objectives
20 Recognizing the importance of compensatory mitigation in achieving “no net 21 loss” and, more generally to assure compliance with regulatory mandates, the SWRCB 22 contracted with the University of California, Los Angeles to conduct this study. The 23 scope and objectives of the contract were:
24 Beneficial uses of wetlands and riparian area in California have been 25 heavily impacted by a variety of projects, with more than 90% of 26 California’s wetlands and riparian areas lost. California’s Wetland 27 Conservation Policy establishes a “no net loss – long term gain” goal for 28 wetland quantity, quality, and permanence (Executive Order W-59-93). The 29 main tool used by the State Water Resources Control (State Board) and the 30 Regional Water Quality Control Boards (Regional Boards) to protect 31 wetlands and riparian areas is the Clean Water Act (CWA) §401 Water 32 Quality Certification (WQC) Program. Section 401 WQC is associated 33 with CWA §404 permits issued by the US Army Corps of Engineers 34 (USACE). A principal means to achieve the “no net loss” goal is the 35 requirement for compensatory mitigation when unavoidable impacts to 36 wetlands and riparian areas occur.
37 Successful compensatory mitigation is technically complex, usually 38 takes years to achieve, and can be expensive. Thus there is a real danger of 39 failure, and a financial incentive for dischargers to avoid or minimize the 40 necessary costs. These considerations argue for an effective compliance

1 Compensatory mitigation is the creation, restoration, enhancement, or occasionally, preservation of wetland resources either onsite or offsite to offset permitted losses in wetland acreage and/or function.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 2 41 mitigation program for compensatory mitigation projects. However, due to 42 staffing constraints, the Regional Boards perform little or no such 43 compliance monitoring. A second concern is that regulatory conditions, 44 even if complied with, may not assure reestablishment of beneficial use 45 quality or permanence. The National Academy of Sciences, in a 46 2001comprehensive review of wetland compensatory mitigation in the U.S. 47 found that the national “no net loss” goal is not being met because (1) there 48 is little monitoring of permit compliance, and (2) the permit conditions 49 commonly used to establish mitigation success do not assure the 50 establishment of wetland functions. The San Francisco Estuarine Institute 51 and the Southern California Coastal Water, working with other concerned 52 State and federal agencies, have developed a California Rapid Assessment 53 Method (CRAM) for assessment of wetland condition. A third concern is 54 that, because we have not integrated compliance monitoring into our routine 55 regulatory practice, the State and Regional Board’s administrative and 56 regulatory procedures may not adequately support effective and efficient 57 compliance monitoring of compensation sites.
58 The objectives of this project are to: (1) determine project-specific 59 and regional compliance with regulatory requirements, (2) assess wetland 60 function and condition at the compensatory mitigation sites, (3) improve 61 administrative and regulatory practice for establishing and monitoring 62 conditions to regulate compensatory mitigation, and (4) determine the need 63 for ongoing compliance monitoring.
64 Compensation sites in the North Coast, San Francisco Bay, Central 65 Coast, Los Angeles, Central Valley, Lahontan, Santa Ana, Colorado Basin, 66 and San Diego Regional Board jurisdictions will be considered for the 67 study.
68 The purpose of this project was to evaluate the compliance and wetland condition 69 of compensatory wetland mitigation projects associated with §401 Water Quality 70 Certifications throughout California. This was done by selecting, reviewing and 71 performing field evaluations for nearly 150 permit files distributed across the 12 Water 72 Board regions and sub-regions of the State. For each permit file we assessed the extent to 73 which permittees complied with their mitigation conditions, including acreage 74 requirements, whether the corresponding mitigation efforts resulted in optimal wetland 75 condition, and if the habitat acreages gained through compensatory mitigation adequately 76 replaced those which were lost through the permitted impacts. 77 The Water Boards’ 401 Program was established in 1990. During the period from 78 which permits were evaluated (1991-2002) and continuing to the present, the 401 79 Program has evolved. A major change was the adoption of new Program regulations, 80 which became effective on June 24, 2000. The new regulations specified the information 81 to be included in an application for certification, eliminated the possibility of waiving 82 certification, identified standard conditions to be included in all certifications, and 83 generally systematized the processing of applications. In addition, regulatory practice Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 3 84 has evolved as field staff have acquired experience with the Program. This study 85 presents analysis of data representing historical practice over the study period. 86 1.2. Previous Studies
87 Wetland mitigation has been the focus of many critical studies (see Race 1985, 88 Zentner 1988, Kentula et al. 1992, Holland and Kentula 1992, DeWeese and Gould 1994, 89 Miller 1995, Mitsch and Wilson 1996, Zedler 1996, Race and Fonseca 1996, Gilman 90 1998, Breaux and Serefiddin 1999, Gwin et al. 1999, Ambrose 2000, Brown and 91 Veneman 2001, Kelly 2001). In 2001, a panel convened by the National Academy of 92 Sciences completed a comprehensive review of compensatory wetland mitigation in the 93 U.S. (NRC 2001).
94 The work reported here follows from a number of previous studies focusing on 95 Section 404 permits. Mary Kentula and her colleagues have conducted a series of studies 96 exploring the effectiveness of Section 404 permitting in the United States (Kentula et al. 97 1992, Holland and Kentula 1992, Sifneos et al. 1992a, 1992b), including California. 98 These studies relied solely on office reviews of permit files. In general, these studies 99 have reported that Section 404 permits have not prevented the continued loss of wetland 100 habitat in the U.S. However, office reviews of permit files are necessarily limited to the 101 intent rather than actual implementation of mitigation. To remedy this limitation, a 102 number of studies have assessed actual compliance with permit conditions in the field 103 (see NRC 2001). In California, for example, DeWeese and Gould (1994) found 50% of 104 the projects evaluated achieved at least 75% compliance with stated permit conditions, 105 while Allen and Feddema (1996) identified a compliance rate of 67% in Southern 106 California. Several studies have suggested that increased enforcement of mitigation 107 permits would improve compliance with permit conditions (Holland and Kentula 1992, 108 Sifneos et al. 1992a, DeWeese and Gould 1994).
109 A few studies have gone beyond compliance assessment to evaluate ecological 110 condition or functions of mitigation sites. The NRC report summarizes 11 of these 111 studies. The most relevant for our work was conducted by Mark Sudol in southern 112 California (Sudol 1996, Sudol and Ambrose 2002). Sudol reviewed Section 404 and 113 Section 10 permits for Orange County and conducted field assessments of each 114 mitigation site to evaluate its compliance with permit conditions as well as how well the 115 wetland performed certain functions (as indicated by the Hydrogeomorphic Assessment 116 Methodology (Brinson 1993)). Sudol found 18% of the mitigation sites complied fully 117 with their permit conditions, but that none of the sites had appropriate levels of wetland 118 function. The joining of an office review of permits with field assessments of permit 119 compliance and wetland function/condition is a powerful combination (Sudol and 120 Ambrose 2002), and provided the model for the approach adopted in this study. 121 Most of these previous studies have focused on mitigation success solely with 122 respect to the Section 404 permit conditions, without considering the contributions of 123 other agencies involved in the greater regulatory process. In particular, few have 124 investigated the successes and failures of mitigation projects with respect to the permit 125 conditions of the Section 401 Water Quality Certification orders. Breaux et al. (2005) 126 studied mitigation success for 20 projects near San Francisco Bay which had been Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 4 127 regulated under the 401 and 404 programs by the local Regional Water Quality Control 128 Board and Corps district, respectively. They found that most projects were in compliance 129 with their permit conditions and were realizing their intended habitat functions. They 130 reported increased habitat functional success at larger sites and argued that regulators 131 should favor regionally integrated mitigation banks because of their improved benefits to 132 wildlife. In a similar study commissioned by the Los Angeles Regional Water Quality 133 Control Board, Ambrose and Lee (2004) investigated this issue within the Los 134 Angeles/Ventura area by evaluating the mitigation projects associated with 135 approximately 55 Section 401 permits issued by that Regional Water Board. For those 136 projects, they found that the assessable 401 permit conditions were mostly being 137 complied with, yet very few mitigation projects could be considered optimally 138 functioning wetlands. About half of the total mitigation acreage consisted of drier 139 riparian and upland habitats that were outside of jurisdictional waters of the United 140 States; about two-thirds of the projects did not fully replace the functions lost, and, thus, 141 “no net loss” was not being achieved. The present study would help determine if the 142 findings of Ambrose and Lee (2004) are unique to the Los Angeles/Ventura Region, or if 143 they reflect mitigation success statewide.
144 2. Background
145 2.1. Definitions and Characteristics
146 Definitions of wetlands and riparian areas vary widely among different groups 147 and for different purposes. A recent NRC panel defined a wetland as below, based not on 148 regulatory requirements but a consensus of wetland scientists; this definition provides 149 context for the important benefits that wetland ecosystems provide:
150 An ecosystem that depends on constant or recurrent, shallow inundation or 151 saturation at or near the surface of the substrate, and the presence of 152 physical, chemical, and biological features reflective of that regime, such 153 as hydric soils and hydrophytic vegetation (adapted from NRC 1995). 154 In general, wetlands are characterized by the presence of biophysical gradients 155 between aquatic and terrestrial habitats and include freshwater marshes, tidal salt 156 marshes, riverine floodplains, riparian wetlands, mangroves, and several types of 157 depressional wetlands. These can be grouped into estuarine (tidal salt marshes), riverine 158 (floodplains and riparian areas), lacustrine (lake affiliated), or palustrine (freshwater 159 marshes and bogs) wetlands. The biological communities present at the various wetlands 160 can take many forms, but one of their predominant characteristics is the presence of 161 hydrophilic (water-loving) vegetation.
162 While the preceding characterization of wetlands reflects an ecological 163 perspective, more restrictive definitions are used for regulatory purposes, with the 164 specific definition depending on the regulatory agency. Of most relevance for this study, 165 wetlands as defined by the U.S. Army Corps of Engineers (USACE) must generally meet 166 a three-parameter test, having appropriate hydrology, hydric soils, and wetland 167 vegetation. According to the USACE, wetlands are defined as:

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 5 168 those areas that are inundated or saturated by surface or ground water at a 169 frequency and duration sufficient to support, and that under normal 170 circumstances do support, a prevalence of vegetation typically adapted for 171 life in saturated soil conditions. Wetlands generally include swamps, 172 marshes, bogs, and similar areas.
173 (More discussion of jurisdictional habitats under the Clean Water Act is given later; see 174 page 25) Many of the activities requiring Section 401 and 404 permits also affect habitat 175 adjacent to jurisdictional wetlands, including non-wetland riparian habitats. Although 176 non-wetland riparian habitat is regulated under the California Department of Fish and 177 Game’s Streambed Alteration agreements, it is outside the jurisdiction of the Corps of 178 Engineers and the Regional Water Quality Control Boards. 179 Riparian habitats are defined in a non-regulatory sense as those areas that are 180 transitional between terrestrial and aquatic ecosystems and are distinguished by gradients 181 in biophysical conditions, ecological processes, and biota (NRC 2002). They are areas 182 through which surface and subsurface hydrology connect waterbodies with their adjacent 183 uplands (NRC 2002). Riparian areas include those areas that are adjacent to perennial, 184 intermittent, and ephemeral streams, lakes, or estuarine-marine shorelines. These habitats 185 often line the margins or banks of streams and lakes and are characterized by the 186 presence of low-growing hydrophytic herbs, shrubs, and tall woody trees. Much of the 187 difference in the regulatory versus ecological definitions of wetlands that we have 188 encountered in this study relates to variations in the definition of riparian areas. 189 2.2. Functions and Services
190 Human activities have encroached on wetlands and river systems. Vast, low191 lying riverine floodplains and coastal wetlands have been key targets for human 192 development because of the relative ease of reclamation and because of their associated 193 fertile soils. These complex drainage systems have often been reduced to straightened 194 channels with tall constructed banks or levees, designed to contain high flood waters. In 195 addition, isolated wetlands have commonly been drained and filled, or converted to 196 livestock watering areas. The result of these impacts has been the diminishment of the 197 beneficial services that these wetland habitats provide (NRC 1995; NRC 2001; NRC 198 2002; Leibowitz 2003), and humans are now beginning to recognize the consequences of 199 their loss. As a result, much of the focus of concern about the loss of wetland habitats 200 revolves around the loss of functions and services they provide.

The functions and services2 201 that wetlands and riparian areas provide fall into three 202 broad categories: hydrology and sediment dynamics, biogeochemistry and nutrient 203 cycling, and habitat and food web support. Each wetland type performs characteristic 204 functions; no particular wetland performs all possible functions. A brief description of 205 wetland functions and services follows; this is a simple overview and not a detailed 206 catalog of all functions and services performed by wetlands.

2 “Functions” refers to natural processes occurring in wetlands; “services” refers to processes or attributes of wetlands that are useful to humans.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 6 207 2.2.1. Hydrologic Functions
208 Water flowing along the surface of the earth naturally flows downhill towards 209 lower areas of the terrain and begins to accumulate in rills, rivulets, streams, and 210 ultimately river channels as it makes its way to the ocean. Water infiltrating into the 211 earth will also flow down-gradient through the interstitial spaces in the soil or rock, 212 eventually emerging back at the surface in topographically lower areas. These areas 213 where the ground water table emerges are commonly adjacent to or within stream 214 channels. The hydraulic connectivity between precipitation source areas and re215 emergence areas results in increased groundwater contributions to streams following 216 storm events, though there is usually a modest time lag and great modulation of flow. 217 The combined flow from overland runoff and emerging groundwater following a storm 218 event results in a pulsed stream discharge pattern with peak flood levels occurring some 219 time after the point of maximum precipitation. Sediment is also a significant proportion 220 of storm runoff as soil eroded from adjacent hillsides enters the stream along with the 221 storm water (Knighton 1998). The destructive force of the storm flow reaches the 222 maximum at the peak of discharge, and these peak flows are what human management 223 strategies have tried to accommodate through the construction of tall levees and often224 straight concrete channels. The general philosophy has been to move the water to the 225 ocean as fast as possible, to minimize flooding during peak flows. 226 But the natural geomorphology of river channels has developed to accommodate 227 these peak flows with appropriately wide floodplains and adjacent wetlands, which serve 228 to modulate high water flow through the short term storage of water and sediment 229 (Knighton 1998). During high flow events, water flows over the banks of the natural 230 channel and spreads out over floodplains, where the velocity is reduced and the sediment 231 settles out. Water percolates into soils and sediments within floodplains and riparian 232 areas, where it is stored until the flow recedes. Then the water slowly flows back out 233 during periods of low flow, helping to maintain baseflow conditions during the dry 234 season. Isolated depressional wetlands collect some of the water that would otherwise 235 flow directly to the stream, thus contributing to the moderation of storm flow and the 236 recharge of ground water. In addition, the vegetation that occurs on floodplains and in 237 riparian zones provides mechanical flow reduction and energy dissipation of high flow, 238 and riparian trees, shrubs, and grasses contribute to the stabilization of the stream banks. 239 Often, the absence of riparian vegetation on the banks can lead the destabilization of the 240 banks and their subsequent erosion and incision, though the presence of riparian trees 241 may contribute to bank erosion in other circumstances (Lyons et al. 2000). 242 2.2.2. Biogeochemical Functions
243 Biogeochemical functions in wetlands and riparian areas include the retention and 244 removal of substances from the water, sediment accumulation, and nutrient cycling, 245 among others. All of these result in the overall maintenance of water quality. For 246 example, a riparian buffer zone located between an agricultural area and a stream channel 247 can absorb much of the nutrients leaching from a nearby agricultural field through either 248 surface flow or through the groundwater (NRC 2002). These nutrients can become 249 adsorbed by hydric riparian soils or may be assimilated by riparian vegetation, thus 250 minimizing their transport to the stream. In many agricultural areas, the absence of a Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 7 251 riparian buffer may result in direct inputs of nutrients to the stream, in which case 252 instream wetland conditions become very important with respect to improving water 253 quality. Many biogeochemical reactions are redox dependent. That is, certain reactions 254 occur in the presence of oxygen while others require the absence of oxygen. Many of the 255 beneficial reactions that contribute to the improvement of water quality require the 256 absence of oxygen and are common in anaerobic wetland soils. 257 2.2.3. Ecological Functions
258 Wetlands are extremely important habitats for migratory birds, which use them 259 for resting and feeding areas as they travel from place to place or for breeding. Wetlands 260 and riparian areas are also important to many other species of plants and animals, 261 including threatened and endangered species, and can be areas of notably high 262 biodiversity. For example, riparian habitats in the Santa Monica Mountains cover less 263 than 1% of the land area yet are the primary habitat for 20% of the higher plant species 264 (Rundel 2002). In today’s heavily fragmented landscape, riparian areas can be extremely 265 important corridors for the movement of animals. Many isolated wetlands that become 266 dry during part of the year means cannot support fish species, making them important 267 habitats for reptiles and amphibians that would otherwise be preyed upon by fish 268 (Gibbons 2003). Further, riparian trees and other vegetation perform important shading 269 functions, providing significant thermal regulation for the community by keeping water 270 and air temperatures cool during warm dry periods.
271 2.3. The Protection of Wetlands
272 When Europeans first arrived in North America, the vast amount of dense 273 woodland and wetland habitat constituted substantial impediments to the settlement of 274 the land (Hawke 1989). Throughout most of our nation’s history, the federal government 275 actively encouraged the conversion of wetlands for useful purposes and for disease 276 abatement, as evidenced by legislation such as the Federal Swamp Land Act of 1850, 277 which promoted their conversion to agricultural land (NRC 1995). The notion that 278 wetlands perform functions or services that can be beneficial to the greater human society 279 has only taken root within the last several decades. Among the suite of landmark 280 environmental laws passed in late 1960’s and early 1970’s was the Clean Water Act, 281 which had the ambitious goal “to protect the physical, chemical and biological integrity 282 of the nation’s waters” (NRC 2001).
283 While the main focus of the Clean Water Act was to prevent water pollution, 284 some aspects of this law extended protection to wetlands, and these remain the most 285 important federal protections for wetlands today. Wetland protections came primarily 286 under Section 404 of the CWA, in which the U.S. Army Corps of Engineers was made 287 responsible for regulating the discharge of dredged or fill material into “waters of the 288 United States,” including wetlands, under the general oversight of the EPA. Under CWA 289 Section 404, restoration and creation practices were to be employed to compensate for 290 impacts to wetlands. Wetlands are often located wholly or partially on privately owned 291 land. This aspect of wetland regulations have made them some of the most contentious 292 elements of environmental law to date (NRC 1995), and the resulting protection of 293 wetland habitat has fallen short of the goals set forth in the Clean Water Act (NRC 2001).

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 8 294 By the mid 1980’s, wetland declines had resulted in approximately 117 million 295 acres of wetland had been lost nationwide, about half the original amount (NRC 1995). 296 In California, declines were much more severe with losses estimated to be about 297 90%.(Dahl 1990) Recognizing this problem, and given the refined understanding of the 298 importance of wetland functions, the EPA called for a National Wetlands Policy Forum 299 in 1987 and asked the participants to make national policy suggestions for the future of 300 wetland protection. The central recommendation of the panel was to create a policy of 301 “no net loss” of remaining wetlands which would be emphasized in the Corps’ Section 302 404 permitting program. In 1990, the first Bush administration adopted this policy of no 303 net loss. Later that year the Corps and EPA produced a guidance document that 304 instructed regulatory personnel how to implement compensatory mitigation requirements 305 (see below) within their 404 permit program such that “no net loss” would be achieved 306 (NRC 2001). The implementation of this policy goal, along with a stronger emphasis on 307 compensatory mitigation practices to offset wetland losses, took effect in 1991. 308 2.4. Clean Water Act Section 404
309 Section 404 of the Clean Water Act prohibits the discharge of dredged or fill 310 material such as sand or soil into waters of the United States, unless a permit is issued 311 under the regulatory authority of the U.S. Army Corps of Engineers. The great majority 312 of permit applications are ultimately approved (NRC 2001). While some projects must 313 be evaluated and permitted on an individual basis, others may fall into more general 314 categories, such as bank stabilization or the maintenance of bridge over-crossings. 315 Numerous regional or nationwide permit categories are available for such projects, which 316 can help to streamline the approval process. In all cases, the Corps personnel must 317 follow a standard three-step sequence in their decision making process. They must first 318 determine if different strategies could be employed in which all or some of the proposed 319 impacts might be avoided or minimized. Given the national goal of “no net loss,” any 320 remaining impacts must be compensated for by creating, restoring, or preserving 321 wetlands or waters in another location (NRC 2001). This is termed compensatory 322 mitigation.
323 With respect to compensatory mitigation, agency guidance documents and 324 regulatory personnel have traditionally preferred nearby, in-kind mitigation to offset 325 losses. However, recognizing the shortcomings of some permittee-responsible 326 mitigation, some regulators have begun favoring the use of alternative third-party 327 strategies such as mitigation banks and in-lieu fee programs where mitigation is likely to 328 be off-site (NRC 2001).
329 Mitigation banks are sites where a large restoration, creation, or enhancement 330 project, is undertaken to provide compensatory mitigation in advance of projects that will create wetland losses.3 331 Credits from these projects can be used to offset losses (debits) 332 permitted under Section 404 on an acreage basis. Mitigation banks may be established 333 by entities that anticipate having large numbers of future permit applications, or by third 334 parties that wish to sell their credits for a profit. Although there is a formal process for

3 Of course, there are many variations on this general description, a common variant being allowing credits from a mitigation bank before it is completed and demonstrated to be successful.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 9 335 establishing mitigation banks, some of the mitigation banks used by permittees with a 336 large number of permits are only informal banks, having never been established through 337 the formal process but nonetheless being used by the permittee and regulatory agencies as 338 a bank. In-lieu fees are payments made to natural resource management entities for implementation of either specific or general wetland development projects.4 339 Mitigation 340 banks have the benefit of avoiding temporal losses of wetland habitat that occur between 341 the time the actual loss occurs at the impact site and the point where complete function is 342 restored at the mitigation site. In-lieu fee programs may or may not avoid temporal 343 losses. Both of these third-party approaches have the potential to restore large areas of 344 relatively high quality contiguous wetland habitat that may be better situated in a 345 landscape context than individual mitigation projects, being placed in proximity to 346 existing functional wetland habitat. However, banks and in-lieu fees often result in off347 site mitigation, with potential negative effects due to spatial shifts in habitat distributions 348 and loss of wetlands within some regions. In addition, the values wetlands provide often 349 are dependent upon their location in the landscape, such as their position relative to one 350 another, to adjacent waters, and to the human population that would benefit from the 351 services provided (Brow and Lant 1999).
352 Most often, the amount of mitigation required is not a simple one-acre mitigated 353 for one-acre lost ratio (NRC 2001). The additional acreage is intended to account for 354 temporal losses and incomplete replacement of function. Therefore, mitigation ratios of 355 2:1, 3:1, or greater are sometimes required.
356 Before a 404 permit can be issued, the applicant must also obtain: (1) A state 357 water quality certification required under CWA Section 401, which, in California, is 358 administered by the State Water Resources Control Board and its nine Regional Water Boards.5 359 This document certifies that the project will not adversely impact water quality, 360 or if it does, those impacts will be mitigated. (2) A California Department of Fish and 361 Game (DFG) streambed alteration agreement, which ensures that a project does not 362 adversely impact the local fish and wildlife, or if it does, those impacts are mitigated. 363 These mitigation requirements are distinct from those required by the Corps. Once all 364 approvals are either met or waived, the Corps can issue the 404 permit. 365 2.5. Assessing mitigation success
366 After a permit is issued, monitoring of the mitigation site is almost always 367 required; however, there is generally little regulatory follow up evaluating what happened 368 at either the impact site or the mitigation site. This is, in part, because there are so few 369 regulatory staff and so many permit applications (NRC 2001). Mitigation reports 370 typically are required to be submitted by the permittee throughout the five-year 371 certification period, but it is not clear how often this is done or how often regulatory staff 372 review them. In addition, record keeping has been identified as an impediment to

4 In the past, in-lieu fees were not necessarily restricted to natural resource management, and as a result became a controversial form of mitigation.
5 The administration and implementation of CWA Section 401 varies from state to state; California is among those states with more developed 401 programs.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 10 373 assessing mitigation practices, with incomplete files and inadequate database tracking 374 systems being a common regulatory problem (NRC 2001, Ambrose and Lee 2004). 375 Few determinations of the regulatory success of compensatory mitigation projects 376 occurred during the first decade of their existence (NRC 2001). Determining mitigation 377 compliance can be difficult. Assessing permit compliance entails an initial permit review 378 and site visit to determine if the project was undertaken, if the actual acreage matched 379 what was proposed, and if the specified performance standards were met. In planning 380 and executing a compensatory mitigation project, the permittee’s focus usually is to 381 satisfy permit conditions. As long as the permittee can demonstrate that the performance 382 standards set forth in the permit have been met, their obligations have been fulfilled. As 383 yet, aspects of wetland function have not been adequately incorporated in performance 384 standards (NRC 2001, Ambrose and Lee 2004), in part because of the legal difficulties in 385 assigning specific targets for function (NRC 2001). Some performance standards that 386 have been developed are intended to be proxies for function, but given the challenges of 387 measuring functions directly, assessments of hydrological, biogeochemical, and 388 ecological function have remained elusive.
389 Data reported by the Army Corps of Engineers indicate that the goal of “no net 390 loss,” as measured by acreage shifts, is not only being met but is being exceeded. 391 According to the Corps, from 1993 through 2000, approximately 24,000 acres of wetland 392 losses were permitted, while 42,000 acres were created through compensatory mitigation 393 (NRC 2001). Thus an average mitigation ratio of 1.8:1 was achieved. However, these 394 statements of mitigation success and the achievement of no net loss were based solely on 395 the acreage of mitigation required in the permits, not on field evaluations of wetland 396 acreage or function present at mitigation sites. In addition, they may have not included 397 existing acreage of wetlands at mitigation sites. Furthermore, they have not addressed 398 functions provided at mitigation sites. One recent study that employed functional 399 assessment methods to evaluate the success of the Section 404 permitting program, 400 conservatively estimated that only 55% of mitigation sites met permit conditions, while 401 only 16% of the sites could be considered successful in terms of function (Sudol and 402 Ambrose 2002). Another study, Ambrose and Lee (2004), found that the majority of 403 mitigation projects met their mitigation acreage requirements and most were in 404 compliance with permit requirements overall, yet few (4%) resulted in optimally 405 functioning wetlands and, with respect to a structured qualitative assessment of the 406 beneficial services lost versus those gained through the mitigation project, 66% failed to 407 achieve “no net loss.” These data suggest that the success of the Clean Water Act and the 408 “no net loss” policy has not succeeded in preserving our nation’s remaining wetlands. It 409 is impossible, however, to determine the extent of wetland losses that would have 410 occurred in the absence of the Section 404 program. 411 3. Methods
412 3.1. Project Management
413 This statewide study was conducted by two research groups: a University of 414 California, Los Angeles (UCLA) research group consisting of Dr. Richard Ambrose 415 (principle investigator), two full-time research technicians, three shorter-term technicians, Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 11 416 and one graduate student/project coordinator (Steven Lee), and a University of San 417 Francisco (USF) research group consisting of Dr. John Callaway (principle investigator), 418 three graduate student researchers working full-time and one shorter term technician. 419 The Principal Investigators maintained oversight over the entire project, including 420 project conception and design and completing the final report. UCLA had primary 421 responsibility for contract administration and project management, project coordination 422 and management, the initial SWRCB database review, regional apportionment and 423 selection of permit files for review, Freedom of Information Act (FOIA) coordination, 424 and progress report generation. The permit review and field efforts for this project were 425 roughly equally divided between the USF and UCLA groups, with USF responsible for 426 the northern half of the state and UCLA the southern half. Considerable effort was spent 427 ensuring consistency between USF and UCLA data collection procedures. Members of 428 the UCLA group participated in the initial file review for the north-central portion of the 429 State and joined the USF group for a number of their field reconnaissance visits and site 430 evaluations, and a member of the USF group participated in some site evaluations 431 conducted by UCLA. After the fieldwork was completed, UCLA was responsible for 432 data management, data analysis and presentation, and producing the initial draft of the 433 final report. UCLA carried out most of the QA/QC procedures and, after finding a range 434 of data and consistency problems, helped the USF group resolve these issues. The USF 435 group incorporated the site GPS coordinates into GIS base maps to create regional and 436 statewide maps showing the distribution of our mitigation site assessments. In addition, 437 the USF group completed an analysis of mitigation banks (see Appendix 9) and a 438 supplemental assessment of wetland condition (the Wetland Ecological Assessment, or 439 WEA) at a subset of their sites and carried out all analyses and reporting of those data 440 (see Appendix 10).
441 3.2. Permit File Selection and Review
442 For this study, our goal was to evaluate the mitigation actions associated with at 443 least 100 Section 401 permit files issued in California between 1991 and 2002. The 444 projects were to be distributed across the 12 regions and sub-regions of the State Water 445 Resources Control Board (SWRCB) in proportion to the total number of 401 permit 446 actions issued within each region (Figure 1). For instance, if a particular region had 447 issued 10% of the total statewide 401 permits in this timeframe, then 10% of our 448 evaluations occurred in that region. The regional targets were exceeded for all regions 449 except for Redding (5R) and Lake Tahoe (6T), for which we met the targets exactly. For 450 those regions with small proportional targets (Region 7 and sub-Regions 5F, 6T, and 6V), 451 we attempted to add more files to increase the sample sizes, but this only was achieved 452 for sub-Region 5F.
453 Files were selected using the SWRCB’s permit tracking database. We used the 454 version dated October, 2004, obtained directly from the State Board. To ensure 455 statistically reliable information, projects were chosen randomly from this database. 456 Initially, we expected to select all projects based on the database fields that indicated 457 compensatory mitigation was required. However, we discovered that the database did not 458 reliably indicate a compensatory mitigation requirement for permits issued before 1998; 459 for these files, a physical inspection of a large number of files at the State Board office Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 12 460 was necessary in order to find the appropriate number of projects requiring mitigation. 461 To account for the difference in information in the database as well as ensure an equal 462 distribution between older and more recent permits, half of the projects were from 1991- 463 1998 and half were from 1998-2002. The permit projects included in our study included 464 401 permits with explicit mitigation conditions as well as permits without conditions but 465 with implicit or explicit requirements that the mitigation conditions of other regulatory 466 agencies be followed. The permit projects were reviewed through multiple visits to the 467 SWRCB, each of the three Army Corps of Engineers district offices (Los Angeles, San 468 Francisco, and Sacramento), and various Regional Boards. There were many 469 complications that had to be resolved in selecting files for this study; a full accounting of 470 the selection process is provided in Appendix 1.
471 3.3. Office Review and Assessment
472 After the initial permit review at the Corps and/or Regional Board offices, the 473 relevant file materials were photocopied and retained for further review and for reference 474 during field visits. Prior to the field visit, each file was subjected to an extensive office 475 review to verify that the project occurred, to gain a general understanding of both the 476 project impact and the expected mitigation activities, and to extract all relevant permit 477 conditions for the ensuing compliance evaluation. To this end, all available 478 documentation was consulted, including any pre-project planning information, the 401 479 order, 404 permit, streambed alteration agreement, mitigation plan, monitoring reports, 480 and any other information reflecting changes in the planned actions since the permits 481 were issued. Often, correspondence with regulatory personnel, the permittee, the 482 permittee’s consultant, or the in-lieu fee recipient was necessary to resolve site access 483 issues, to determine if the impact or mitigation projects were undertaken, or to verify fee 484 payments.
485 Office evaluations were a significant element of the condition assessment 486 methodology (discussed below); the information gained from this evaluation improved 487 the understanding of the landscape context of the site, including the surrounding land 488 uses and the stressors associated with those land uses and helped to identify the 489 boundaries of the assessment area. One important component of the office review was 490 the acquisition of web based aerial photographs (http://terraserver.microsoft.com/), which 491 provided landscape context and aided in the location of project sites. 492 As we performed the office reviews, some files were deemed un-assessable and 493 were excluded from further study. Reasons for such exclusion varied but included 494 confirmation that the impact and/or mitigation project never happened and denial of 495 access to the project site.
496 3.4. Site Visits
497 Given the broad geographic scope of this statewide study, combined with the time 498 limitation imposed by the contract and the protracted permit review process, logistics and 499 efficiency were critical aspects of the field phase of the project. Early site visits and 500 methodological refinements occurred close to the home bases of the two research groups; 501 more distant sites were assessed later. Once the assessment procedures were established Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 13 502 and the initial list of permit files was obtained, the project locations were marked on state 503 and regional maps and organized into local or multi-day research trips based on the 504 proximity and clustering of the sites. Next, seasonal and other factors were considered, 505 and the trip clusters were prioritized and scheduled. In advance of a trip, the relevant 506 files were reviewed, the permit conditions extracted, data forms were generated, access 507 issues were anticipated and pursued, and other logistical arrangements were made. 508 Upon arrival at the general project area or the mitigation site location, we looked 509 for evidence of mitigation activities such as plantings, irrigation systems or disturbed 510 earth to confirm the presence of mitigation activities. The permit paperwork and aerial 511 photographs were helpful in establishing the presence of the mitigation site and 512 determining its boundaries. For each of the fully assessed files, a considerable amount of 513 time was spent onsite deciphering the language of the permit file paperwork to 514 understand the nature of the impacts, to identify all discrete mitigation projects involved, 515 to identify and map the boundaries of those discrete projects. A site was considered 516 onsite if it was on the same property as the impact, and this determination was relative to 517 the scale of the greater project area. For a large development project, two mitigation 518 actions located a kilometer or more apart could both be considered onsite, while the 519 mitigation site for a small utility crossing might be considered offsite even if separated by 520 just 100m.
521 Occasionally, we found that the impact project was currently under construction 522 and the mitigation activities had not yet been initiated, or there was no evidence that the 523 impact or mitigation project occurred. It was also common, especially with the newer 524 permits, that the impact project had occurred, but the construction of the mitigation site 525 was still under way. There were a few instances where the impact project had been 526 completed, but we found no evidence that the required mitigation had occurred. In each 527 of these cases, the file was excluded from further consideration in this study. A list of all 528 such files with the reasons for exclusion has been provided separately to the SWRCB. In 529 addition to these excluded permit files, there were 14 files for which compliance 530 evaluations could be made, but where wetland condition evaluations were not performed 531 either because of ambiguities inherent in the mitigation banking and/or in-lieu fee process 532 or for logistical reasons. These files, provided in Appendix 2, are included in our 533 compliance results but not the results of our condition evaluations. We refer to these 14 534 files as “compliance only” files, while files that were evaluated for permit compliance, 535 acreage, and wetland condition (CRAM) are referred to as “fully assessed” files. 536 3.5. Acreage Determinations using GPS
537 The acreages of mitigation sites were determined by mapping the perimeter of 538 each site. After initial site reconnaissance, we walked the site perimeter using a mapping 539 grade GPS to establish the outline of the site. GPS data were collected with a Trimble 540 Pro XR GPS receiver and a TSCE handheld interface. Many permits (70 of the 129 541 permit files we assessed) involved multiple mitigation sites. In these cases, we surveyed 542 and evaluated the discrete mitigation sites separately. 543 Although simple in concept, the actual acreage determinations were complex. 544 The reasons for this are varied. In many permits, there were ambiguities in the Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 14 545 identification of mitigation habitat types and no site positioning information. The 546 boundary between mitigation wetlands and adjacent existing wetlands was often not 547 easily discerned. Many mitigation project sites blended together several different habitat 548 types (e.g., wetlands, alluvial scrub, riparian areas, etc.). In addition, multiple mitigation 549 strategies were often used (e.g., creation, restoration, enhancement, and preservation) and 550 were difficult to distinguish. Even where site boundaries could be determined, they were 551 usually not clearly delineated as they transitioned into the surrounding landscape. GPS 552 coordinates of mitigation sites were almost never available in the permit files, and stakes, 553 flags or other survey markers were seldom present. We attempted to be as accurate as 554 possible in our surveys of site perimeters, but we erred toward overestimation rather than 555 underestimation of site area. That is, we walked the widest boundary possible as 556 determined by disturbed earth, irrigation systems or obvious vegetation plantings to 557 provide a “best case” acreage estimate.
558 We were sometimes unable to determine even the approximate boundaries of a 559 mitigation site. (See Section 6.2.1.7 for a recommendation to address this problem.) 560 This was common for older sites and for re-vegetation projects in active channels or 561 floodplains. When the evidence of mitigation activities was scant or absent, and when 562 these activities blended into the surrounding landscape, it was not possible to delineate 563 the perimeter of the project site. We attempted to confirm the general location of the 564 mitigation site from evidence of mitigation activities at the expected site location and/or 565 through information gleaned from the permit files. If it was possible to confirm a general 566 location for the mitigation site, a single GPS point was taken to identify the approximate 567 location of the site and our corresponding evaluations. 568 After field mapping, GPS data were downloaded to office computers and 569 managed using Trimble’s Pathfinder Office Version 3.0 software. GPS data were 570 differentially corrected (yielding sub-meter accuracy) using data collected from the base 571 station provider nearest to the mitigation site, as determined by an automated internet 572 search. The acreage values were obtained from the corrected files within Pathfinder 573 Office. Occasionally small perimeter adjustments were made to these files or polygon 574 fragments were added or subtracted using the measuring tool function in that program. 575 Acreage values were recorded and compared to the permit requirements to determine 576 acreage compliance. There may have been a number of discrete mitigation sites 577 associated with a file, and there were mapped separately. However, permit requirements 578 generally included only a single acreage requirement per file (or per habitat type), so we 579 combined the acreages of separate mitigation sites to determine compliance. 580 In situations where the site perimeters were clear and unambiguous, we always 581 reported our survey values as the obtained acreage. However, where the site perimeters 582 were less clear, and especially where single GPS points were taken, a judgment had to be 583 made to determine whether there was compliance with acreage requirements. In such 584 cases, we considered all available information, including visible features of the site and 585 information from the permit file such as acreage values reported in mitigation plans and 586 monitoring reports, to judge whether the acreage requirement was met. Ultimately, a 587 decision regarding acreage compliance was made for all files with acreage requirements. 588 It should be noted that the target acreage outlined in the mitigation plan is intended to Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 15 589 compensate for all agency requirements (including the Army Corps, and CA Dept. of 590 Fish and Game), and often exceeds that required by the 401 permit alone. 591 For every file, a single representative GPS coordinate was selected and recorded 592 in Pathfinder as the best description of the location of the mitigation sites (Appendix 4). 593 Also included in this appendix is a compact disc containing all GPS-related computer 594 files associated with this project.
595 3.6. Compliance Evaluations
596 In theory, permit compliance would be determined by considering each of the 597 specific and general conditions listed in an agency’s permit, assessing whether each 598 condition had been met or not met, and then assigning an overall compliance score based 599 on the percentage of conditions met. In practice, a third party assessment of permit 600 compliance, especially one that attempts to follow the standard conventions of scientific 601 rigor, is complicated by the idiosyncratic nature of regulatory permits. 602 Most of the conditions listed in 401 orders were administrative in nature or 603 involved impact avoidance measures to be implemented during the construction phase of 604 the impact and mitigation projects. This was especially true of the standard conditions 605 that are often attached to the 401 order, but many of the special conditions fell into this 606 category as well. Most of these conditions were impossible to assess in an after-the-fact 607 review, such as the present study, because one would need to be present during the 608 construction phase or have detailed post-construction compliance reports documenting 609 how each condition had been satisfied. While compliance monitoring reports were often 610 required, they were infrequently available.
611 Since the focus of this study was on the success of compensatory mitigation 612 projects, the conditions we considered in our compliance evaluation were limited to those 613 dictating the mitigation actions to be taken, any performance standards meant to ensure 614 the success of the mitigation project, and any submission requirements for mitigation615 related documents. The 401 permits we reviewed included relatively few conditions in 616 these categories. The most commonly encountered were descriptions of the proposed 617 mitigation actions and acreages, submission requirements, references to the mitigation 618 plan or specific phraseology that the plan be followed, and conditions invoking the permit 619 requirements of other regulatory agencies (e.g., the 404 permit issued by the U.S. Army 620 Corps of Engineers, the Streambed Alteration Agreement issued by the California 621 Department of Fish and Game (DFG), and occasionally, other agency requirements such 622 as those specified in the U.S. Fish and Wildlife Service (FWS) Biological Opinion). 623 Our determinations of 401compliance included all mitigation conditions 624 specifically outlined in the 401 permit order, plus any additional compliance goals or 625 conditions found in the mitigation plan and other agency permits when the 401 permit 626 included explicit statements requiring that those documents be followed. With respect to 627 the mitigation plan, if the 401 permit contained a submission requirement or included 628 language indicating that the plan had already been obtained and reviewed by the Regional 629 Board prior to permit issuance, we considered it to be implied and enforceable that the 630 plan be followed as a condition of the permit. We did not consider other agency Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 16 631 requirements as implied and enforceable conditions of the 401 permit unless there was 632 specific language mandating that those permits be followed. At the same time, we 633 recognized that during the mitigation planning process, the permittee must consider all 634 agency requirements (not just the 401), and that the mitigation plan represents a blending 635 together of these conditions into a single project. Therefore, we completed a second 636 compliance evaluation that considered how well the assessable goals and performance 637 standards of the mitigation plan were met. In addition, in the field we assessed 638 compliance with all agency conditions contained in the file, even for permits not 639 explicitly invoked by the 401 order. Due to time limitations and the fact that these latter 640 analyses were beyond the contractual scope of this project, they are not included in this 641 report.
642 As part of our general office assessment, each permit file was subjected to a 643 thorough review during which all appropriate mitigation requirements were extracted 644 from the available paperwork. Beginning with the 401 order, each regulatory permit was 645 carefully read to allow for a full understanding of the project requirements and to 646 distinguish mitigation-related conditions from the other conditions of the permit. All 647 relevant conditions were entered into a Microsoft Access database and tracked according 648 to the source permit. Many of these conditions were entered verbatim, but it was often 649 necessary to paraphrase or dissect the permit text because the permit requirements were 650 written in an ambiguous fashion or not amenable to a direct assessment of compliance. 651 (See Sections 6.3.2 and 6.3.3 for recommendations the deal with this issue.) For 652 example, a single line-item condition including two or more discrete requirements that 653 could not easily be assessed or scored together would be separated into assessable 654 conditions. In other cases, long passages were condensed down to the essential 655 compliance elements. All relevant mitigation-related conditions were entered, even 656 conditions that would likely be un-assessable.
657 In addition to the regulatory permits, the mitigation plan, if present, was carefully 658 read to extract the essential compliance elements. Though it may implicitly or explicitly 659 be mandated that the mitigation plan be followed as a condition of the permit, there is no 660 simple prescription for assessing mitigation plan compliance. Mitigation plans are not 661 written as lists of assessable conditions; both permit-mandated and permittee-initiated 662 objectives, actions, and success criteria are blended together and presented diffusely 663 throughout the pages of the mitigation plan. (See Section 6.3.3 for a recommendation 664 addressing this issue.) This complication required that we establish criteria for extracting 665 discrete compliance elements from the mitigation plans. A full accounting of these 666 conventions and lists of typical conditions extracted are presented in Appendix 6. All 667 relevant objectives, actions, and success criteria taken from the mitigation plans were 668 entered into our Access database and recorded as coming from the mitigation plan. 669 Prior to the field visit, lists of conditions by source were printed as data sheets and 670 permit conditions were assessed for compliance though a combination of field and office 671 assessments. There are at least two equally justifiable methods of assessing permit 672 compliance. The first is to score each condition as either met or not met, and to calculate 673 an overall compliance score as the percentage of conditions met. This approach is 674 consistent with the regulatory perspective and has been used in other studies of mitigation Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 17 675 compliance (Sudol 1996). The approach employed in this study departed from this met676 not met perspective because we recognized that permittees may attempt to meet a 677 particular condition even if they fall short of the success criterion needed to meet that 678 condition to 100% satisfaction. In other words, a not met score does not allow the 679 distinction between a permittee who obtained 95% of the required mitigation acreage and 680 a permittee who made no mitigation attempts at all. Since our goal was to understand the 681 critical factors influencing compliance success, we were interested in incorporating this 682 distinction. Thus, we scored each condition as a percentage on a scale from 0% (no 683 attempt to comply) to 100% (condition fully met).
684 In most cases, compliance was assessed within five scoring categories: 100%, 685 75%, 50%, 25%, and 0%. A 100% score was assigned if the condition had been clearly 686 met or exceeded. The 75% scoring category was applied if the condition fell short of 687 being fully met, but had been mostly met. If the condition was about half, or partially 688 met, it received a 50% score. The 25% category was used if some level of compliance 689 effort had been made, but the outcome fell far short of expectations, and the condition 690 was mostly not met. Finally, a 0% score was assigned if there was clear evidence that the 691 permittee made no effort to comply with the condition. These broad categories were used 692 to distinguish different degrees of compliance with a particular condition but avoid 693 difficulties that could arise from trying to distinguish between fine-scale categories (e.g., 694 85% versus 90% compliance).
695 For some conditions, the score could readily be calculated as a percentage relative 696 to the desired outcome. For instance, if the target mitigation acreage was 0.75 acres but 697 our surveys revealed that only 0.50 acres had been obtained, then the compliance score 698 would be 67% (0.50/0.75). Acreage compliance was almost always calculated in this 699 way. This approach was used for other variables that were continuous in nature (such as 700 survivorship or percent cover), but only when our assessments could be made with a high 701 degree of certainty. Otherwise, the condition was assessed using the above scoring 702 categories.
703 In scoring compliance, we were careful to distinguish between compliance with 704 the explicit verbiage of the condition and the ecological outcome that the condition was 705 directed towards. For example, if a condition required that “non-natives be removed 706 prior to planting,” then as long as we found evidence that this task was done, the 707 condition would be assigned a high score, even if the site was currently dominated by 708 non-natives. However, if the condition required that “non-natives be eradicated from the 709 site,” then a site dominated by non-natives would yield a low score. 710 A large number of mitigation conditions could not be assessed because there was 711 not enough evidence to confirm or deny that a required action had been taken. In such 712 cases, we had no choice but to score the condition as “not determinable.” These 713 conditions were not included in our analyses of overall compliance score. Many of these 714 conditions could not be assessed because one would have had to be present during project 715 implementation or have access to detailed information verifying compliance. For 716 example, it is commonly required that any non-native species be removed prior to 717 restoration, stripped or exposed areas be hydroseeded with native grasses, and mulch 718 applied around plantings. Sites rarely contain evidence of such activities a few years Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 18 719 after construction, so without photo-documentation or written verification, none of these 720 conditions can be assessed in an after-the-fact review such as the present study. A full 721 accounting of the compliance issues we experienced, along with our resolutions and 722 scoring conventions, is provided in Appendix 6.
723 3.7. Evaluations of Wetland Condition
724 3.7.1. California Rapid Assessment Method (CRAM)
725 Permit compliance alone may not guarantee that mitigation actions result in 726 ecologically functional wetlands or riparian habitats. To evaluate existing wetland 727 condition, we performed the California Rapid Assessment Method (CRAM; Collins et al. 728 2005) at all assessable compensatory mitigation sites associated with our permit files. 729 CRAM is a semi-quantitative method for the rapid assessment of wetland and riparian 730 condition. The following excerpts from the CRAM 3.0 manual (Collins et al. 2005), with 731 some paraphrasing, provides the basic conceptual framework of this methodology:
732 The objectives of CRAM development are to provide a rapid, 733 scientifically defensible, and repeatable [assessment of wetland 734 condition] that can be used routinely in wetland monitoring and 735 assessment programs, [notably in the] evaluation of wetland restoration 736 project performance under the Coastal Zone Management Act, Section 737 1600-1607 of the California Fish and Game Code, Sections 401 and 404 738 of the Clean Water Act, and local government wetland regulations, [and 739 in the] assessment of restoration or mitigation progress relative to 740 ambient conditions, reference conditions, and expected ecological 741 trajectories.
742 The CRAM methodology consists of scoring wetlands of any of 743 several different classes based on four attributes: hydrology, biotic 744 structure, physical structure, and buffer/landscape context. Within each 745 of these attributes are a number of metrics that address more specific 746 aspects of wetland condition. Each of the metrics is assigned a score 747 based on either narrative or schematic descriptions of condition, or 748 thresholds across continuous, numerical values. Scores assigned are 749 aggregated up to the level of attributes as well as into a single, overall 750 score. In addition to assessing wetland condition, CRAM provides the 751 practitioner with guidelines for the determining the types of stressors 752 that may be affecting a given wetland, and may therefore help explain 753 low condition scores.
754 During our previous study of mitigation success (Ambrose and Lee 2004), we 755 used an earlier version of CRAM (CRAM Version 2.0; Collins et al. 2004) to evaluate 756 wetland condition at mitigation sites in SWRCB Region 4 (Los Angeles/Ventura). At the 757 time of that study, CRAM was in an intermediate stage of development and some aspects 758 of the method had not been resolved. We made a number of modifications to that version 759 of CRAM to improve its utility for evaluating mitigation wetland sites, many of which 760 were subsequently incorporated into CRAM. By the beginning of the present study, a Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 19 761 new draft version of CRAM was available and ready for field calibration. Early in the 762 project, the UCLA and USF research groups participated a calibration meeting that 763 included several field tests of the revised method. Issues identified during that calibration 764 meeting were incorporated into the new version (Version 3.0, Collins et al. 2005), which 765 was distributed to the CRAM calibration teams for further field testing. As we entered 766 the fieldwork phase of this study, we began using CRAM 3.0 in our site evaluations. 767 During the course of this study, a few additional modifications were proposed by 768 members of the CRAM development team and an unofficial revision of CRAM (termed 769 Version 3.5) was implemented. We adopted the proposed modifications and incorporated 770 them into our remaining site evaluations; we also rescored all previous evaluations to 771 ensure consistency among all mitigation site assessments. 772 Despite changes to CRAM incorporated after our study for Regional Board 4, the 773 delineation of the assessment area still required modification or adaptation. CRAM was 774 designed to evaluate complete wetland systems, including larger estuarine or depressional 775 wetland complexes or for riverine sites, the entire riparian zone consisting of the stream 776 channel and the vegetation along both banks. However, mitigation sites are rarely 777 complete wetland systems. For example, it was very common for riparian mitigation 778 projects to occur outside the active channel and to involve plantings along only a single 779 bank, or within an area above the bank that previously was upland habitat. While CRAM 780 has rules for establishing the limits of the assessment area (including the appropriate 781 reach length and the lateral limits of the riparian zone), our assessment areas had to 782 conform to the boundaries of the mitigation sites. Thus, if the mitigation efforts occurred 783 on a single bank, most of our ecological evaluations (such as plant cover) would be 784 limited to that bank area alone. However, several aspects of the riverine CRAM 785 evaluation were dependent upon the characteristics of the main stream channel. 786 Specifically, the assessment criteria for all three hydrology metrics (water source, 787 hydroperiod, and upland connection), two of the abiotic structure metrics (abiotic patch 788 richness and topographic complexity), and two of the biotic structure metrics (biotic 789 patch richness, and interspersion and zonation) were focused on channel and floodplain 790 characteristics. If CRAM was applied strictly, assessment areas that did not include the 791 stream channel would always score poorly for those metrics. Consistent with the 792 approach used by Ambrose and Lee (2004), the convention we adopted was to consider 793 the channel as part of the assessment area for these metrics, provided that the mitigation 794 site was in direct proximity to, and hydrologically connected with, the stream channel. 795 As a result, mitigation sites or portions of sites that occurred higher on the banks, and 796 were clearly not wetlands, received relatively high scores for these metrics. While this 797 may have inflated the CRAM scores for some mitigation sites, we adopted this 798 convention to allow mitigation sites adjacent to a stream channel to be assessed as part of 799 the entire riverine system, even if the mitigation action did not alter the channel. 800 Mitigation sites not directly associated with a channel, such as “riparian” plantings in 801 upland areas above and beyond the banks, were not scored based on a (distant) channel; 802 such sites were given the lowest scores for channel-dependent metrics. Aside from this 803 convention for including channel characteristics in the evaluation of riparian sites, all 804 other aspects of CRAM related solely to the actual site of the mitigation actions.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 20 805 For every file, we determined whether the permit requirements resulted in one or 806 more mitigation projects that could be assessed appropriately using CRAM through our 807 permit review, site reconnaissance, and compliance investigations. Restoration, creation, 808 and enhancement projects that were post-construction and for which the initial vegetation 809 efforts had been made were evaluated using CRAM. As a convention, we did not 810 perform CRAM at any wetland preservation or conservation sites because there was no 811 mitigation action to assess. Such files were evaluated for compliance only (e.g., payment 812 of fees).
813 When a permit file contained a single discrete mitigation site, a single CRAM 814 evaluation was made. Many files, however, included two or more distinct sites involving 815 fundamentally different habitats or mitigation strategies. For example, the mitigation 816 requirements of a given file might include a depressional wetland creation project and a 817 riparian restoration project, or the file might include two separate “riparian” sites, one of 818 which involved the reconfiguration and planting of a stream bank while the other 819 involved “riparian” plantings in a separate location that was beyond the stream banks in 820 an upland area. As another example, a file might involve mitigation bank payments for 821 both tidal wetland and seasonal wetland credits. Separate CRAM evaluations were done 822 for each of these distinct mitigation sites.
823 When an individual mitigation site was small and homogeneous, we assessed the 824 entire site with a single CRAM evaluation. If the site was larger and more complex but a 825 central location appeared to be representative of the entire site, we performed a single 826 CRAM evaluation in the central location. However, there were many mitigation sites 827 that were so large and/or complex that we needed to perform two or more CRAM 828 evaluations in different locations in order to characterize the entire site. Decisions about 829 how to subsample were dictated by the physical and biological features of the sites. For 830 example, if a site consisted of a series of excavated wetland depressions occurring 831 diffusely throughout the site or in groupings across the general mitigation project area, 832 we would assign numbers to each of the depressions and randomly select two or more 833 individual sites to evaluate. Alternatively, we would break the site into like groupings 834 and randomly subsample one depression per grouping. As another example, for a long 835 and complex stream/riparian system that was too extensive to integrate into a single 836 CRAM evaluation, we might perform three separate evaluations, one at each end and one 837 in the middle of the reach. Often, up to five or more evaluations were performed for a 838 single mitigation site. In all cases where multiple CRAM assessments were made for a 839 single mitigation site, the CRAM scores were averaged to arrive at a single CRAM per 840 site.
841 One change that occurred between the earlier version of CRAM used in Ambrose 842 and Lee (2004) and CRAM 3.0 was an increased emphasis on assessing the vegetation 843 community at the site. The greater level of detail required for the identification of 844 individual plant species and determining the relative percent cover for each of those 845 species added considerable time to the field evaluations, demanded increased expertise 846 regarding the statewide flora, and created numerous complications in the assessment of 847 the percent invasive plant species, and native plant species richness metrics. The 848 consistent identification of plants to a given taxonomic level was problematic for such a Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 21 849 large study. In general, we tried to identify all plants to the species level, but for some 850 individuals, we were only able to reach the genus or family level. During our field visits 851 across the state, the diversity of flora encountered often required that we photograph or 852 collect plant samples so they could be later identified and/or categorized. Cover 853 estimates for those unidentified individuals were made in the field, however. Grasses 854 were particularly challenging in this regard. As with other plants, we attempted to 855 identify grasses to the species or genus level, but given the great morphological and 856 ontogenetic variability of grasses, this task often exceeded our collective expertise. As 857 such, we commonly combined individuals of questionable division into a generic “grass 858 spp.” category, or where individual species could be discerned, they were arbitrarily 859 named (e.g. grass sp. 1; grass sp. 2, etc.). These grasses were categorized as native or 860 non-native, given the best information available to us regarding the local flora. 861 We also had to adapt CRAM guidelines for the timing and seasonality of 862 assessments. CRAM was designed to be performed during the growing season, which for 863 different wetland types in different locations might occur at different times of the year. 864 However, the timing of this project required that our field evaluations be made during the 865 summer and early fall of 2005, when many annual plants had already senesced for the 866 season. To reduce the effect of this off-season sampling, we departed from the written 867 CRAM methodology and included senesced annual plants in our cover estimates. Such 868 individuals were identified to species where possible, any unidentified individuals were 869 combined into larger unidentified categories according to our best judgment of 870 native/non-native status, and cover estimates were made. Although we tried to identify 871 all species that would have been included if the site had been assessed during the growing 872 season, some herbaceous plants undoubtedly had decomposed or were unrecognizable at 873 the time of our site evaluations.
874 As indicated earlier, Ambrose and Lee (2004) modified the previous version of 875 CRAM by superimposing a numerical scale over the CRAM letter grades and developing 876 algorithms for combining metric scores into scores for each of the four attributes plus a 877 Total-CRAM score for the entire file. For CRAM 3.0, the CRAM development team 878 opted against the 1-12 scoring scale used by Ambrose and Lee (2004) and adopted a 879 modified system of letter grading instead. This system allowed for the application of “+” 880 and “–“ designations to add refinement to the existing letter grades. For most metrics, 881 which are scored on an A-D scale, this system is analogous to the 1-12 scale. However, a 882 few of the CRAM metrics are limited to an A-C scale and one has been expanded to an 883 A-E scale. The CRAM developers intend that these letter grades be combined into a 884 single CRAM score, but a convention for doing so has not yet been developed. For our 885 site evaluations, we followed the new protocol and scored the CRAM metrics as letter 886 grades, adding + or – designations as appropriate. Once all CRAM data were finalized, 887 entered and checked for quality control, we converted these letter grades to numerical 888 scores for analysis. The majority of the metrics, which were on a D- through A+ range, 889 were converted using a corresponding 1-12 scale. Metrics with a C- through A+ scale 890 were converted using a 1-9 scale, and E- through A+ metrics were converted using a 1-15 891 scale. Details regarding our conversion conventions are provided in Appendix 7. To 892 normalize these scores so they could be combined, the scores were converted to Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 22 893 percentages (e.g. 9/12 = 75%) so that all metric scores would be on the consistent 0- 894 100% scale.
895 CRAM scores were combined in three stages. First, a single score was 896 determined for each metric. For mitigation sites with a single CRAM, no further 897 adjustments were needed. For CRAM evaluations that were subsamples for a large or 898 complex mitigation site, a mean metric score was calculated by averaging each of the 899 separate metric scores. For example, if three depressional wetlands were randomly 900 selected and assessed within a larger complex of depressions, then these would be 901 averaged together at the metric level in order to arrive at a single set of CRAM scores for 902 that mitigation site.
903 Next, the individual metric scores were combined to arrive at a single CRAM 904 score for the mitigation site. To do this, the metrics were first combined by attribute (e.g. 905 buffer/landscape context and hydrology) and then into a single CRAM score fore each 906 mitigation site. For the hydrology and physical structure attributes, the metric scores 907 were treated as equal and independent, so they were simply averaged. The 908 buffer/landscape context and biotic structure metrics were more complicated and were 909 treated differently. For biotic structure, the two plant community metrics (percent 910 invasive plant species and native plant species richness) were clearly related to one 911 another (high non-natives usually meant low natives). Therefore, before averaging with 912 the rest of the biotic structure metrics, a geometric mean was calculated for these two 913 scores. Within the landscape context category, the percent of the assessment area with 914 buffer and the average width of buffer metrics jointly determined the general buffer 915 extent, and these in combination with buffer condition, reflected the overall buffer 916 quality. To clarify this point, it is possible to have a very high quality buffer that is 917 adjacent to just a small portion of a site. Conversely, most of a site may have extensive 918 buffer areas that are of very low quality. To account for the complex relationship among 919 these three metrics, we first took the geometric mean of the percent of assessment area 920 with buffer and the average width of buffer metrics to determine general buffer extent, 921 then took the geometric mean of this result and buffer condition. Once we determined 922 this overall buffer score, it was averaged with the remaining landscape context metric, 923 connectivity, to determine the landscape context category score. The four attribute scores 924 were averaged to obtain an overall Total-CRAM score. 925 Finally, a single CRAM score was calculated for each permit file. For files with a 926 single mitigation site, the final CRAM score for the file was the same as the score for the 927 site. For files with multiple mitigation sites, a final CRAM score was calculated using a 928 weighted average of the scores for the individual mitigation sites. The individual CRAM 929 scores were weighted by the area of the mitigation site. Weighting the CRAM scores by 930 acreage prevented a small mitigation site from having a disproportionate effect on the 931 score for the file. For example, if a file had a very small wetland creation site that 932 received a high CRAM score and a very large wetland restoration site that received a 933 marginal CRAM score, a simple average of these two CRAM scores would not reflect the 934 combined wetland condition because of scale differences between the component sites. 935 To account for this, we multiplied the individual CRAM scores by the proportional 936 acreage of each mitigation site.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 23 937 Determining the acreages for each mitigation site required a careful review of the 938 permit files, which we accomplished after all sites had been assessed. There was no 939 simple procedure for making the acreage determinations since the permit files are 940 complex and each poses a unique set of circumstances concerning the component site 941 acreages. In some cases these acreages were taken from our GPS data, sometimes they 942 were obtained from the permit file paperwork, and sometimes both sources of 943 information were used. As an example, suppose a file involved 1.0 acre of onsite riparian 944 enhancement and a payment for 0.25 acres of vernal pool creation credits at a 10-acre 945 mitigation bank. We might have used the GPS to delineate the boundaries of the riparian 946 site and measured an area of 0.95 acres. We considered how confident we were in our 947 GPS surveys before deciding whether to apply the expected or the measured acreage. If 948 there was a very clear perimeter to the site and we had good satellite coverage, we would 949 use the measured value; otherwise, we would use the expected value from the permit 950 paperwork. For the mitigation bank, even if we had done a series of CRAM evaluations 951 at the mitigation bank to represent the 10 acre site, and these were later combined for a 952 single score for that site, we would still use only the 0.25 acres of credit for our acreage 953 proportions because that was the fraction of the entire site that related to the permit file. 954 Had we applied the expected riparian acreage from the permit file, then the total file 955 acreage would be 1.25 acres, which would yield acreage proportions of 0.8 and 0.2 to be 956 multiplied by the respective riparian and vernal pool CRAM scores. Using a similar 957 procedure, we established the acreages associated with every mitigation site, which were 958 then used to weight the CRAM scores for each mitigation site in order to calculate a 959 single CRAM value for each permit file.
960 3.7.2. Reference Sites
961 As part of CRAM development, CRAM was to be calibrated through extensive 962 sampling of a range of wetlands within each wetland class, including high quality 963 reference sites. Without some calibration of wetlands in optimal condition, the 964 appropriate target for judging mitigation sites was not clear. Performing CRAM at 965 reference sites and viewing the resulting distribution of scores would help define the 966 appropriate target range for mitigation success. Unfortunately, the CRAM calibration 967 effort had not yet been done before our field assessments had to be completed; it was 968 scheduled to take place concurrently with this project. In our previous mitigation study, 969 Ambrose and Lee (2004) assigned numerical cutoffs for optimal, sub-optimal, marginal, 970 and poor wetland function/condition based on the quartiles of the scoring scale and on the 971 findings of Sudol (1996), who established similar numerical limits through the use of the 972 Hydrogeomorphic Method (HGM) at reference sites. To provide a sound foundation for 973 evaluating mitigation sites in this study, we performed CRAM at a series of reference 974 sites distributed throughout the state.
975 In general, we took an opportunistic approach to finding reference sites, sampling 976 reference sites that were close to mitigation sites as time allowed. Discussion with local 977 agency staff, environmental consultants, or private citizens were helpful in identifying 978 potential reference sites, but we also consulted maps or aerial photographs and conducted 979 internet searches to identify wetland sites in preserves or other open space areas of 980 limited human influence. We sought reference sites that reflected the best attainable Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 24 981 conditions for the various wetland classes in a particular region. We explicitly did not 982 search out the best possible wetland sites in the state; although this would be useful for 983 CRAM calibration, they would not necessarily be the best standard for comparing to 984 mitigation sites. For this study, we wanted to use reference sites of comparable condition 985 to natural wetlands in the area (and, presumably, similar to the conditions of the wetland 986 sites that were impacted). Thus, our sites were relatively unimpacted by human activities 987 compared to other wetlands in a region, but were not pristine. We generally avoided 988 wetlands with distinct development in the watershed, but some reference sites certainly 989 had been influenced by human activities. For example, in the southern Central Valley, 990 there is essentially no portion of the lower valley floor that has not been modified in some 991 way by human activities, yet this is where most of the permitted impacts occur and where 992 most mitigation sites are located. With respect to vernal pools, there are relatively 993 pristine sites occurring on the higher table lands of the western Sierra Nevada Mountains, 994 but these are fundamentally different from the vernal pools being impacted and mitigated 995 on the valley floor and so would not have been appropriate for comparing to vernal pools 996 on the valley floor.
997 The UCLA group sampled 22 reference sites throughout the state, including 5 998 high gradient riverine, 11 low gradient riverine, 2 lacustrine, 2 vernal pool, 1 999 depressional, and 1 seep/spring wetland (Table 1). Three of these sites were in northern 1000 California, but most occurred in the southern half of the State. The USF group planned to 1001 sample a similar number of reference sites in the northern half of the State, but they were 1002 unable to do so because of time limitations. To provide data for reference sites in the 1003 northern half of the state, we used data from the CRAM calibration teams, who had 1004 completed their calibration field work by the end of the field season. Their calibration 1005 trials involved just two wetland classes: estuarine and riverine. The CRAM calibration 1006 evaluations were done for a wide range of wetland conditions, from high quality sites to 1007 lower quality sites. To select appropriate reference sites from this data set, we used the 1008 qualitative assessments of overall wetland condition made by the calibration teams to 1009 select sites that were relatively unimpacted by human activities. The CRAM calibration 1010 teams provided us with data for 7 estuarine sites and 18 riverine sites (Table 1), resulting 1011 in a total sample of 47 reference CRAM evaluations (Figure 2). All reference CRAM 1012 data were incorporated into our Access database, subjected to standard QA/QC 1013 procedures, and analyzed for comparison with our mitigation site data. 1014 3.7.3. Wetland Ecological Assessment
1015 In our previous mitigation study for SWRCB Region 4, Ambrose and Lee (2004) 1016 performed an alternative condition assessment methodology called the Wetland 1017 Ecological Assessment (WEA), developed by Breaux and Martindale (2003) to assess 1018 mitigation sites in Region 2. We performed a separate WEA assessment for every 1019 mitigation site evaluated in Region 4 to compare to the CRAM assessments. We found a 1020 strong correlation between the WEA scores and the corresponding CRAM scores, with 1021 WEA yielding slightly higher condition scores. In the present study, we decided not to 1022 repeat a WEA/CRAM comparison for the southern California sites, but the USF group 1023 performed WEA at their sites in northern California. The WEA evaluation is presented 1024 in Appendix 10.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 25 1025 3.8. Mitigation Habitats Analysis
1026 Evaluating wetland condition at compensatory mitigation sites through CRAM 1027 provides some measure of mitigation success. However, taken alone, these assessments 1028 do not indicate whether the mitigation actions resulted in “no net loss” of wetland acreage 1029 and function. In order to understand “no net loss” of wetland function, one would need to 1030 perform an assessment at the mitigation site before and after the mitigation actions were 1031 made to understand the true functional gains, and before/after evaluations of the impact 1032 site would be necessary to understand any functional losses. Indeed while some 1033 mitigation projects convert upland habitats to wetlands, most mitigation actions are 1034 undertaken at locations that already include some wetland acreage and exhibit some 1035 degree of wetland function. Clearly, before/after evaluations of wetland function are not 1036 possible in a study like this because the projects have already occurred. 1037 In our previous study of mitigation success, Ambrose and Lee (2004) investigated 1038 this “no net loss” question by performing qualitative assessments of the beneficial 1039 wetland services gained through mitigation activities, compared to what was lost through 1040 project impacts. We were unable to perform similar assessment in the present study. 1041 However, we were able to expand another aspect of the Ambrose and Lee (2004) study, 1042 the jurisdictional habitats evaluation, which allowed us to investigate “no net loss” with 1043 respect to individual types of wetland habitat.
1044 3.8.1. Jurisdictional Habitat Assessment
1045 While wetland delineations at proposed impact sites are a required step in the 1046 permit process, there is seldom a requirement that similar wetland delineations be 1047 performed at mitigation sites to ensure that adequate acreage of jurisdictional habitat is 1048 created, restored, or enhanced. (For a definition of these terms, see Section 6.3.4.) 1049 Performing full legal wetland delineations at mitigations sites was beyond the scope of 1050 this contract. However, at each mitigation site we made a qualitative assessment of the 1051 approximate proportions of jurisdictional and non-jurisdictional habitat types that would 1052 have been recorded had such wetland delineations been made. In this assessment, the 1053 first distinction we made was between that portion of the site that was within the ordinary 1054 high water mark of the water body, including adjacent wetlands (federal waters), and the 1055 remaining portion of the site. The “non-waters” area was apportioned into riparian 1056 habitats and upland habitats. The “waters of the US” area was apportioned into wetland 1057 habitats and non-wetland waters. These jurisdictional habitat categories are listed in a 1058 hierarchical fashion in (Table 2).
1059 Our wetland estimates did not conform exactly to the three parameter test 1060 (hydrology, hydric soils, and hydrophytic vegetation) because for younger sites, we 1061 factored in the potential for future development of soils and plants, provided that the 1062 hydrology was appropriate. Therefore, our data likely represent a slight to moderate 1063 overestimate of jurisdictional wetland habitat, since some of these sites might not develop 1064 hydric soils. In many cases, the established site vegetation was used to delineate wetland 1065 perimeters. However, for other sites with more sparse vegetation, site topography and 1066 hydrological indicators aided our boundary determinations.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 26 1067 In both 401 and 404 permits, non-wetland waters are often, but inconsistently, 1068 described in more specific categorizations such as “streambed,” “open water streambed,” 1069 “unvegetated streambed” and “vegetated streambed” habitats, but are sometimes simply 1070 referred to by some other description such as “riparian waters.” We followed this same 1071 approach in subdividing the non-wetland waters category, but in a hierarchical way that 1072 would enable grouping in an unambiguous way. Non-wetland waters categorized as 1073 “other” were almost exclusively those riparian waters habitats that were within the 1074 ordinary high water mark of the water body, but beyond the channel or adjacent wetlands. 1075 The clearest definition of “riparian” specifies those areas “…adjacent to perennial, 1076 intermittent, and ephemeral streams, lakes, and estuarine-marine shorelines” (NRC 1077 2002). But in regular use, and in the permit files, there is substantial ambiguity in the 1078 application of “riparian,” with reported impacts to riparian waters that may or may not 1079 include the channel itself. This ambiguity makes it difficult for us to compare our 1080 riparian waters category to those from the permit files. 1081 3.8.2. Habitat Acreage Analysis
1082 Many of the 401 permits that we analyzed were issued early in the regulatory 1083 process, before aspects of impact and mitigation planning were finalized. As we carried 1084 out the early phases of this project, we noticed that the impact acreage and mitigation 1085 requirements reflected in the 401 orders frequently did not agree with the impact, 1086 required, and obtained acreage that ultimately occurred through project implementation. 1087 This lack of agreement would be manifested in the SWRCB database as well, since those 1088 data are derived from the information in the 401 orders. To determine the extent of this 1089 difference between the 401 order and actual implementation, we conducted a formal 1090 comparison.
1091 After all the fieldwork was completed, we performed another review of all “fully 1092 assessed” and “compliance only” files to extract the most accurate information available 1093 regarding acreage losses and gains. We considered all relevant information, including all 1094 regulatory permits, the mitigation plan, monitoring reports, correspondence reflecting 1095 planning adjustments, and the dates of all such documents. The final acreages for project 1096 impacts, permit requirements, and the “obtained” acreage values determined through our 1097 study were recorded. For the impact acreage data, permanent versus temporary impacts 1098 were distinguished. In addition, acreage data were further categorized into their 1099 respective jurisdictional habitat categories (see Table 2) to analyze the individual habitat 1100 types lost versus gained. As with the more general information mentioned above, the 1101 impact and required data were obtained through our acreage analysis permit review, and 1102 the values for each habitat type were classified as permanent or temporary impacts. The 1103 “obtained” acreage data for the site were either taken from the permit files or from our 1104 GPS surveys, depending upon which values were deemed the most accurate. As 1105 mentioned earlier, when the site perimeters were clear and unambiguous, the data from 1106 our GPS surveys would be used, but when the exact perimeter of the site could not be 1107 delineated, judgments were necessary to decide whether to accept the acreage value 1108 reported in the permit files. Once the appropriate mitigation site acreage value was 1109 determined, it was sub-divided into its component habitats multiplying it by the 1110 jurisdictional habitat proportion values from our jurisdictional habitat assessment. These Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 27 1111 data were further divided into created versus enhanced acreage to distinguish acreage 1112 gains from habitat enhancements.
1113 3.9. Digital Photographs
1114 Digital photographs were taken at all of the mitigation sites. Our objective in 1115 taking these photos was to capture the essential features of the site at the time of our site 1116 visit. In many cases, only a few photos were necessary to accomplish this, while many 1117 photos were needed at other sites. It was difficult to cover some sites adequately because 1118 of the sheer size or complexity of the site. In addition to the general site photos, close up 1119 pictures of individual plants were taken for the purposes of subsequent identification, or 1120 for other reasons. The digital images were organized within computer folders labeled 1121 with the appropriate file identification number. All digital images are provided in 1122 Appendix 13 of this report, on DVD media.
1123 3.10. Data Management and Analysis
1124 All permit review, compliance, CRAM, and supplemental data were entered into a 1125 series of Microsoft Access databases developed for this project. The UCLA and USF 1126 groups maintained separate databases for their respective files, and these were later 1127 combined into a single version. The CRAM data were entered into a database obtained 1128 from the CRAM developers to ensure that the results of this study could feed back into 1129 the ongoing CRAM development process. As indicated earlier, CRAM version 3.0 was 1130 used, but with certain interim modifications implemented by the CRAM development 1131 team (unofficially termed version 3.5). Data extracted from queries of the Access 1132 databases were typically imported into Microsoft Excel for processing, graphed using 1133 SigmaPlot, and statistical analyses performed in Systat v.11. 1134 Most of the data analysis procedures have already been discussed in earlier 1135 portions of this Methods section. In general, the data in this report are organized and 1136 analyzed in two distinct ways: (1) by file, and (2) by individual mitigation site. As stated 1137 earlier, a number of permit files consisted of two or more discrete mitigation sites that 1138 could not appropriately be combined into a single evaluation. Thus, separate functional 1139 evaluations and habitat analyses were made for each of these sites to yield a total sample 1140 of 204 individual mitigation site evaluations for the 129 assessable permit files included 1141 in our study. Individual CRAM scores were combined into a single overall Total-CRAM 1142 score by factoring the proportional acreage of each respective mitigation site. The permit 1143 requirements transcended these individual mitigation actions, and thus, a single 1144 compliance evaluation was performed per file. Where necessary, the CRAM and 1145 “habitat” results are presented by mitigation site with a sample size of n=204. In other 1146 cases, such as comparisons between CRAM and compliance, they are given by file with a 1147 sample size of n=129. While the compliance date were collected exclusively by file, but 1148 in certain analyses, they are combined with the “compliance only” files (where no CRAM 1149 evaluation could be conducted but compliance could be assessed) with the larger sample 1150 size of n=143.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 28 1151 3.11. Quality Assurance/Quality Control
1152 The quality assurance and quality control (QA/QC) measures required for this 1153 project were uniquely complex. This was mainly due to the interface between our needs 1154 regarding scientific rigor and objectivity and the inherently non-scientific regulatory 1155 practices we are studying. While several previous studies have investigated wetland 1156 mitigation success, the geographic scope and multi-agency aspects of this study were 1157 without precedent, and much of our methodology had to be developed and adaptively 1158 managed as the project progressed. Timing limitations were a factor here since we had 1159 just a single field season to implement what was originally conceived as a three year 1160 study. Given the extensive decisions and interpretations that were required in this study, 1161 splitting the effort between the UCLA and USF research groups compounded the QA/QC 1162 challenges. For many ecological studies, the QA/QC procedures simply involve 1163 checking for mathematical and data entry mistakes by reviewing 10% or so of the data 1164 sheets and calculations. For this project, the QA/QC procedures spanned the entire effort, 1165 from the earliest aspects of our permit review to data analysis. Many of these procedures 1166 have already been discussed in the above portions of this Methods section, but several 1167 more specific aspects of our QA/QC are provided here. 1168 Throughout the permit file selection process, we developed and refined a series of 1169 rules and conventions for determining which files to pursue and which to consider 1170 outside the scope of this mitigation study. After our list of prospective files was 1171 generated, we went back through the original source list to ensure consistency. After all 1172 files were reviewed and categorized, we made sure that our conventions for excluding 1173 files were consistent. Several files ended up being excluded because of an incorrect 1174 interpretation of the permit file paperwork.
1175 The task of extracting the relevant mitigation compliance requirements from a 1176 permit file was exceedingly complex and difficult to standardize. While the permits 1177 usually follow a standard format, most permit conditions are not clearly delineated but 1178 are mentioned diffusely throughout the text of the permits, mitigation plans and other 1179 documents. Our rules and conventions for extracting these requirements evolved 1180 considerably throughout the course of the study. After the initial lists of conditions were 1181 developed and entered into the database, they were modified repeatedly as each permit 1182 file was subjected to subsequent reviews. In some cases, conditions that had been 1183 included were removed when we determined they were really procedural in nature or had 1184 to do with minimizing impacts during project implementation. In other cases, relevant 1185 conditions were added after they were missed in an earlier review, sometimes because 1186 they were in obscure portions of the file paperwork. Many permit conditions that were 1187 extracted verbatim were later divided when we determined they involved two or more 1188 distinct assessable conditions. The rules for scoring the permit conditions were also 1189 developed and refined throughout the course of this study and many site evaluations had 1190 been completed before the methods were finalized. Later in the project, after all data 1191 were collected, every condition of every file was reconsidered to ensure a consistent 1192 scoring approach.
1193 Despite attempts in CRAM development to reduce decision-making in the field 1194 and to improve scientific defensibility, there remained instances where differences in Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 29 1195 interpretation could lead to differences in data collection. Our previous experience with 1196 CRAM (Ambrose and Lee 2004) helped reduce these interpretation and decision-making 1197 issues substantially, as did the early field trials with members of the CRAM development 1198 team. After all the CRAM data were collected, we went back through all of the data 1199 sheets for every file to ensure that we had followed a consistent approach in all the 1200 evaluations. Numerous changes were made through this process, most in relation to the 1201 vegetation data and for the physical and biotic patch types. The plant community data are 1202 particularly noteworthy, as many species identification and consolidation issues were 1203 resolved through this process. For example, it was mentioned earlier that grasses and 1204 senesced annual plants presented unique challenges in our CRAM assessments. Through 1205 our QA/QC of the CRAM data, we discovered that the UCLA and USF groups diverged 1206 in their approaches to these issues and in their level of taxonomic resolution. The UCLA 1207 group had taken a more general approach to grass identification and had not included 1208 senesced annual plants in their evaluation. To maintain consistency, they went back 1209 through their data sheets and used site photos and other information to increase their 1210 resolution regarding grasses and senesced annual plants. The current version of CRAM 1211 included a provision that + or - modifiers be added to each of the letter grades; however, 1212 no rules for this procedure had been developed. After all other CRAM issues were 1213 resolved, we revisited our scoring decisions for every metric of every file to ensure that 1214 these grade modifiers were applied consistently.
1215 The outcome of the CRAM evaluation was profoundly influenced by the correct 1216 interpretation of the assessment area. As discussed earlier, the CRAM methodology was 1217 designed to assess complete wetland systems, and conventions had to be established 1218 regarding the application of CRAM for the evaluation of discrete mitigation sites. A 1219 considerable amount of time was spent ensuring that our project researchers understood 1220 these conventions. After the field season, the habitat acreage analysis forced us to go 1221 back through every file to carefully consider the actual acreage losses and gains that 1222 occurred through project implementation. One objective of this analysis was to assign a 1223 proportional acreage value to each CRAM evaluation within a particular file. During this 1224 procedure, numerous inconsistencies were discovered in the way our established CRAM 1225 conventions were applied. For example, a particular mitigation action might have 1226 involved restorative plantings on or above the stream banks, yet the channel itself was 1227 included in the assessment area. Alternatively, the CRAM evaluation for this project 1228 may have involved the correct mitigation site assessment area, but a second CRAM 1229 evaluation was done just for the channel. As we reconsidered these issues for every 1230 permit file, several changes were made, ranging from simple data adjustments to entire 1231 permit files being moved from the “fully assessed” category to the “compliance only” 1232 category or being excluded altogether.
1233 Measures were also taken to ensure that the data for our habitats analysis were 1234 consistent throughout. Understanding how to apportion a particular mitigation site into 1235 its component habitat types required some understanding of regulatory jurisdictions and 1236 wetland delineation. Fortunately, at least one member from each research group had 1237 taken a course in wetland delineation and we had intensive internal discussions regarding 1238 the jurisdictional framework of the regulatory agencies. Yet during the habitat acreage 1239 analysis that we performed after the field season, several inconsistencies were discovered Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 30 1240 in these jurisdictional habitat data. While some of these errors were related to the 1241 apportioning of individual habitat elements, most were caused by the same 1242 misinterpretations of assessment area that beset our CRAM evaluations. One consistent 1243 misinterpretation of particular relevance to this habitat assessment was the restricting of 1244 the assessment area to the wetland portion of the site. As a hypothetical example, if the 1245 permit requirements and mitigation planning documents indicated that a 1-acre wetland 1246 site would be created, then our assessments should include the mapped boundaries of that 1247 1-acre creation site, even if only one half of that area was actually wetland. While the 1248 purpose of the jurisdictional habitat assessment was to address this specific issue, many 1249 sites had been erroneously delineated as 100% wetland, even though the entire 1-acre site 1250 had been mapped. As we went back through every file to review the CRAM assessment 1251 area issues, we also resolved these jurisdictional habitat inconsistencies and then carried 1252 out the remaining portions of the habitat acreage analyses. 1253 After the field data collection phase was complete, the paper data sheets were 1254 scrutinized by the field team to ensure that all information was filled in correctly, 1255 consistently and legibly. Any calculated values (e.g., acreage or percentage calculations) 1256 were double-checked with a calculator, and then the data were entered. In order to 1257 reduce human error during data entry, the CRAM Access database was designed to only 1258 allow data entry in the appropriate format specific to that data table. For example, one 1259 electronic CRAM data form only allows the entry of letter grades A, B, C, D, etc. when 1260 entering data into this form. Each research group entered the data for their respective 1261 field evaluations.
1262 Once all data were entered, all computer files were double-checked against the 1263 paper data sheets to ensure that no errors occurred. Initially, 10% of the files were 1264 randomly selected and all data from those files were reviewed for completeness and 1265 accuracy in data entry. Through this process, enough errors were detected to warrant 1266 checking 100% of the files. This involved checking the data in our Access database both 1267 visually and using queries to ensure that there were no duplicate entries, blanks, or 1268 improper values (e.g., data that were out of the allowed range), and that data were 1269 completely entered into all relevant tables. These QA/QC procedures extended beyond 1270 our Access database and included a thorough review of all data relating to our GPS 1271 surveys. The GPS data were treated separately from the remainder of the field data and 1272 were not included in the Access database. The QA/QC measures taken with respect to 1273 the GPS data include ensuring adequate satellite geometry, maintaining a PDOP value 1274 around 2.00, differentially correcting the data using the nearest base station provider, and 1275 keeping a record of all base stations used in the differential correction of all files. In the 1276 end, every datum from every field form was double-checked against the databases, and 1277 all mistakes discovered were corrected. We are confident that the resulting dataset is free 1278 from significant data management errors.
1279 As mentioned above, ensuring consistency between the UCLA and USF research 1280 groups was challenging. Early in this project, both teams participated in a CRAM 1281 calibration meeting that involved field testing of the method to ensure user consistency. 1282 Then, to ensure that both groups were employing a consistent approach, a member of the 1283 USF team joined the UCLA group for the first round of mitigation site field visits, and Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 31 1284 the project coordinator from UCLA later joined the USF group for two separate weeks of 1285 field work at northern California sites. Extensive phone and email correspondence also 1286 helped in this regard. After the field season, both groups were responsible for the 1287 QA/QC of their respective permit files. Then, after the majority of the QA/QC 1288 procedures were completed, members of the UCLA group traveled to USF to help them 1289 finalize their remaining data tasks. During that visit, enough data errors and 1290 inconsistencies in approach were discovered to warrant a second round of QA/QC 1291 procedures between groups. Through this process, every USF file was subjected to a 1292 thorough re-review, which involved rechecking all aspects of the data for consistency, 1293 including the permit review, permit compliance, CRAM, habitat acreage analysis, and 1294 GPS data. Once all data modifications were complete, they were re-entered into the 1295 computer databases and all relevant files were checked one last time to make sure that 1296 every datum was correct.
1297 4. Results
1298 This section presents results for the four principal components of the study: (1) 1299 permit review, (2) permit compliance evaluation, (3) evaluation of wetland condition, and 1300 (4) habitat acreage analysis. A final section combines elements from the individual 1301 sections to provide a synthesis of some of the study’s results. 1302 4.1. Permit Review
1303 As noted in the Methods section, we experienced numerous difficulties in 1304 selecting, identifying, and locating an adequate number of permit files distributed by 1305 region and year. The details of these complications are provided separately in Appendix 1306 1.
1307 Between 1991 and 2002, a total of 9,924 CWA Section 401 permit orders were 1308 generated by the 12 SWRCB regions and sub-regions. The greatest numbers of 401 1309 permits were issued in Region 2 and sub-Region 5S, followed by Regions 4, 9, 3, 8, and 1 1310 (Figure 1). Our initial goal was to assess at least 100 permit files across the state, 1311 apportioned by region according to the percentage of the total state 401 orders that each 1312 region had issued. The percentage values displayed in Figure 1 reflect the proportions of 1313 files issued within each region; these regional proportions were used to calculate the 1314 target number of files to be assessed by region, given our initial goal of 100 assessed 1315 files. In the end, we assessed 143 permit files (Table 3). Narrative descriptions of each 1316 assessed project are provided in Appendix 12. Of these, 129 were fully assessed for 1317 compliance, habitat acreage and condition, while 14 were assessed for compliance only 1318 (e.g., fees paid). In addition, we identified 13 permit files with either clear compliance 1319 shortcomings (i.e., impacts occurred but mitigation project was never undertaken), or 1320 expected shortcomings suggested by denials of site access. A list of these files has been 1321 provided to the State Board.
1322 Of the 429 permit files randomly selected and pursued at either the Corps or 1323 Regional Board offices, a large percentage (40%) could not be positively identified in the 1324 agency databases or located in the file archives (Table 3). Many files that were located 1325 (104 files) were excluded after further review because they did not have assessable Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 32 1326 mitigation projects. We had difficulties finding assessable files in all regions, but 1327 particularly in Region 9, Region 7, and the two sub-regions of Region 6 (the reasons for 1328 this are discussed in Appendix 1). Files that were potentially assessable but were not 1329 assessed for lack of time are included in this table for completeness, as are two multiregional files that had been issued directly by the State Board.6 1330 1331 Mitigation sites were more heavily concentrated in portions of the state with 1332 greater development pressure over the past 10-15 years (Figure 3), particularly the San 1333 Francisco Bay area, north of Los Angeles, Orange County, and San Diego. Several sites, 1334 especially those in the Central Valley (Region 5) involved a collection of shared 1335 mitigation banks, so there are fewer than 129 mitigation points on the map. Most regions 1336 had some “compliance only” files (Figure 4), with no particular pattern among regions 1337 except Region 4 having a somewhat larger number than the other regions. Surprisingly, 1338 the projects regulated by the various Regional Board offices (see regional tallies in Table 1339 3) did not always fall within the boundaries of those regions. For example several of the 1340 401 permits located in the southern portion of sub-Region 5R were issued by the 1341 Sacramento office (5S); two in the southern portion of sub-Region 5S were issued by the 1342 Fresno (5F) office and the San Francisco office (Region 2) permitted some of the projects 1343 within areas designated as Region 1. Alternatively, the perimeters of the regions and sub1344 regions, as indicated by the SWRCB GIS base maps, might not reflect their true 1345 jurisdictional boundaries. For the purposes of this study and our respective analyses, 1346 such permit files remained associated with the issuing regional office. 1347 The 143 assessed permit files involved 204 distinct mitigation sites or actions 1348 (Table 4). Of these, 62% (127 sites) were within or immediately adjacent to the greater 1349 project boundaries (onsite), while the remaining 38% (77 sites) were offsite. There was 1350 no obvious geographic pattern to the offsite mitigation sites (Figure 5). While the 1351 majority of permit files involved independent, file-specific mitigation projects, others 1352 involved third-party mitigation strategies such as mitigation banks or in-lieu fee 1353 payments. Some mitigation projects included both onsite file-specific mitigation and 1354 offsite payments for mitigation bank credits. In total, about 75% of the mitigation actions 1355 were file-specific, while the remaining 25% purchased or applied acreage credits at some 1356 larger restoration, creation, or preservation site. Of these latter actions, 30% involved the 1357 application of acreage credits within informal permittee-controlled mitigation banks. For 1358 the remaining 70%, a third-party approach was employed that included credit purchases 1359 at formal mitigation banks or in-lieu fee programs. Payments for acreage at formal 1360 mitigation banks recognized by the Corps and/or FWS made up the majority of these 1361 credit purchases, while three mitigation actions involved in-lieu fee payments to invasive 1362 species eradication programs. While several regions applied such strategies, the use of 1363 mitigation banks was especially prevalent in Region 5 (Figure 5). Of the 24 fully 1364 assessed files in Region 5S, 17 involved credit purchases at five mitigation banks. One 1365 of these mitigation banks was used by 13 files. Further details on mitigation bank 1366 projects are given in Appendix 9.

6 These two files were obtained inadvertently since multi-regional projects were not part of our file selection/regional apportioning methodology. Even though the files were potentially assessable, the files were excluded from our study because they were not selected in accordance with our selection protocol.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 33 1367 The files we assessed included both older and newer mitigation projects (Figure 1368 6). The number of 401 orders issued by the SWRCB gradually increased from 1991 to 1369 1998, declined through 2000, and then increased again through 2002. We had initially 1370 selected a roughly even distribution of files throughout the years, except for the early 1371 years prior to 1995 for which fewer 401 orders were issued. The distribution of assessed 1372 files roughly followed the distribution of certifications, but with disproportionately more 1373 1996-1998 and 2000 files, and disproportionately fewer 1992 through 1995 and 2002 1374 files. We did not assess any files with 401 orders issued in 1991, which is not 1375 unexpected given the low number of files available from that year. As is discussed in 1376 Appendix 1, we had a difficult time obtaining assessable files from the earlier years 1377 (1991-1994) due to the prevalence of unconditioned waivers issued during that period. 1378 For these 401 actions, the compensatory mitigation requirements of other regulatory 1379 agencies were often explicitly or implicitly invoked by the Regional Boards, but such 1380 requirements were not clearly indicated in the 401 certification orders, or in the SWRCB 1381 database. It is not clear why our sample included so many 1997 and 2000 permit files; 1382 for some unexplained reason, files from these years were more easily located and more 1383 frequently contained assessable mitigation projects. The reason that proportionally few 1384 2002 files were included might be because many mitigation projects had not yet been 1385 undertaken.
1386 Nearly half (46%) of the 143 files we assessed represented permits given to 1387 developers (Figure 7). Municipal permits comprised almost a quarter of the files (24%). 1388 The California Department of Transportation (Caltrans), industry, private, and 1389 state/federal agencies each comprised 6-9% of the total number of files. Caltrans was 1390 distinguished from other state and federal permittees because of the large number of 1391 permits they received and the uniformity in the types of projects involved (mostly bridge 1392 crossings).
1393 In the following paragraphs we provide an analysis of assessed files by habitat 1394 type, impact type (permanent or temporary), and several aspects of the impact and 1395 required mitigation acreage. The data used in this analysis are not simple extractions of 1396 401 permit information taken directly from the SWRCB database or the 401 permits. 1397 Instead, they were derived from detailed reviews of all project-related information found 1398 in the permit files, including the 401 permit, the 404 permit and other agency permits, all 1399 mitigation planning documentation, and post-construction monitoring reports. Taken 1400 together, this information provided us with the most complete picture possible of the “as 1401 built” impacts and mitigations that occurred under the 401 program. During our permit 1402 reviews we discovered that the information obtained in this way frequently differed from 1403 the corresponding information taken directly from the 401 permits or the SWRCB’s 1404 permit tracking database. Through a specific analysis performed to understand the nature 1405 of these discrepancies, we found that the source of the differences ranged from simple 1406 data management issues to more substantive issues of potential regulatory concern. The 1407 results of that analysis are presented below, near the end this section. 1408 Wetlands were the habitat type impacted by the most files (Figure 8), although 1409 there were substantial impacts to habitats classified as “riparian” and “streambed,” as 1410 well as combinations of these three. A few files had impacts to non-streambed open Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 34 1411 waters, such as, lake and ocean habitats. Some files reported impacts to a single habitat 1412 type while others impacted multiple habitat types. For several files, the impacts were not 1413 well specified. Some of these listed impacts to unspecified “waters of the U.S.” while 1414 others did not provide any specificity for the impacts. 1415 For the overall acreage impacted and required, data from the files were 1416 consolidated and displayed by logarithmic size categories as appropriate for the wide 1417 range of acreages involved (Figure 9). These figures show that most files involved 1418 impact and/or required acreage values in either the 0.1 to 1 acre range or in the 1-10 acre 1419 range. However, a substantial number of files had acreages in the 0.01 to 0.1 acre range 1420 and, overall, the acreages involved ranged from 0.002 to 60 acres. The total acreage 1421 impacted and required for these 143 projects, as determined by our detailed file review, 1422 were 216.8 and 445.2 acres, respectively. Permanent impacts, totaling 166 acres, far 1423 outweighed the 51 acres of temporary impacts (Figure 10). 1424 In most years, more acres were required for mitigation than were allowed to be 1425 impacted (Figure 11). Ten percent of the projects (14) had fewer acres required for 1426 mitigation than were allowed to be impacted. The overall mitigation ratios were 1427 particularly large in 1996, 2000, and 2002. When the required mitigation ratios were 1428 calculated on an individual project basis and averaged by year, there also was no 1429 consistent temporal pattern in mitigation ratios through the years (Figure 12). The higher 1430 mean mitigation ratio in 1994, 2000, and 2002 were largely due to single files in each of 1431 these years with relatively large ratios (23:1, 70:1, and 123:1, respectively). 1432 The Regions differed in the amount of impacts and mitigation included in the 1433 permits we reviewed. Among the well represented regions (those with greater numbers 1434 of file assessments), the combined acreages of impact were relatively high in Regions 2, 1435 4, 5S and 8 while the combined impacts within Regions 1, 3, and 9 were relatively low 1436 (Figure 13). Regions 2, 5S, and 7 required the highest total mitigation acreage. Region 2 1437 also reflects the highest mean mitigation ratio, whereas Regions 5S and 7 had mitigation 1438 ratios that were similar to other regions. When considering the mean mitigation ratios 1439 required across the State, the regional patterns appear differently (Figure 14). For 1440 example, the mean mitigation ratio for Region 4 is the second highest, despite the fact 1441 that the total acreages impacted and required suggest a lower ratio similar to other files. 1442 The results for Region 7 (Figure 13) are notable in that the disproportionately 1443 high amount of impact and mitigation acreage occurred through just three permit files. 1444 This was primarily due to a large restoration project initiated by the United States Fish 1445 and Wildlife Service, wherein twenty acres of wetlands adjacent to the Colorado River 1446 were to be dredged to form a deepwater lake. The mitigation for this project was to 1447 include 40 acres restoration (invasive removal and riparian plantings around the lake), 1448 plus the lake conversion itself (20 acres). Although it was discussed in the 401 permit, 1449 the wetland acreage lost was not specified as impacts by the Regional Board and was thus 1450 not included in the SWRCB database. Even though there was no impact acreage listed, 1451 the permit (and database) included the 20-acre lake conversion as compensatory 1452 mitigation. The 40 acres of required restoration were not recorded as compensatory 1453 mitigation in the permit or database.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 35 1454 4.1.1. Discrepancies between file information and SWRCB database 1455 As indicated above, we discovered numerous discrepancies between the 1456 information obtained through our detailed file reviews and the corresponding information 1457 found in the 401 permits and the SWRCB database. Two examples illustrate such 1458 discrepancies: (1) for approximately 25 files, the database indicated wetland or 1459 streambed impacts that either did not occur or occurred in combination with other habitat 1460 impacts that were not recorded in the database; (2) according to the database, the selected 1461 files involved a little over 2 acres of temporary impacts, while we determined that, in 1462 fact, there were over 50 acres temporarily impacted. In addition, there were 1463 approximately 34 fewer acres of permanent impacts than reflected in the database. Data 1464 entry errors at least partially influenced these results. In the SWRCB database, there are 1465 data entry fields for habitat impacts (e.g. “Wetland,” “Riparian,” etc.), and temporary 1466 impacts (e.g. “WTEMP,” RTEMP,” etc.). According to the written conventions of the 1467 SWRCB, the former data fields are to be analogous to “total impacts,” and the latter 1468 fields are supposed to include the subset of the total impacts that are temporary. In 1469 practice, the ambiguity that is inherent in these data entry labels has led to substantial 1470 inconsistency in data entry. While we did not do a file by file analysis of this issue, our 1471 file information reviews identified numerous examples where the permanent and 1472 temporary acreage data were entered separately such that the sum of these data fields 1473 would equal the total impact acreage.
1474 There were considerable differences between the impact and required acreage 1475 values reflected in the database and the corresponding acreages that were ultimately 1476 involved. According to the SWRCB database the total acreage impacted and required for 1477 these 143 permit files was 198.9 and 241.0 acres respectively, while the corresponding 1478 values reported above were 216.8 and 445.2. Several files for which zero impacts were 1479 indicated did involve clear impacts. To understand how these differences varied among 1480 the files, we subtracted both impacted and required acreage values obtained through our 1481 detailed file review from the corresponding database values and plotted the resulting 1482 distributions (Figure 15). Of the 143 projects, approximately 48% (68 projects) had 1483 impact acreage differences between our file review and database. Twenty-one percent 1484 had fewer impacts indicated in the files than the database and 27% had greater impacts. 1485 The differences for most projects were below 1 acre, but the differences exceeded 1 acre 1486 for 10 projects. For required acreage, 63% (90 projects) had differences between the file 1487 review and database. For 53% percent of the projects (76 projects), information in the 1488 file indicated that more mitigation acreage was required than was indicated in the 1489 SWRCB database, while less acreage was required for 10% of the projects. For most of 1490 the projects, the discrepancy in acreage requirements was less than 1 acre. The 1491 discrepancies exceeded 1 acre for 31 projects.
1492 In order to understand the nature and source of these variations, a comprehensive 1493 acreage discrepancy analysis was performed. Every file for which our reported impact 1494 and/or required mitigation acreage differed from the database values was thoroughly 1495 reviewed. Impact and mitigation acreage data were extracted from each document in the 1496 file, including the 401 permit, 404 permit, streambed alteration agreement, biological 1497 opinion, and mitigation plan, plus monitoring reports and correspondence. The relevant Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 36 1498 dates were noted and the text of each document was read, in detail, for context. Based on 1499 the review, the final impact and mitigation acreage values were confirmed (our reported 1500 values), and a brief narrative was written for each file to explain the source of the 1501 discrepancy. Then the files were categorized according to the type of discrepancy. Files 1502 commonly contained two or more discrepancy categories. 1503 The complete results of this acreage discrepancy analysis, including narratives, 1504 are provided in Appendix 3. The main findings are summarized in Table 5. Among the 1505 143 randomly selected 401 permit files, discrepancies between our reported values and 1506 the SWRCB database values occurred in 101 files (71%). For 9 files (6.2%), the 1507 discrepancies were due to simple rounding issues and were inconsequential. For 26 files 1508 (18.2%), the discrepancies were caused by data entry or interpretation errors when the 1509 401 permit information was entered into the SWRCB database. Data interpretation errors 1510 were usually the result of unclear permit language and the lack of unambiguous acreage 1511 fields; other data entry errors included inputted values that were incorrect by a factor of 1512 10 (e.g., 0.07 acres instead of 0.7 acres). While database entry issues are troublesome, it 1513 is the content of the 401 orders that the Regional Boards rely on for compliance 1514 considerations. In comparing our results to the information extracted directly from the 1515 401 orders, discrepancies were still found for 60% of the files (86 files). For 19 files 1516 (13.4%), another regulatory agency simply required more mitigation acreage than the 1517 Regional Board, and we reported this greater acreage; these discrepancies are not errors, 1518 but simply reflect differences among agencies. These above categories amount to 1519 relatively minor quality assurance and quality control (QA/QC) issues. 1520 For 27 files (18.9%), the discrepancy was due to an accounting difference. For 1521 example, the Regional Board may have only considered wetland or permanent impacts 1522 while the project included impacts to non-wetland waters and temporary impacts, 1523 respectively. For 24 files (16.8%), the information in the 401 orders contained 1524 transcription, typographical, or interpretation errors indicating impact or mitigation 1525 acreage values that were clearly different from the planning documents available prior to 1526 401 issuance. Both of these categories reflect inconsistencies in the writing of 401 1527 permits and indicate that under the 401 program, the SWRCB may not always be 1528 regulating the full suite of jurisdictional impacts that are occurring. The extent to which 1529 these inconsistencies are understood and intentional is not known. 1530 Legally, it is the 401 permit, as written, that defines the regulatory scope of the 1531 SWRCB and the permittee must comply with the terms of that permit. Realistically, 1532 planning changes regularly occur following the issuance of the 401 permit, and we 1533 observed that the 401 permits did not always reflect the most current information 1534 regarding the project impacts and mitigation. Substantive changes in project planning or 1535 implementation that occurred after the 401 was issued resulted in discrepancies in 40 1536 (30%) of the files. For 12 of these files (8.4% overall), the impacts were not altered but 1537 there were changes in the context or acreage of the mitigation project. For five of these 1538 files, another agency approved modifications that resulted in greater mitigation acreage, 1539 but for the other seven, the approved changes resulted in lower acreage or a 1540 fundamentally different mitigation strategy (e.g., offsite purchase vs. onsite creation; 1541 riparian enhancement vs. wetland creation). These latter files would seem of regulatory Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 37 1542 concern to the SWRCB. The other 28 files involved changes in impact acreage. For three 1543 of these files (2.1%), the project impacts were reduced after the 401 was issued but the 1544 mitigation stayed the same. For another 13 files (9.1%), lower impacts were 1545 accompanied by a change in mitigation required by other agencies. Of these latter files, 1546 most had lower mitigation acreage than required in the 401 permit as a result of 1547 decreased impacts. However, at least two files contained a fundamentally different 1548 mitigation strategy. If the mitigation acreage undertaken was lower than that specified in 1549 the 401 permit, then this may be of concern to the SWRCB. However, if the lower 1550 mitigation was the result of impact avoidance understood and approved by other 1551 regulatory agencies, then such departures from the written 401 requirements might be 1552 judged less important. For the remaining 12 files (8.4%) out of the 28 files involving 1553 changes in impact acreage, changes during project planning or implementation resulted in 1554 greater impacts than reflected in the 401 permits and SWRCB database. Such files would 1555 seem of regulatory concern by the SWRCB.
1556 In all cases where the 401 permit information did not reflect later impact and/or 1557 mitigation adjustments, the planning modifications were approved by another regulatory 1558 agency (i.e., Corps, Fish and Game, or Fish and Wildlife Service). For most projects, we 1559 could find no evidence that the Regional Board was consulted or copied on the 1560 modifications; while one or more of the other agencies were regularly addressed on 1561 correspondence, listed on the documents as responsible parties, or included in copy-to 1562 lists, the Regional Board seemed to be largely omitted from the decision-making process 1563 after the initial 401 review. Note that our review was often based on files from the Corps 1564 rather than Regional Board files, so we might not have seen some correspondence. 1565 However, the Regional Board should nonetheless have been named on copy-to lists and 1566 other documents. These examples indicate that communication between the Regional 1567 Board and the permittees, consultants and other agency staffs involved in ongoing project 1568 planning and implementation occurring after 401 issuance could be improved. 1569 Among the 40 files with substantive changes in project planning or 1570 implementation after the 401 was issued, there were a few for which the Regional Board 1571 was copied on the changes, but these did not result in a modified 401 order. When 1572 modified 401 orders are created, they supersede the original order and the SWRCB 1573 database is to be updated with the revised impact and mitigation acreage information 1574 (also, the term “CERTMOD” is to be included in the notes field). We have found that 1575 this database updating is regularly done correctly. However, through the acreage 1576 discrepancy analysis, we found that for 7 of the 143 randomly chosen permit files (5%, or 1577 17.5% of the 40 files we reviewed that had changes after the initial 401 certification), the 1578 information from these revised certification orders (dates, acreages, etc.) was erroneously 1579 recorded redundantly in the database as separate records. 1580 The sources of the acreage discrepancies we found fall into three broad 1581 categories: (1) data management and QA/QC issues; (2) inconsistencies in the writing of 1582 401 permits; and (3) deficiencies in communication and follow-up after 401 issuance. 1583 Discrepancies falling into the first group, while notable, do not raise substantive 1584 regulatory/compliance concerns, while those from the other groupings may or may not 1585 raise regulatory concerns. To understand the extent of the regulatory/compliance issues Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 38 1586 indicated by the discrepancies, we performed a specific analysis considering the context 1587 and nature of the discrepancies for every file, judging whether they represented a 1588 substantive regulatory/compliance concern for the RWQCB/SWRCB. If the source of 1589 the discrepancy was limited to (1) a minor rounding error, (2) a database entry error, (3) 1590 another agency requiring greater mitigation acreage, or (4) reduced impacts with either 1591 no change in mitigation acreage or increased mitigation, then the discrepancy was not 1592 deemed a regulatory/compliance concern. However, if the source of the discrepancy fell 1593 within any of the other categories of Table 5, then the project was deemed of 1594 regulatory/compliance concern. The guiding principle that we employed here was 1595 whether the 401 order would have differed if the 401 manager had (1) seen, correctly 1596 interpreted, and correctly transcribed all the impact and mitigation information we found 1597 through our file review, and (2) employed an approach consistent to that of other 1598 managers regarding the accounting of temporary versus permanent impacts and wetland 1599 versus non-wetland waters impacts. Through this analysis, we judged that there was a 1600 regulatory issue for 60 files (42%). While some of these files involved transcription, 1601 interpretation, or accounting issues involving information available prior to 401 issuance, 1602 the discrepancies for 38 files were caused by 401 permits that did not reflect planning 1603 and/or implementation changes that occurred after 401 issuance. This highlights an 1604 important fact: because the Corps requires proof of 401 certification (or waiver) prior to 1605 issuing the 404 permit, permittees seek their 401 certification early in the regulatory 1606 process before some avoidance and minimization of wetland impacts occurred and before 1607 the mitigation planning is finalized. In such cases, communication and follow-up 1608 between the Regional Board and permittees, consultants and other agency staffs is 1609 essential if the project changes, and our results indicate that it often was insufficient. 1610 When the 401 order is issued based on preliminary planning information, the order (and 1611 the corresponding database information) could become outdated unless the Regional 1612 Board maintains an active role in the remaining aspects of regulatory planning and 1613 modifies the 401 certification if necessary. Our definition of “regulatory/compliance 1614 concern” assumes that the SWRCB would wish to regulate and track all wetland and 1615 riparian impacts (permanent and temporary) that occur within its jurisdiction. The permit 1616 files we documented with impacts exceeding those approved by the 401 permit would 1617 surely be of concern to the SWRCB; some of the other cases may be less important 1618 because, ultimately, it is the text of the 401 permit that the permittee must comply with in 1619 order to remain in compliance with the terms of the permit. 1620 4.2. Status of Regulatory Compliance of Compensatory Mitigation Sites 1621 Thirteen of the 257 permits we located had to be excluded because of potential 1622 compliance issues. This indicates that up to 5% of the files we reviewed may have 1623 significant compliance problems (such as the impact occurring but no mitigation being 1624 undertaken).
1625 For the files we were able to evaluate, the majority met most of their permit 1626 requirements (Figure 16), although fewer met all conditions to 100% satisfaction. Of the 1627 143 assessed permit files, 19 did not have any assessable 401 conditions (the 401 permit 1628 could not be located for 13 of these, although enough information was available from the 1629 Corps to locate and assess the site; whether these would have had assessable conditions is Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 39 1630 not known). For the remaining 124 files, the average 401 compliance score was 84% 1631 (Table 6). As described in detail in the methods, the average 401 compliance score 1632 (hereafter, average 401 score) was calculated as the mean of the compliance scores for all 1633 of the permit conditions; the potential scores for each of these conditions ranged from 0 1634 to 100%. Almost half (46%) of the files achieved perfect (100%) average 401 scores, 1635 indicating that they were in full compliance with all 401conditions; 57% had an overall 1636 score of 90% or greater, and 77% had average 401 scores of 75% or more. Three files 1637 received average 401 scores of zero.
1638 Compliance was also assessed by determining the percentage of permit conditions 1639 that were met completely (100% score) for a particular file (hereafter, average 401 1640 percent-met score). This approach to measuring compliance with 401 conditions is more 1641 consistent with regulatory evaluations, even though it is a more difficult standard with 1642 which to comply since the permittee is not given any credit for partially meeting permit 1643 conditions. According to this approach, on average 73% of a file’s 401 permit conditions 1644 were fully complied with (Table 6). Forty-eight percent of the files fully met more than 1645 90% of their conditions, and 57% completely complied with at least 75% of their 1646 conditions (Figure 16). Seven files did not meet any of their conditions to 100% 1647 satisfaction.
1648 Characterizing these files in terms of success or failure for compliance is not 1649 straightforward. For some files, the 401 requirements may have involved a single 1650 mitigation condition, such as an acreage requirement. Other files might have multiple 1651 conditions, including highly specific planting requirements and performance standards if 1652 the 401 permit had included a condition to follow the mitigation plan. There is no simple 1653 prescription for determining which aspects of the mitigation plan to include as assessable 1654 conditions; these documents are not organized in a way that makes this tractable. The 1655 “conditions” extracted from these plans were often difficult to assess and for many, the 1656 100% compliance criterion is unrealistic; nonetheless, we judged these to be in full 1657 compliance only if they were completely met. We placed near-misses in the 75% (mostly 1658 met) scoring category; therefore, we defined the lower limit of this category as the cutoff 1659 for “success.” Likewise the cutoff for “failure” was defined by the upper limit of our 1660 25% (mostly not met) scoring category. Given this convention, 76% of the permit files 1661 were considered successful according to the average 401 score and 4% were considered 1662 failures (Table 6). The remaining 20% were partially successful. According to the 1663 average 401 percent-met score, 57% were successful, 40% were partially successful, and 1664 13% were failures. Although a simple success/failure evaluation is not as informative as 1665 the numeric evaluations given in the previous paragraphs, we made success 1666 determinations to facilitate a simple summary of the compliance results. 1667 Although compliance with mitigation plans was included in the 401 compliance 1668 assessment if the mitigation plan was invoked (directly or indirectly) by the 401 permit, 1669 we also conducted a separate compliance evaluation for mitigation plans, since they can 1670 be viewed as a proxy for all agency requirements for file-specific mitigation projects. 1671 The majority of projects (57%, or 81 of the 143 permit files) contained mitigation plans. 1672 Mitigation plans were not included in the remaining files for a variety of reasons. For 1673 some files, plans were not required (e.g., mitigation bank credits purchased); for others, Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 40 1674 the plan was not in the agency’s file, presumably because it was misplaced or never 1675 submitted. Of the mitigation plans that were reviewed, some were relatively simple 1676 documents that described the general mitigation strategies; 16% of the 81 files had fewer 1677 than five conditions. The majority (84%) of the mitigation plans were detailed 1678 documents containing implementation plans and mitigation goals from which we 1679 extracted more than five conditions. The mitigation plan conditions for most (63%) files 1680 (44 of the 70 files for which we had conditions from both 401 permits and mitigations 1681 plans) had been invoked by the 401 permit and were included in the above 401 1682 compliance evaluation. The mitigation plan conditions for the remaining 37 files are 1683 unique to this analysis.
1684 The average mitigation plan scores for these 81 files (Figure 17) were somewhat 1685 lower than the 401 compliance scores for the total sample of 124 files (Figure 16). The 1686 average mitigation plan score was 81% (Table 6), only slightly lower than the average 1687 401 score of 84% for all 124 files. However, only 16% of the files had perfect scores (all 1688 conditions 100% met) compared to 46% for the 401 permits; 42% had scores of 90% or 1689 greater. On average 68% of a particular file’s mitigation plan requirements were fully 1690 complied with. In comparing the distributions, the files scored significantly lower for 1691 mitigation plan compliance than for 401 compliance both for the average scores 1692 (Kolmogorov-Smirnov 2 sample test, p<0.001) and for average percent-met scores 1693 (p<0.001). It would seem that mitigation plan conditions are more difficult to fully 1694 comply with than 401 permit conditions. This may be true; however, part of this 1695 discrepancy is due to the large percentage of the 401 permits with just one or two permit 1696 conditions (e.g., acreage requirements or credit purchases) with which compliance was 1697 relatively easy. Seventy of the files for which we had mitigation plan scores also had 401 1698 scores, so we could compare their scores by both of these measures. The average 1699 mitigation plan scores for these 70 files were significantly lower than the average 401 1700 scores (Wilcoxon signed-rank test, p=0.030), but the average percent-met scores were not 1701 significantly different (p=0.252). Of the 81 files with mitigation plans, 68% were 1702 considered successful for mitigation plan compliance using the criteria established above, 1703 32% were partially successful, and none were considered failures (Table 6). The average 1704 mitigation plan percent-met score was 68%. A total of 18 files (22%) had scores of 90% 1705 or higher. Two files did not meet any of the permit requirements to 100% satisfaction. 1706 Using this approach, 48% of the files were successful, 35% were partially successful, and 1707 6% were failures (Table 6).
1708 For the 124 files evaluated for 401 compliance, on average 30% of the permit 1709 conditions were not determinable (Figure 18). All permit conditions could be determined 1710 for 40 files (32%). Eighty-four files had at least some conditions that could not be 1711 determined, with an average of 45% non-determinable conditions per file. When 1712 mitigation plan compliance was considered separately, 30% of mitigation plan conditions 1713 were non-determinable (similar to the 401 compliance result). All conditions could be 1714 assessed for only 12 out of 81 (15%) files (Figure 19). Sixty-nine files had at least some 1715 mitigation plan conditions that could not be determined, with an average of 35% non1716 determinable conditions per file. The results from these two figures are indicative of the 1717 differences between the types of conditions listed in the 401 orders versus typical 1718 mitigation plan conditions. Aside from invocation conditions (those requiring that the Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 41 1719 mitigation plan or other agency permits be followed), the mitigation conditions specified 1720 in the 401 permit often consist of a single acreage requirement. Those containing more 1721 mitigation conditions often include a range of other requirements that, like acreage, tend 1722 to be addressed in a yes/no fashion or are not determinable (e.g., revegetation 1723 requirements, and monitoring and submission requirements). Mitigation plans include 1724 many more specific “conditions,” such as requirements for site preparation, 1725 implementation, and performance standards. While such conditions are less frequently 1726 complied with at the level of 100% satisfaction, they are also more frequently assessable 1727 in an after-the-fact assessment, such as the present study. 1728 One might expect compliance with 401 permit conditions to have increased 1729 through the years as the regulatory practices evolved; however, we did not find this to be the case (Figure 20; r2 1730 =0.000, p=0.845). There was no significant difference in 401 1731 permit compliance by year (ANOVA, p=0.959). Mitigation plan compliance was more 1732 variable through the years (Figure 21), and the correlation between compliance and year also was not significant (r2 1733 =0.030, p=0.119). As with 401 permit compliance, there was 1734 no significant difference by year (ANOVA, p=0.357). The scatterplot in Figure 21 1735 suggests a general increase in compliance through 1999 or 2000, followed by no further 1736 improvement. However, the leveling out after 2000 appears to be due to “maxing out” at 1737 100% compliance, so the only way to have continued improvement in compliance would 1738 be to have fewer low-compliance files. In any case, any temporal trend, if it exists at all, 1739 is slight, since there is no significant difference between the early files (1992-1997) and 1740 the more recent files (1998-2002) in 401 compliance (Mean±SE= 84.9±2.9 for 92-97 and 1741 84.0±2.7 for 98-02; t=0.223, P=0.824) or mitigation plan compliance (78.6±2.9 for 92-97 1742 and 82.4±2.7 for 98-02; t= -0.944, P=0.348).
1743 Overall, there was no significant difference in 401 compliance among regions 1744 (Figure 22; ANOVA, p=0.882). Similarly, there were no significant differences among 1745 regions for mitigation plan compliance (Figure 23; ANOVA, p=0.198). 1746 Average 401 permit compliance did not differ significantly by 401 certification 1747 type (Figure 24; ANOVA, p=0.159). Section 401 orders fell into four general categories:
1748 certifications, certifications with conditions, waivers, and conditional waivers. 1749 Regulatory practice evolved over the study period, and after June 24, 2000, issuance of 1750 waivers was no longer authorized by the State Board. Some of the regulatory orders also 1751 comprised waste discharge requirements (WDRs), either standard WDRs, conditional 1752 WDRs, WDR waivers, or conditional WDR waivers. We treated these as equivalent to 1753 the corresponding 401 certification categories and grouped them accordingly. In terms of 1754 a Regional Board’s level of involvement in the mitigation planning, one would expect 1755 certifications to include more involvement than waivers, and conditional orders more 1756 than standard orders. In practice, we found that the number of conditions from the 1757 various order types varied widely. From this study, it is unclear which certification 1758 category represents greater involvement by Regional Board staff. 1759 There were notable differences in the frequency of use of the various categories of 1760 permit conditions (Table 7). In general, the majority of mitigation requirements dictated 1761 the actual tasks to be completed during the preparation and construction of the mitigation 1762 site (i.e., site implementation tasks). For 401 compliance, site implementation tasks Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 42 1763 comprised the most conditions (30%), followed by monitoring & submission 1764 requirements (19%), success & performance standards (15%), and acreage requirements 1765 (12%). While acreage requirements comprised 12% of the conditions, only one or two 1766 such conditions were necessary for any particular file. Of the 143 permit files, 89 (61%) 1767 included at least one acreage requirement. For other condition categories, a given permit 1768 file may have had 10 or more conditions per category, especially when the mitigation 1769 plan was invoked by the 401 order. Fifty percent of the 401 orders invoked the 1770 requirements of other regulatory agencies or required that the mitigation plan be 1771 followed. Conditions involving mitigation site maintenance and the protection of the site 1772 from degrading influences, plus third party requirements (mostly credit purchases), made 1773 up a relatively low percentage of the conditions. For mitigation plan compliance, most of 1774 the “conditions” involved site implementation (39%), success & performance standards 1775 (21%), monitoring & submission requirements (16%), and acreage requirements (9%). 1776 Excluding the miscellaneous “other” category, the average number of conditions per 1777 category ranged from 1.4 to 6.0 for 401 compliance, and 1.6 to 7.9 for mitigation plan 1778 compliance.
1779 Compliance across the condition categories was variable. Third party 1780 requirements were almost always complied with fully (Figure 25). Monitoring and 1781 submission requirements had considerably lower compliance (about 60%), although this 1782 could be due to the fact that some monitoring documents were submitted but were not 1783 located in our review. The other categories had compliance scores of 75-85%. Except 1784 for third-party requirements, the percent-met scores were considerably lower than the 401 1785 scores. Acreage and credit purchasing conditions could usually be determined, while the 1786 conditions for other categories more frequently could not. Relatively few of the 1787 conditions in the success and performance standards category were non-determinable. 1788 Monitoring and submission requirements were more frequently non-determinable than 1789 other conditions, which is interesting since this category also had the lowest compliance 1790 scores when we could assess it. The patterns of compliance and non-determinability 1791 were similar for compliance with mitigation plan, although for mitigation plans, there 1792 was somewhat less variability among the categories (Figure 26). 1793 Because many of the permit, and even mitigation plan, conditions include purely 1794 administrative requirements (such as submitting reports) or actions that are only 1795 peripherally connected to the ecological functioning of a mitigation site, we analyzed 1796 compliance for a combination of condition categories deemed most relevant to the 1797 success of the actual mitigation project. These categories, shown in the last line of Table 1798 7, include the Site Implementation, Maintenance, Protection, and Success/Performance 1799 Standards categories. For this grouped category, the mean compliance scores were about 1800 80% for both 401 and mitigation plan compliance. The mean percent-met score was 1801 considerably lower, 63% for 401 compliance and 66% for mitigation plan compliance. 1802 All of the above 401 compliance results included the conditions found in 1803 mitigation plans and other agency permits that had been explicitly or implicitly invoked 1804 as a requirement of the 401 permit. In order to understand the contributions of the 1805 Regional Boards per se to the outcome of mitigation projects, we considered only those 1806 conditions specifically required by the 401 permits. A single mitigation-related permit Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 43 1807 condition was required for 27% of 401 permits (Figure 27). Another 18% percent of the 1808 permits contained two mitigation conditions, and 15% had three conditions. Ten permits 1809 (8%) specified 7-12 conditions, while eleven permits (8%) did not contain any 1810 mitigation-related permit conditions. These data do not include the eleven permit files 1811 for which no 401 permit was obtained. Among the 12 Regional Boards, Regions 6T and 1812 6V required the most mitigation requirements per 401 order (Figure 28), but there were 1813 just two permits for each of these sub-regions. Of the regions with larger sample sizes, 1814 Regions 2 and 4 included relatively more mitigation conditions per file while Regions 5S 1815 and 8 included relatively few.
1816 Of the mitigation conditions included in 401 permits, the majority involved 1817 acreage and third party acreage credit requirements, site maintenance requirements, and 1818 monitoring and submission requirements (Figure 29). Relatively few conditions 1819 specified the actual mitigation tasks to be implemented, protective measures, or success 1820 and performance standards. These data represent the conditions found in all 132 permit 1821 orders combined. When mitigation conditions from a given category were included in 1822 the permit order, there was, on average, between one and two conditions of that category 1823 per order (Figure 30). When present, there were close to two site maintenance and two 1824 monitoring and submission conditions on average per order, close to 1 site maintenance 1825 condition per file, and for acreage requirements, third party acreage credit requirements, 1826 and success and performance standards, there were approximately 1.5 conditions each per 1827 order.
1828 As indicated above, most 401 permit orders included 1 to 3 mitigation-related 1829 conditions. When a single condition was included, it involved a simple acreage or 1830 acreage credit requirement almost 90 percent of the time (Figure 31). Three single1831 condition orders contained site maintenance requirements and one contained a monitoring 1832 and submission requirement. Similar breakdowns are provided in Figure 31, for 401 1833 orders with up to four mitigation-related permit conditions. As the number of conditions 1834 increased, the proportion of maintenance and monitoring/submission conditions 1835 increased. Site protection, site implementation, and success and performance 1836 requirements were always a minor proportion of the conditions. These data demonstrate 1837 that most 401 permit orders included in this study contained relatively few permit 1838 conditions dictating the actions to be taken at the mitigation sites, or the success criteria 1839 upon which those sites would be judged. Instead, most permits specified the mitigation 1840 acreage requirements, included some site maintenance requirements, and mandated that 1841 mitigation and monitoring related documents be submitted. 1842 As we reviewed the files, extracted the relevant permit conditions, and 1843 consolidated the various agency conditions for our compliance analyses, we noted 1844 substantial overlap between the 401 conditions and the conditions required by other 1845 regulatory agencies. We performed a separate analysis to understand the extent of these 1846 redundancies. The conditions extracted from each relevant agency’s permit were aligned 1847 with those extracted from the 401 permit orders. Each 401 condition was scrutinized for 1848 equivalency with the other permit conditions. Some were verbatim copies of other 1849 agency conditions, while others were different in verbiage but equivalent in context. In 1850 all cases, our test was whether the greater mitigation responsibilities would have differed Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 44 1851 had a particular condition not been included in the 401 order. Overall, 62% of 401 1852 conditions were either redundant or invoking (Figure 32). Thirty-eight percent of the 401 1853 conditions were unique to the 401 permit. Those conditions unique to the 401 permit 1854 included all 401 conditions involving monitoring and submission requirements, which 1855 were 25% all 401 conditions. Excluding these since other agencies had their own 1856 submission requirements as well, about 13% of all 401 conditions were unique 1857 requirements of the 401 program. A breakdown of redundant and invoked conditions by 1858 region is given in Figure 33. Regions 6T, 6V, and 7 had the lowest percentage of 1859 redundant and invoked conditions, but these regions had very small sample sizes. 1860 Among the other regions with larger sample sizes, Region 2 included a relatively greater 1861 percentage of unique conditions in their 401 orders. Region 8 was unique among these 1862 latter files as having a relatively low percentage of invoking conditions. 1863 Considering the full set of conditions explicitly specified in the 401 orders, the 1864 mean permit compliance score was 84% (Figure 34). This score is identical to the overall 1865 mean compliance score given earlier (including invoked conditions from other permits). 1866 In addition, the distribution of scores is essentially the same as the earlier distribution. 1867 Because of these similarities, no further analyses were performed on these 401-specific 1868 conditions.
1869 4.3. Function and Condition of Compensatory Mitigation Sites 1870 CRAM evaluations were completed for 129 of the 143 permit files (14 files 1871 included in the above compliance evaluations did not contain assessable mitigation 1872 projects). These 129 files had 204 discrete mitigation sites due to multiple mitigation 1873 actions (e.g., depressional wetland creation plus riparian enhancement) that needed to be 1874 evaluated separately (Figure 3). Fifty three of these mitigation sites were sub-sampled 1875 because they were too large or complex for a single CRAM evaluation. These resulted in 1876 a total of 321 separate CRAM evaluations for this study. In addition, we performed 1877 CRAM evaluations for 22 reference sites across the State and added 25 more reference 1878 sites from the CRAM development team for a total of 47 reference site evaluations 1879 (Figure 2). CRAM results are presented below in two ways: one is by mitigation site 1880 with a sample size of 204, and the other is by file with a sample size of 129; for the latter, 1881 the scores of multiple mitigation sites were combined into a single overall score per 1882 permit file. Additional CRAM results that were too detailed for inclusion in the main 1883 report are provided in Appendix 7.
1884 The 204 mitigation sites were largely represented by low gradient riverine (46%) 1885 and depressional (36%) wetland classes (Figure 35). The remaining 18% of assessed 1886 mitigation sites, in decreasing order of occurrence, were vernal pool, estuarine, lacustrine, 1887 seep and spring, high gradient riverine, and lagoon wetland classes. Although mitigation 1888 sites were distributed throughout the state, the occurrences of each wetland class vary by 1889 region (Figure 36), with vernal pool and seep and spring mitigation sites only present in 1890 central to northern portions of the State. Similarly, estuarine sites were primarily in the 1891 north, though two estuarine sites were located on the south coast of California. While 1892 depressional and low gradient riverine sites were common throughout the state, 1893 depressional sites were more prevalent in the north, and low gradient riverine sites 1894 dominated in the South.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 45 1895 4.3.1. Total-CRAM Scores
1896 The total-CRAM scores for the 129 permit files assessed had a mean±SE of 1897 59%±1.1, with a median of 61% (Figure 37; Table 8). In comparison, the total scores for 1898 the 47 reference sites had a mean±SE of 79%±1.4, with a median of 82%. Based on the 1899 distribution of reference site CRAM scores, we classified sites in categories of wetland 1900 condition. The vast majority of the reference sites (89%) had total-CRAM scores of 70% 1901 or greater. For this reason, we established a 70% score as the cutoff for “optimal” 1902 wetland condition. We evenly distributed the remaining attainable CRAM scores into the 1903 three remaining categories. Thus, we defined the “sub-optimal” cutoff at 49%, and 1904 distinguished “marginal” from “poor” categories at 28%; in most cases, we have 1905 combined these categories and refer to them collectively as “marginal to poor.” 1906 Using these criteria, only 19% of the mitigation files were optimal, just over half 1907 were sub-optimal, and approximately one-quarter were marginal to poor (Table 8). Files 1908 with optimal and sub-optimal scores were distributed throughout the state, though there 1909 was a prevalence of marginal to poor files in northern California around the greater Bay 1910 Area (Figure 38) [see Appendix 5 for detailed mapping of mitigation and imact locations 1911 by region]. In our previous study of mitigation success in SWRCB Region 4, Ambrose 1912 and Lee (2004) found that just 2% of the files assessed had optimal wetland condition. 1913 However, in that study, optimal condition was defined as an 80% or above CRAM score. 1914 We established that criterion based on the quartiles of the 1-12 scoring scale, since 1915 reference site evaluations were not available for that study. The reference site 1916 evaluations included here suggest that the 80% criterion used in that study may have been 1917 too high; more of the permit files included in that study would have been considered 1918 optimal had a standard of 70% been applied.
1919 There was no relationship between CRAM score and certification year (Figure 39; r
2 1920 =0.005, p=0.415). Given evolving regulatory practices, one might expect more recent 1921 permit files to have mitigation sites with higher CRAM scores if more recent regulatory 1922 practices resulted in more successful mitigation projects. Alternatively, older sites have 1923 had more time to develop, so higher scores might be expected of these sites. Neither of 1924 these expected trends can be discerned for the actual relationship, with one possible 1925 exception. The CRAM scores for 2002 do not range as high as earlier years, which could 1926 be because these younger sites did not have enough time to develop sufficiently to score 1927 highly on CRAM.
1928 There were significant differences in Total-CRAM scores by region (ANOVA: F 1929 = 2.642; p = 0.005) with relatively low median scores in Regions 1, 2, and 6V, and 1930 relatively high scores in Regions 8, 9, and sub-Regions 5F, 5S, and 6T (Figure 40; Table 1931 9). Sub-Regions 6T and 6V had the highest (74%) and lowest (43%) median scores, 1932 respectively; however, these sub-regions had only two permit files each. When 1933 combined, the overall Region 6 score was comparable to the other regions (64%). A 1934 Tukey post hoc analysis revealed the differences between the low scores in Region 2 and 1935 the relatively high scores in sub-Region 5S (p = 0.006) to be responsible for the overall 1936 differences among regions. Region 2 had the highest percentage of marginal to poor files 1937 (52%), while Region 9 and sub-Region 6T had the highest percentage of optimal files 1938 (sub-Region 6T had only two permit files, both of which had optimal condition) (Figure Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 46 1939 41). Neither Region 7 nor sub-Region 6V had any optimal files, but they had very few 1940 files. Sub-Region 5R did not have any marginal to poor files, and the percentage for sub1941 Region 5S was low, even with a large number of files. However, the majority of files for 1942 these sub-regions had sub-optimal rather than optimal condition. The results for sub1943 Region 5S are notable due to the high percentage of those files that used formal 1944 mitigation banks. The standard error of scores from this sub-Region was low (Table 9) 1945 and this likely influenced the significance region effect. However, 17 of the 24 fully 1946 assessed permit files from this sub-region used 5 mitigation banks (13 files used a single 1947 bank; see Figure 5), and so the CRAM scores of those banks were repeated across these files.7 1948 A more in-depth analysis and discussion of mitigation banks is provided in 1949 Appendix 9.
1950 4.3.2. CRAM Attribute Scores
1951 We determined “optimal” cutoffs for each of the four CRAM attributes with the 1952 same criteria that were used to establish the overall “optimal” cutoff. Because the overall 1953 “optimal” cutoff contained 89% of reference sites above that score, we set each of the 1954 four attribute “optimal” cutoffs to the score with approximately 89 percent of reference 1955 sites above that score. For each attribute, we established the three remaining categories 1956 by evenly dividing the remaining attainable CRAM scores by three. Thus, for buffer and 1957 landscape context we established an “optimal” cutoff at 74%, “sub-optimal” at 52% and 1958 distinguished “marginal” to “poor” at 30%. We established a hydrology “optimal” cutoff 1959 at 76%, “sub-optimal” at 53% and distinguished “marginal” to “poor” at 30%. Physical 1960 and biotic structure attribute cutoffs were markedly lower than the overall CRAM 1961 cutoffs. Physical structure had an “optimal” cutoff at 53%, “sub-optimal” at 38% and 1962 distinguished “marginal” to “poor” at 23%, while biotic structure had an “optimal” cutoff 1963 at 47%, “sub-optimal” at 34% and distinguished “marginal” to “poor” at 21%. 1964 4.3.2.1. Buffer and Landscape Context
1965 The mitigation sites scored better for buffer and landscape context than for Total1966 CRAM. The median landscape context score for the 129 files was 72% (mean 66%) with 1967 a distribution that was skewed towards higher scores (Figure 42, Table 8). Similarly, 1968 reference sites scored well on landscape context with a mean score of 87% and a median 1969 of 90%. Most files had optimal scores, while roughly a quarter of files each were in the 1970 sub-optimal and marginal to poor categories. Region 7 and sub-regions 5S and 6T scored 1971 particularly well in the landscape context attribute while files for Region 1 and sub1972 Region 6V scored lower (Table 10). Overall, five of the regions had the majority of their 1973 files with optimal scores, and four regions (Region 7 and sub-Regions 5R, 5S, and 6T) 1974 did not have any files scoring in the marginal to poor category for landscape context. 1975 Despite criticism that mitigation projects are too often placed in proximity to 1976 development, these results indicate that the mitigation projects we assessed have been 1977 undertaken at sites that were reasonably well positioned in a landscape context.

7 Rather than report the score for a particular mitigation bank site just once, the score was assigned to all files that purchased credits from that bank since the functional losses from those projects were to be offset by mitigation bank site function.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 47 1978 4.3.2.2. Hydrology
1979 Hydrology attribute scores for the mitigation sites had a mean and median score 1980 of 63% (Figure 43, Table 8). In contrast, the reference sites scored well in hydrology 1981 with an overall median of 91%. Most (43%) permit files had sub-optimal scores, while 1982 27% had optimal, and 30% had marginal to poor scores. The Total-CRAM scores for 1983 sub-Regions 6T and 6V were reflected in their hydrology scores with the highest (81%) 1984 and lowest (36%) scores of all regions (Table 11), but these two regions had only two 1985 files each so these extreme values are likely a consequence of the small sample size. 1986 Two sub-regions of Region 5 (5F and 5R) also had higher scores, but when these were 1987 combined with large number of files from sub-Region 5S, the overall Region 5 hydrology 1988 mean was similar to other files. Regions 3 and 4 had the lowest hydrology scores, as 1989 Region 3 had the majority of files being sub-optimal and no optimal files, while 80% of 1990 Region 4 files were evenly split between sub-optimal and marginal to poor for hydrology. 1991 Improper hydrology has often been cited as the major shortcoming of mitigation 1992 project design (NRC 2001). The mitigation sites sampled during this project had lower 1993 hydrology scores than the reference sites, yet when compared to other CRAM attributes 1994 the site hydrology scores were not disproportionately poor. However, approximately 1995 50% of the assessed mitigation projects were classified and evaluated as riverine 1996 wetlands, and our conventions for employing CRAM were quite liberal with respect to 1997 stream-associated mitigation. Many of the riverine/riparian projects we evaluated did not 1998 include the channel itself. Instead, they occurred along the sloping banks of stream 1999 channels, frequently extending some distance away from the top of the banks. Others 2000 began at the top of the banks and extended outward from there, with even less connection 2001 to the channel. If the site was in direct proximity and seemingly hydrologically 2002 “connected” to the stream channel, the channel-dependent aspects of CRAM were scored 2003 as if the channel was part of the assessment area. Hence, many riverine sites that largely 2004 lacked wetland hydrology on the site were given more favorable scores for hydrology 2005 than the restoration site alone would have warranted. If we had taken a more narrow 2006 scope in defining the CRAM assessment area, hydrology scores would have been much 2007 lower. This is an important point regarding the utility of CRAM in evaluating mitigation 2008 sites, and it will be necessary to establish a standard approach for identifying assessment 2009 areas for future riverine mitigation reviews.
2010 4.3.2.3. Physical and Biotic Structure
2011 Mitigation sites yielded relatively low scores for both the physical structure and 2012 biotic structure attributes, with mean and median scores just above 50% (Table 8). 2013 However, the reference sites also scored lower for these two attributes and had wide 2014 variability in their scores (Figure 44 and Figure 45). For reference sites, the median 2015 physical structure score was 79% (mean 76%) and the median biotic structure score was 2016 68% (mean 67%). The overall low physical structure scores were mainly driven by low 2017 scores in the physical patch richness metric, while vertical biotic structure and biotic 2018 patch richness scores lowered the overall biotic structure attribute. Most files scored 2019 optimally in physical structure, with approximately a quarter of files in the sub-optimal 2020 and marginal to poor categories. The majority of files were optimal for biotic structure, 2021 about one quarter were sub-optimal, and only 12% were marginal to poor. As with Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 48 2022 hydrology, certain aspects of the physical and biotic structure attributes were channel2023 dependent. That is, the metrics were designed around physical and biological aspects of 2024 the stream channel. In cases where a hydrological link between mitigation site and 2025 channel existed, the channel was treated as part of the assessment area for those metrics. 2026 Region 2 had the lowest median score for physical structure (40%), with 48% of 2027 its files considered marginal to poor (Table 12). Similarly, only 25% of sub-Region 5F 2028 files were optimal, while neither of the Region 7 files was optimal. In contrast, Region 8 2029 had the highest mean score for physical structure (67%) and this region was joined by 2030 Regions 3, 4, 9, and sub-Region 5S in having a larger percentage of optimally scoring 2031 files.
2032 Regions 2, 3, 4, 7, and sub-Regions 5R and 6V all had a median biotic structure 2033 scores lower than 50%, with the two Region 7 files having particularly low scores (Table 2034 13). Region 2 and 4 had only 40% of files score in the optimal category, while 9 of the 2035 remaining 10 regions and sub-regions had the majority of their files score optimally. 2036 Similar to physical structure, Region 8 scored comparatively high for biotic structure, 2037 with a median score 65% with the vast majority of its files scoring optimally. 2038 With respect to physical structure, these results are not surprising. Most 2039 mitigation sites do not emphasize topographic complexity and physical patch types as 2040 design elements. However, the results for biotic structure are interesting given that most 2041 mitigation activities seem to focus on habitat improvement, namely the enhancement, 2042 creation, restoration, or preservation of plant communities. The focus of the biotic 2043 structure metrics was on these plant communities, requiring time intensive investigations 2044 into the diversity and cover of native and non-native plant species. These poor results 2045 from the reference sites for biotic structure suggest that CRAM is not calibrated to these 2046 design goals. (CRAM calibration efforts were being conducted at the same time we were 2047 assessing mitigation sites, so the results of those efforts could not be incorporated into 2048 our analyses.) However, even lower scores at mitigation sites indicate that they are 2049 falling short of design goals in this regard. The following sections highlight the main 2050 findings with respect to each of the 15 individual CRAM metrics. 2051 4.3.3. Individual CRAM Metrics
2052 The distribution of scores for individual CRAM metrics scores varied widely. For 2053 example, the percent of assessment area with buffer metric had a median score of 92%, 2054 while physical patch richness, biotic patch richness, vertical biotic structure, and native 2055 plant species richness had a median of only 42% (Table 14). In general, the majority of 2056 metrics had mean scores between 60 and 70%. The mitigation sites scored lower than the 2057 reference sites for all 15 individual CRAM metrics (Figure 46). Differences were most 2058 pronounced for the average width of buffer, buffer condition, water source, hydroperiod, 2059 hydrologic connectivity, and physical patch richness metrics. There was less difference 2060 between mitigation and reference sites for the six biotic structure metrics, percent of 2061 assessment area with buffer, and organic matter. However, the reference sites scored 2062 relatively low for the six biotic structure metrics and physical patch richness. This 2063 indicates a problem with CRAM calibration for those metrics, which will likely be 2064 resolved after CRAM is recalibrated. In the meantime, the relatively small difference Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 49 2065 between mitigation and reference sites for the biotic structure metrics could be either 2066 because the mitigation sites are doing relatively well in these areas or that the CRAM 2067 metrics are not sensitive to differences in condition that may be present at mitigation sites 2068 (perhaps because the reduced range of reference scores). We cannot distinguish between 2069 these two possibilities from the data.
2070 The 15 individual CRAM metrics scores varied by SWRCB region (Figure 47). 2071 Region 7 shows a particularly distinct pattern, perhaps due to the low sample size (only 2072 two files). Although it scored high (similar to the reference sites) for connectivity, 2073 percent of assessment area with buffer, and average width of buffer, it scored low on all 2074 biotic structure metrics. Region 2 scored particularly low in topographic complexity 2075 (46%) compared to the eight other regions, which averaged between 63 and 71%. 2076 Although Region 9 did not score especially high in the overall biotic attribute, it did 2077 remarkably well in the two plant metrics, exceeding the reference sites scores. 2078 4.3.4. Wetland Class
2079 The overall Total-CRAM scores varied widely within most wetland classes 2080 (Figure 48). Although CRAM was developed for use in a variety of wetland classes, it 2081 has not yet been calibrated for all wetland classes. Even the recent calibration effort 2082 focused on only two wetland classes, riverine and estuarine. Thus, it is not clear whether 2083 differences among wetland classes are due to differences in mitigation success among 2084 classes, or differences in how CRAM scores difference wetland classes. Since CRAM 2085 has been tested most extensively for riverine wetlands, we expect wetland condition to be 2086 most accurately reflected for this class. Appendix 8 discusses differences in CRAM 2087 scores for different wetland classes in more detail. 2088 4.4. Habitat Acreage Analysis
2089 The 143 Section 401 orders authorized approximately 217 acres of impacts and 2090 required that 445 acres of mitigation be provided; our analyses indicate that 417 acres of 2091 actual mitigation acreage was obtained (Figure 49). Overall, 94% of the required 2092 mitigation acreage was met. For the individual files, 72% met or exceeded their acreage 2093 requirements. Twenty percent (28 files) of the files exceeded their acreage requirements. 2094 For 52% of the files (73 files), we determined that the acreage requirements had been met 2095 exactly. Twenty-eight percent (40 permit files) of the files did not meet their acreage 2096 requirements. As noted in the methods, the obtained acreage values were based on GPS 2097 survey of sites where possible, review of files for mitigation bank purchases and other 2098 evidence of acreage met, and a combination of field visits and file review where GPS 2099 survey of sites was not possible. Roughly one third of acreage determinations were based 2100 on each of these approaches.
2101 There was no clear temporal pattern in how well the required acreage was met. 2102 The cumulative acreage requirements were shy of being met in most years with the 2103 exception of 1992, 1993, and 2001 (Figure 50). In 2001, the acreage requirements were 2104 exceeded by 3%, and the acreage requirements were met for the few 1992 and1993 files. 2105 These data are comparing total acreage obtained to total acreage required. When the 2106 average required mitigation ratios were compared to the average obtained ratios Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 50 2107 (gain/loss) by year, the results were more variable (Figure 51). The data in this figure 2108 represent the averages, by year, from one file to the next, whereas the previous figure 2109 showed the total sum of acreages by year. For about half the years the average gains 2110 exceeded the requirements, while for the other half they did not. There were two years 2111 (1992 and 1993) that met the requirements exactly. Although there were some 2112 differences from year to year, there was no general trend, such as earlier years achieving 2113 less than the required ratio or later years exceeding it, nor was there ever a very large 2114 difference between required and obtained mitigation ratio. 2115 Regions 2 and 8 exceeded their acreage requirements by 2 and 3%, respectively 2116 (Figure 52). All other regions fell slightly short of their acreage requirements, meeting 2117 from 38% (Region 6V) to 97% (Region 9). The regions that met the lowest percentage of 2118 their acreage requirements were Regions 6T and 6V which each had only two files—the 2119 lowest sample sizes of all the regions.
2120 While the mitigation acreage fell short of meeting the permit requirements, the 2121 regulatory process nonetheless yielded an apparent “gain” of 200 acres on 217 acres of 2122 impacts, which is an overall mitigation ratio of 1.92:1 (Table 15). However, this simple 2123 ratio is based on the assumption that mitigation sites included no existing wetland 2124 acreage before the mitigation project was undertaken. In fact, many mitigation actions 2125 consist of site preservation or simple vegetative enhancement to existing habitats without 2126 any changes in site hydrology; these types of mitigation actions cannot be considered 2127 acreage “gains” because there is no increase in wetland area. Since the simple mitigation 2128 ratio includes mitigation actions that do not actually increase wetland area, the ratio 2129 overestimates the contribution of compensatory mitigation towards achieving a goal of no 2130 net loss of wetland area. Details regarding acreage gained versus lost for particular 2131 projects are provided in Appendix 11. Also provided in this appendix are the raw habitat 2132 proportion data collected for each individual mitigation site. 2133 4.4.1. Riparian Jurisdictional Issues
2134 In addition to the problem of including mitigation actions that did not increase 2135 wetland area as a wetland “gain,” losses in certain habitat types were often compensated 2136 for by “gains” in other habitat types, and it was not always clear that the difference was 2137 an intended regulatory outcome. In this section, we separate the acreage losses and gains 2138 by their component jurisdictional and non-jurisdictional habitats, and attempt to 2139 distinguish true losses and gains in area from simple alterations of habitat. 2140 A substantial issue in evaluating acreage shifts is the consideration of riparian 2141 habitats that may not necessarily be jurisdictional wetland habitats. While essentially all 2142 impacts considered in the wetland regulatory process were to jurisdictional “waters of the 2143 United States” (two projects contained mitigation requirements for a combined total of 2144 4.40 acres of upland habitat), 27% of mitigation acreage consisted of drier “riparian” and 2145 upland habitats that were outside jurisdictional “waters” (Figure 53). Our “obtained” 2146 acreage assessments focused on mitigation habitats and did not include obvious buffer 2147 acreage or large conservation tracts that were built into the mitigation requirements. For 2148 individual files, part of this non-jurisdictional mitigation acreage may have been 2149 unanticipated by regulatory personnel (i.e., site location or mitigation action was different Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 51 2150 than proposed). However, the majority of this acreage involved site locations and actions 2151 that were proposed and subsequently approved. Of the acreage required to compensate 2152 for jurisdictional losses directly (buffers excluded), only 64% clearly involved 2153 jurisdictional mitigation acreage. Of the remaining acreage, 14% was to include creation, 2154 restoration, enhancement, or preservation of upland habitats and the other 22% was 2155 ambiguously listed as “riparian” mitigation without distinguishing whether jurisdictional 2156 or non-jurisdictional habitat was intended.
2157 It should be understood that “riparian” can be defined from an ecological or 2158 regulatory perspective. In determining riparian impacts, a regulatory definition is 2159 employed that considers only those riparian habitats within the ordinary high water mark 2160 (OHWM) defining “waters of the U.S.” (Under state law, the jurisdiction of DFG is 2161 extended to the outer drip line of the riparian vegetation.) However, in considering 2162 riparian mitigation, permittees and their consultants often use an ecological definition of 2163 riparian, which includes the entire zone of transition to fully terrestrial habitats. The 2164 lateral limits of “riparian” under this definition are vague and can include extensive areas 2165 that are beyond jurisdictional “waters.” When the mitigation requirements include the 2166 ambiguous term “riparian,” it is unclear whether the habitats mitigated were intended to 2167 be jurisdictional or non-jurisdictional riparian habitat. It should also be mentioned that 2168 impacts listed as “riparian” usually involved the entire riverine zone, including the 2169 channel itself and the portion of the floodplain and banks deemed within the OHWM. 2170 This usage does not conform to the most widely accepted definition of “riparian,” defined 2171 as the area between fully aquatic and fully terrestrial habitats and not including the actual 2172 riverine channel. Additionally, the term riparian wetland has been applied loosely and 2173 has often referred to both three-parameter wetlands and/or non-wetland waters habitats 2174 within the OHWM. Our determinations of “riparian waters” were limited to those non2175 wetland portions of the banks and floodplains between the channel and the OHWM. 2176 Aside from the non-jurisdictional acreage found in our site evaluations, the 2177 remaining mitigation acreage yielded a net “gain” of jurisdictional acreage with an 2178 overall gain/loss ratio of 1.43:1 (Table 15). Given the breakdown of habitat types, the 2179 mitigation associated with these 143 permit files resulted in overall net “gains” in both 2180 wetland and “non-wetland waters” acreage (Figure 54). There were 181 acres of wetland 2181 mitigation compared to 121 acres of wetlands impact, resulting in a net “gain” of 60 2182 wetland acres and a gain/loss ratio of 1.50:1. There were 75 acres of non-wetland waters 2183 impacted and 105 mitigation acres mitigated for a total gain of 30 acres (mitigation ratio 2184 of 1.40:1). The replacement ratio for “non-wetland waters” acreage was slightly lower 2185 than that of wetland acreage, but this might be expected given that the “no net loss” goal 2186 is focused on wetland habitats. Of the non-jurisdictional mitigation acreage, 70% was 2187 identified as non-waters riparian habitat and the remaining 30% was upland. While the 2188 acreage associated with these latter habitat types seems inconsistent with “no net loss” 2189 goals, the overall acreage of non-jurisdictional habitats was over and above net “gains” in 2190 jurisdictional wetland and non-wetland waters habitat. It is possible that some amount of 2191 this additional habitat was due to the increased jurisdictional requirements of the DFG; 2192 too few streambed alteration agreements were present in the permit files to test this. 2193 However, mitigation ratios are often proposed as a buffer, a way to account for 2194 uncertainty in the success of wetland creation or restoration, or to accommodate Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 52 2195 temporary losses occurring between impact and the completion of the mitigation project, 2196 and other sources of uncertainty. The inclusion of non-jurisdictional habitat in acreage 2197 considerations obscures the amount of buffer being incorporated into mitigation 2198 requirements.
2199 4.4.2. Permanent vs. Temporary Impacts
2200 To better understand acreage loss and gain, we distinguished permanent from 2201 temporary impacts and mitigation involving creation or restoration from preservation 2202 areas and habitat enhancements that did not increase the acreage of wetlands or waters. 2203 Comparing permanent impacts (true losses) to creation mitigation (closer to true gains), 2204 there was a net gain in overall acreage, and in the acreage of jurisdictional “waters” 2205 habitat (Table 16). In total, 76% of the impact acreage was permanent and 24% was 2206 temporary. In contrast, 65% of the total mitigation acreage was “created,” 24% involved 2207 habitat enhancement, and 11% was preservation (Figure 55). We did not include any 2208 large upland conservation/preservation areas associated with these permit files since these 2209 were usually required by FWS for impacts to endangered species and were tangential to 2210 the wetland impact/mitigation requirements. For jurisdictional “waters,” the overall 2211 gain/loss ratio was 1.37:1. For creation projects, the majority (82%) involved 2212 jurisdictional acreage. The jurisdictional acreage proportion was lower for enhancement 2213 projects (58%) and preservation areas (48%).
2214 Considering permanent impacts and creation mitigation, both wetlands and “non2215 wetland waters” habitats experienced gains of acreage (Figure 56). The overall 2216 replacement ratio for permanent wetland impacts was 1.38:1 while the ratio for non2217 wetland waters was 1.35:1. These data suggest that at least for overall acreage, 2218 mitigation required by the SWRCB and other regulatory agencies appears to be resulting 2219 in net gains of wetland acreage across the State. However, there is a caveat: many sites 2220 categorized as “creations” were in fact enlargements of existing wetlands, with both the 2221 created and pre-existing waters included in the reported mitigation acreage. 2222 It also is not clear how well “no net loss” of acreage is being achieved by 2223 individual mitigation projects, or if large gains from certain projects are compensating for 2224 net losses in others. In fact, 20% of the permits resulted in net losses (Table 17). 2225 Seventeen percent of the projects met their acreage requirements exactly, and 64% had 2226 net acreage gains. Thirty-three percent of the projects had net acreage losses in 2227 jurisdictional “waters,” while 22% had losses for wetlands. When permanent impacts 2228 (true losses) were compared to creation mitigation, only 41% of the projects yielded 2229 acreage gains while 20% met their acreage exactly and 39% resulted in net losses of 2230 acreage (Table 18). Almost half of the projects indicated net losses of jurisdictional 2231 “waters” habitats, and over one quarter of the projects (28%) resulted in net losses of 2232 wetlands. To determine if the projects with disproportionately large acreage gains or 2233 losses were skewing the results, we removed the five projects with the biggest acreage 2234 gains and the five with the biggest acreage losses from the analysis. Following this step, 2235 net acreage gains were still found with an overall gain/loss ratio of 1.67:1. For 2236 jurisdictional waters, the gain/loss ratio was lower (1.35:1), but for wetlands it was 2237 higher, at 1.68:1. While there were substantial deficiencies in habitat acreage for 20% of Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 53 2238 the projects, the large mitigation ratios required by the regulatory agencies have been 2239 successful in achieving overall net gains in wetland acreage within California. 2240 4.4.3. Regional Comparisons
2241 In our previous study, Ambrose and Lee (2004) found that net gains in overall 2242 acreage and in wetland acreage had been obtained within SWRCB Region 4. The results 2243 from this project indicate that these findings were consistent across the State. However, 2244 in that Region 4 study, Ambrose and Lee found an overall net loss in jurisdictional 2245 acreage, with roughly 50% of the mitigation acreage consisting of drier riparian and 2246 upland habitats that were outside “waters of the U.S.” This finding was not consistent 2247 across the State. When separated by the 12 Regions and sub-Regions of the SWRCB, our 2248 habitat acreage data show that most regions yielded net gains in both overall and 2249 jurisdictional acreage (Figure 57). Consistent with Ambrose and Lee (2004), Region 4 2250 experienced a net loss of jurisdictional “waters of the U.S.,” with over half (53%) of the 2251 mitigation acreage consisting of non-jurisdictional habitat. Sub-Region 5F and the two 2252 sub-regions of Region 6 also had net losses in jurisdictional acreage, though Region 6 2253 included just four files, and the loss for six projects of sub-Region 5F would not be 2254 apparent if all three sub-regions of Region 5 were combined. Sub-Region 5S was similar 2255 to Region 4 in that approximately 50% of the mitigation acreage (46%) was non2256 jurisdictional. However, unlike Region 4, Regional 5S had a net gain in jurisdictional 2257 acreage. For Region 7, 28% of the mitigation acreage was non-jurisdictional; however, 2258 like sub-Region 5S, this was in addition to net jurisdictional gains. Region 2, for which 2259 we assessed more permits than any other region, experienced the greatest “gain” in 2260 jurisdictional acreage. Sub-Region 5S had almost the same number of assessments as 2261 Region 2, and nearly as many impact acres. However compared to Region 2, sub-Region 2262 5S had relatively low jurisdictional gains. This region also has the largest number of 2263 mitigation bank projects, and had a mean required mitigation ratio lower than Region 2 2264 (Figure 14). Regions 5S and 7 achieved the highest cumulative gain/loss ratio of all the 2265 regions (2.91:1 and 2.90:1, respectively). Region 4 was also unique in requiring 2266 mitigation for impacts to non-“waters” habitat (coastal sage scrub and alluvial fan scrub 2267 uplands).
2268 For three of the southern California regions, wetland acreage made up a relatively 2269 low percentage of the regulated impacts and mitigated “gains” (Figure 58). The impacts 2270 in Region 4 were mostly to “non-wetland waters” habitat (79%). In Regions 8 and 9, 2271 wetlands comprised just 45% and 29% of impacts, respectively. On the other hand, 2272 wetland habitats comprised 9%, 49% and 61% of the respective jurisdictional “gains” in 2273 Regions 4, 8, and 9. Nearly all impacts in Region 1 were to jurisdictional wetlands, and 2274 these were compensated almost entirely through comparable wetland mitigation. Region 2275 9 had the highest overall gain/loss ratio (3.20:1), while Regions 4 and 7 and sub-Regions 2276 5F, 6T, and 6V all experienced net losses of wetland acreage. While all Regions except 2277 7, 5R, and 6T had some amount of upland mitigation acreage, Regions 2, 4, and sub2278 Region 5S were notable in this regard.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 54 2279 4.5. Combined Acreage, Compliance and CRAM Results 2280 Throughout the preceding sections, we have condensed our results into simple 2281 summaries of success, partial success, and failure. Although these summaries do not 2282 reflect the richness of the full results, they simplify comparisons across different aspects 2283 of the project. Most (72-76%) of the assessed permit files were successful in meeting 2284 their acreage requirements and other responsibilities related to permit compliance, but 2285 few (19%) were considered optimal in terms of wetland condition (Table 19). Thus, 2286 permittees are largely following their permits (although one-quarter to one-third of the 2287 time these are not met), but the permit conditions that are being met are not resulting in 2288 compensatory mitigation projects that are similar to natural wetlands. 2289 Since acreage and overall permit compliance are normally used as the primary 2290 indicators of regulatory mitigation success (i.e., post-mitigation functional evaluations are 2291 rarely performed), it is important to explicitly evaluate the relationship between these 2292 indicators and the condition of the mitigated wetland. Simply meeting acreage 2293 requirements did not ensure overall permit compliance (Figure 59; p=0.612, r²=0.002); 2294 not only was there no overall trend, there was a wide range of compliance values for 2295 projects meeting 100% of their acreage requirement. Similarly, there was no relationship 2296 between percent acreage met and CRAM score for wetland condition (Figure 60; 2297 p=0.169, r²=0.015). The range of CRAM conditions for projects with 100% acreage met 2298 was even broader than for compliance. Clearly, including sufficient acreage in a project, 2299 which is relatively easy to accomplish, had little influence on whether the project would 2300 be accomplished as required or if it would produce a high-quality wetland. 2301 Although compliance with the acreage requirement was not correlated with 2302 CRAM score, general compliance with permit conditions was. Mean 401 compliance 2303 score (Figure 61; p=0.000, r²=0.126), mean percent of 401 conditions met (Figure 62; 2304 p<0.001; r²=0.207), and mitigation plan compliance (Figure 63; p=0.001, r²=0.150) were 2305 all significantly correlated with wetland condition. However, the low r² values indicate 2306 the relationships between the variables were not very strong, with the compliance data 2307 explaining only 13-21% of the variance in the overall CRAM scores. Clearly, other 2308 factors influence the condition of mitigation wetlands, but compliance with permit 2309 conditions appears to have some influence.
2310 Since some permit conditions are more administrative in nature while others are 2311 directly focused on mitigation site performance, it is possible that certain categories of 2312 permit conditions might have a stronger relationship to wetland condition than others. 2313 Separate regression analyses were performed to compare the four condition categories 2314 deemed the most relevant to the CRAM outcome (Figure 64). No significant 2315 relationships were found between the overall Total-CRAM scores and the mean scores 2316 for the site implementation (p=0.219, r²=0.027), site maintenance (p=0.297, r²=0.068), 2317 site protection (p=0.743, r²=0.005), or success & performance standards (p=0.052, 2318 r²=0.091) condition categories. Most of the “conditions” included in these categories 2319 came from mitigation plans, rather than the regulatory permits themselves. When 2320 additional regressions were performed just for the set of conditions found in the 2321 mitigation plans, the relationship with the Total-CRAM score became significant for 2322 success & performance standards (p=0.024, r²=0.086). However, as with the other Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 55 2323 significant compliance relationships, the r² value was very low. This suggests that while 2324 compliance with performance standards is somewhat correlated with a positive CRAM 2325 outcome, the relationship is not very strong. Given the recent emphasis on success and 2326 performance standards in permitting and mitigation requirements, this latter result might 2327 seem surprising. However, the lack of a relationship highlights the fact that CRAM 2328 condition success means achieving the appropriate hydrological, physical, and ecological 2329 conditions at the site, while most performance standards are focused primarily on 2330 vegetation success. As a final test, we investigated the relationship between performance 2331 standard compliance and the CRAM biotic structure attribute scores: this is the portion of 2332 CRAM most closely focused on vegetation success. No significant results were found 2333 (p=0.196, r²= 0.042, for average 401 compliance; p=0.639, r²= 0.006, for average 401 2334 percent-met). Thus, it seems safe to conclude that while compliance was weakly 2335 correlated with CRAM, adequately meeting the permit conditions, even those 2336 performance-based standards, does not guarantee the mitigation site will be a well 2337 functioning wetland. This implies the need for on-going development of more 2338 appropriate standards which will ensure a stronger connection between permit conditions 2339 and overall functional development of mitigation wetlands. 2340 An analysis of these 143 files by permittee type (developer, industry, Caltrans, 2341 municipal, private, and state/federal) revealed notable differences in both mitigation 2342 requirements and outcomes (Table 20). As was mentioned earlier, Caltrans was 2343 distinguished from other state and federal permittees because of the large number of 2344 permits they receive and the uniformity in the types of projects involved (mostly bridge 2345 crossings). State/federal permittees had the highest mean impact acreage, were assigned 2346 among the lowest mitigation ratios, had the lowest obtained mitigation ratios, and had the 2347 lowest 401 compliance scores, though they had slightly better scores for mitigation plan 2348 compliance. Despite having lower permit requirements and compliance, state/federal 2349 permittees achieved the highest Total-CRAM scores. On the other hand, developers and 2350 industry-related permittees had lower mean impact acreages but were assigned the 2351 highest mitigation ratios, scored in the middle for permit compliance, and had the lowest 2352 Total-CRAM scores. Municipal and private entities had lower mean impacts (private had 2353 the lowest of all permittee types), while their mitigation requirements and mitigation 2354 outcomes were near the middle of the range. Caltrans projects had impact acreages near 2355 the middle of the range, but like other state/federal agencies had low required mitigation 2356 ratios, lower obtained ratios, and higher CRAM scores. 2357 It is not clear if the regulatory agencies assign mitigation requirements differently 2358 depending on the type of applicant, or if these mitigation ratios reflect the different types 2359 of impact or mitigation projects. For Caltrans, most permitted impacts involved bridge 2360 installation and repair projects. Due to the prevalence of temporary impacts for such 2361 projects, the mitigation required was often a 1:1 ratio and involved mere vegetation 2362 plantings in the associated channel. The CRAM scores for such mitigation projects are 2363 often high because of the pre-existing conditions in the channel. Other state or federal 2364 permittees might blend their mitigation responsibilities into larger restoration objectives 2365 and their actions are not as constrained by the typical concerns of “for profit” entities.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 56 2366 Industry permittees stand out in Table 20 as having exceptionally high mitigation 2367 ratio requirements, up to an order of magnitude higher than some other permittee types. 2368 This was due primarily to two files. The first involved the complete relocation of a 2369 stream channel from one side of a landfill site to the other. Only the loss of the channel 2370 itself was considered impacts (2.9 acre narrow strip of “waters” with no accounting of 2371 floodplain impacts), while the mitigation requirement included the new channel plus a 2372 wide non-“waters” floodplain and the banks of the stream, for a total of 44.0 required 2373 acres (required ratio of 15.2:1). The other involved 0.035acres of impacts and 4.3 acres 2374 of mitigation, a required mitigation ratio of 122.9:1. Without these two outliers, the 2375 required mitigation ratio for industry permittees was 2.0:1 and the obtained ratio was 2376 2.9:1. Overall, industry, municipal, and private permittees exceeded their mitigation 2377 acreage responsibilities, while developer, Caltrans, and state/federal permittees fell short. 2378 We include in Table 20 a summary statistic calculated by multiplying each file’s 2379 obtained acreage value by its respective Total-CRAM score (“Average CRAM-Adjusted 2380 Acreage” in the last row of the table). The purpose of this calculation was to qualify the 2381 mitigation acreage according to the condition of the site. For example, if a one-acre 2382 mitigation site had a 100% CRAM score, it would get “credit” for one acre. On the other 2383 hand, if the CRAM score was 50%, the site would get “credit” for only one-half acre, 2384 since its condition was not optimal. This is a simple, but relatively crude, method for 2385 adjusting raw acreages to account for the condition of the habitats produced. 2386 Because CRAM scores were less than 100%, the Average CRAM-Adjusted 2387 Acreage was substantially lower than the simple acreage gain estimate. We reported 2388 earlier that these 143 permit files impacted a total of 217 acres of impacts and obtained 2389 417 of mitigation acreage for an overall mitigation ratio of 1.92:1. We used the same 2390 approach of adjusting acreages by CRAM scores for these summary ratios, and the 2391 resulting mitigation ratio dropped to 1.04:1 (Figure 65). Although the mitigation ratio is 2392 substantially lower, it still indicates more adjusted acreage required as compensatory 2393 mitigation than acres lost, with the aforementioned caveat that any existing wetlands at 2394 mitigation sites are not incorporated into this ratio. Unfortunately, were not able to break 2395 these numbers down further by permanent gain and loss to jurisdictional acreage or 2396 wetlands, since this would be most relevant to the question of no net loss. 2397 5. Conclusions
2398 Impacts to wetlands in California are regulated by a variety of different agencies 2399 and regulations. Although the principle objective of this study was to investigate 2400 statewide mitigation success under the CWA Section 401 Water Quality Certification 2401 program, it is not possible to evaluate the success of the State’s 401 Program in isolation 2402 from the actions of other agencies, particularly the U.S. Army Corps of Engineers and the 2403 California Department of Fish and Game. This is particularly true because most 401 2404 permits “invoke” the mitigation plan for the project, which encompasses the requirements 2405 for the suite of agencies regulating the project. To a large degree, then, the findings of 2406 this project relate to the general compensatory wetland mitigation process in California. 2407 We have organized this discussion into a series of major issues. We start with the 2408 two major components of the 401 Program we evaluated, permit compliance and wetland Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 57 2409 condition. Included in the section on wetland condition is a discussion of how permit 2410 conditions could influence the success of wetland mitigation. Next, we discuss how 2411 mitigation replaced different habitat types and differences among the different Regional 2412 Boards. We then discuss issues related to mitigation banks. The final section considers 2413 the question of whether “no net loss” is being achieved in California. 2414 5.1. Permit Compliance
2415 Overall, compliance with 401 permit conditions relating to compensatory 2416 mitigation was reasonably high, though by no means perfect. Using a strict interpretation 2417 of compliance as having to meet each condition to 100% satisfaction, only 46% of the 2418 files with 401 conditions met 100% of those conditions, with another 50% at least 2419 partially in compliance. On average, 73% of a project’s 401 permit conditions were fully 2420 complied with. Although this percentage is fairly high, it is worth noting that the legal 2421 standard would be 100% compliance for all conditions, so fewer than half of all 2422 mitigation projects were in full compliance.
2423 The comparable figures for mitigation plan compliance were lower, with only 2424 16% of the files with mitigation plan conditions meeting all their permit conditions, and a 2425 mean by-file score of 68% of conditions met. Ambrose and Lee (2004) found that about 2426 2/3 of files for the LARWQCB met 100% of their permit conditions. This value is not 2427 directly comparable to the current study, however, because the compliance evaluations of the two studies differed substantially8 2428 . In the current study, fully meeting all conditions 2429 is a fairly high standard, particularly considering the fact that some of the conditions were 2430 extracted from the mitigation plan. In reviewing the mitigation plan, we had to judge 2431 what was a “condition” rather than having the conditions described explicitly. In 2432 addition, in many cases there were more than 20 or 30 conditions ranging from 2433 straightforward implementation conditions to complex performance standards. Even a 2434 relatively minor shortcoming in one standard would prevent a project from achieving 2435 perfect compliance.
2436 A more flexible way to judge permit compliance is to evaluate how well a 2437 condition was met, allowing for a fraction of perfect compliance (e.g., 75% met). The 2438 average 401 scores, according to this definition of compliance, were slightly higher than 2439 the corresponding “percent-met” scores, with a mean score of 84% across all files. For 2440 mitigation plan compliance, which includes the requirements of all regulatory agencies, 2441 the overall mean score was 81%. Regardless of which aspect of compliance was used 2442 (average scores or percent-met scores, 401 permit or mitigation plan) most projects 2443 largely met their permit requirements.
2444 When separated by compliance category, most of the average 401 compliance 2445 scores ranged from about 76% to 85%. Conditions relating to third-party mitigation 2446 requirements (mostly acreage or credit requirements) had a high average score (around

8 In the Ambrose and Lee study, conditions from the 401 permits that were not related to mitigation were included in the assessment and the evaluation did not include any “invoked” conditions from other permits.

We altered our methods for assessing compliance in the current study to provide more focus on compensatory mitigation, at the same time examining the entire set of mitigation requirements.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 58 2447 99%) while monitoring and submission requirements yielded a lower average score 2448 (about 59%). Acreage requirements were usually assessable, but for the other condition 2449 categories, a significant number of the conditions (regularly between 25% and 50%) 2450 could not be determined. Many of the permit conditions did not directly relate to 2451 mitigation actions that promote proper site functioning. When those condition categories 2452 were removed from the analysis (i.e. only those conditions relating to site 2453 implementation, site maintenance, site protection, and performance/success standards 2454 included), both 401 and mitigation plan compliance scores averaged about 80%. 2455 With compliance scores averaging about 80%, it appears that permit compliance 2456 has not been a substantial impediment to the success of compensatory wetland mitigation 2457 required by 401 certifications. While we encountered a few files with significant 2458 compliance shortcomings (13 such files were excluded from our study because the 2459 mitigation projects were never undertaken, despite project impacts), most mitigation 2460 projects met most of their permit conditions, or at least met the permit conditions we 2461 could assess.
2462 5.2. Wetland condition
2463 Understanding how wetland mitigation sites function is a key component of 2464 assessing whether the goal of no net loss of wetland acreage and functions has been met. 2465 In this project, we used the California Rapid Assessment Method (CRAM) to assess the 2466 condition of mitigation wetlands (as well as reference wetlands). Although CRAM is 2467 specifically designed to assess wetland condition rather than function, since it is based on 2468 a one-time “snapshot” of the assessment wetland, we view it as a reasonable indicator of 2469 wetland function.
2470 Only about 19% of the permit files we assessed were considered successful with 2471 respect to overall wetland condition. This was based on overall CRAM score as 2472 compared to the scores of relatively undisturbed reference wetlands, with “success” 2473 identified as an overall score greater than 70% (i.e., “optimal” category). These results 2474 indicate that the vast majority of wetland mitigation projects are not resulting in wetlands 2475 in optimal condition. Although 19% is a low success rate, it is actually somewhat higher 2476 than that found in previous studies but likely due to differences in how success criteria 2477 are identified. Sudol (1996), using a different assessment method (the Hydrogeomorphic 2478 [HGM] Assessment Method), reported 0% success in wetland mitigation projects in 2479 Orange County, California. Ambrose and Lee (2004), using a previous version of 2480 CRAM, reported a very low success rate (2%) for the Los Angeles/Ventura region. 2481 Although it is possible that the statewide success rate is somewhat higher than reported 2482 by Ambrose and Lee, more likely the difference is due to Ambrose and Lee’s previous 2483 use of 80% rather than 70% as the cut-off for optimal condition, suggesting that their 2484 results for LA/Ventura are likely comparable to the current results for the entire state. 2485 CRAM is still under development and future refinements will undoubtedly occur, so it 2486 may be difficult to compare directly the early applications of CRAM. Nonetheless, it is 2487 clear that very few mitigation wetlands have the same conditions as relatively 2488 undisturbed natural wetlands.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 59 2489 Mitigation sites tended to have relatively high CRAM scores for the “buffer and 2490 landscape context” metrics, but lower scores for hydrology, physical structure, and biotic 2491 structure. As discussed above, some of this variation may be due to differences in the 2492 relative effectiveness of CRAM for each of these metrics, but when compared with 2493 reference site scores, median mitigation scores were substantially different across the 2494 attributes. For example, for buffer and landscape context, the median score was 72 for 2495 mitigation projects vs. 90 for reference sites; the mitigation score was 80% of the 2496 reference. For hydrology, the median score was 63 for mitigation projects vs. 91 for 2497 reference sites; the mitigation score was 69% of the reference. For physical structure, the 2498 median score was 53 for mitigation projects vs. 79 for reference sites; the mitigation 2499 score was 67% of the reference. For biotic structure, the median score was 52 for 2500 mitigation projects vs. 68 for reference sites; the mitigation score was 76% of the 2501 reference. Mitigation sites appear to do worst in this comparison for hydrology and 2502 physical structure. As CRAM is calibrated and refined, more detailed comparisons 2503 among attributes will be possible.
2504 There was no relationship between year of certification and total-CRAM score. 2505 At least two factors might be expected to influence this relationship, and they probably 2506 work in opposite directions. On the one hand, since regulatory practice has evolved since 2507 401 certifications (or waivers) were first issued, one might expect CRAM scores to 2508 improve over time. That is, as regulators have changed the way they reviewed projects or 2509 the conditions they added to permits in order to improve the success of the mitigation 2510 projects, these improvements should lead to higher CRAM scores if they were effective. 2511 On the other hand, one might expect older mitigation projects to score higher because 2512 they have had more time to mature and develop optimal wetland conditions. Certification 2513 date is not the same as construction date since there is a variable lag between certification 2514 and when a mitigation project is actually completed, but certification date should be a 2515 reasonable proxy for age of a mitigation project. Other studies (e.g., Craft et al. 2003) 2516 have demonstrated that wetland structure and functions increase over time since 2517 restoration. In addition, some workers have argued that monitoring should be required 2518 for at least ten years to give the mitigation wetland time to develop so that any 2519 deficiencies would be more apparent. There is a slight suggestion that the youngest 2520 mitigation sites (certification date of 2002) did not achieve as high a CRAM score as 2521 older sites; however, no other pattern is apparent. Because there was no trend in CRAM 2522 score over time, it is not clear if either – or both – of these factors were acting. However, 2523 it is clear that any improvements in wetland condition that might have been caused by 2524 improved regulatory practice were swamped by other factors. 2525 The average compliance scores were not correlated with the CRAM scores, even 2526 when compliance with performance standards was compared to CRAM biotic structure. 2527 In other words, permit compliance did not guarantee optimal, or even high, wetland 2528 condition.
2529 5.2.1. Permit conditions
2530 Permit conditions guide mitigation projects to produce the types of wetlands 2531 needed to compensate for losses due to impacts. The conditions set the parameters of the 2532 mitigation project and, in theory, as long as these conditions are complied with, the Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 60 2533 mitigation project should provide appropriate compensation. In practice, compliance 2534 with permit conditions was not correlated with CRAM score, even when we considered 2535 only the conditions most directly related to mitigation performance. 2536 Does this mean that permit conditions do not influence the success of wetland 2537 mitigation? Probably not. However, it does appear that the conditions typically included 2538 in 401 permits and mitigation plans do not ensure that the mitigation wetlands have 2539 optimal condition, even when there is compliance with the permit requirements. 2540 Although more detailed examination of the relationship between compliance and wetland 2541 condition might provide some additional insight into this relationship, the general 2542 conclusion is likely to remain: a permittee can do everything required by a 401 permit 2543 and mitigation plan yet still produce a mitigation wetland lacking important 2544 characteristics. There are three areas of permit conditions that we suggest could be 2545 improved.
2546 First, permit conditions need to focus on a more important set of wetland 2547 characteristics. Currently, permits and mitigation plans focus largely on the vegetation 2548 component of wetlands, in particular the percent cover and survivorship of native species. 2549 Extensive planning goes into the species to plant, planting configurations, survival and 2550 growth, and prevention of non-native plant species. All of these are important. However, 2551 wetland ecosystems incorporate many aspects beyond simply plant cover, and the 2552 production of a well-functioning, sustainable wetland requires broader consideration 2553 (Ambrose 1995). Permit conditions should focus on the full suite of wetland functions 2554 and services (see Section 6.1.1).
2555 In general, the metrics incorporated into CRAM could serve as an initial guide to 2556 the types of wetland characteristics addressed by 401 permits. These metrics were 2557 selected by an experienced group of wetland experts to reflect wetland condition hence 2558 and they reflect the suite of characteristics a wetland should posses in order to have 2559 optimal condition. CRAM metrics do not include all aspects of a wetland that should be 2560 considered in permit conditions, however. Regulatory staffs should explicitly consider 2561 the functions and services lost at the impact site and ensure that the mitigation actions to 2562 be taken adequately compensate for those losses, so that the “no net loss” goal can be 2563 achieved (Ambrose and Lee 2004).
2564 Second, permit conditions should support closer tracking of jurisdictional losses and 2565 gains. In previous work in Region 4 (Ambrose and Lee 2004), we found that 2566 jurisdictional habitats (those within jurisdictional “waters of the United States), were 2567 being replaced with non-jurisdictional habitat, with the net effect of a loss of 2568 jurisdictional habitats. The current study confirmed that result for Region 4, but did not 2569 find an overall net loss of jurisdictional habitat statewide. Nonetheless, 401 certifications 2570 are rarely clear and precise about the types of habitats being impacted and replaced 2571 through mitigation. If a simple habitat classification scheme (e.g., 2572 Table 2) was used consistently in 401certifications, file documents, and the 2573 database, the accounting between habitat types impacted and created, restored, enhanced, 2574 or preserved through mitigation would be much clearer. This would help ensure that 2575 permit conditions will require compensation appropriate to the permitted impacts.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 61 2576 Finally, wetland mitigation might be improved if 401 permits included more 2577 conditions concerning the implementation and protection of mitigation projects and 2578 specifying success criteria/performance standards. Remarkably few permits included 2579 these types of permit conditions, and even when they were included in a permit, there 2580 were not many separate conditions specified.
2581 5.3. Changes in habitat types and acreage
2582 In previous assessments of the success of wetland mitigation projects, there has 2583 been little consideration of the fact that the habitats under consideration vary in their 2584 regulatory status. To remedy this problem, in Ambrose and Lee (2004) we distinguished 2585 between different types of habitats, and especially between jurisdictional and non2586 jurisdictional habitats, which allowed us to investigate “no net loss” with respect to 2587 acreage and individual types of wetland habitat. In the present study, we again evaluated 2588 impacts and mitigation according to the different types of habitats they affected. 2589 Our jurisdictional habitat evaluations demonstrate that, while essentially 100% of 2590 the regulated acreage losses were to jurisdictional waters of the United States (including 2591 wetlands, jurisdictional riparian habitats and other non-wetland waters), almost 30% of 2592 the mitigation “gains” involved riparian and upland habitats that were not jurisdictional 2593 “waters.” After isolating the jurisdictional waters portion of the mitigation acreage, the 2594 resulting overall gain (permanent losses versus creation gains) still gave an overall 2595 mitigation ratio of 1.4:1, but when the individual files were considered, only 36% had net 2596 acreage gains, 17% replaced their acreage exactly, and 47% of the files resulted in net 2597 acreage losses. This issue appears to be particularly important for riparian habitats, 2598 where there are wide-ranging definitions of wetland/upland boundaries used across 2599 agencies, and in a regulatory versus ecological context. 2600 For wetlands specifically, more acres were created than impacted. Forty percent 2601 of individual files resulted in net acreage gains (permanent losses/creation mitigation) 2602 and 28% resulted in net losses of wetland acreage. Our estimates of wetland habitat at 2603 mitigation sites represent the best-case scenario because we assumed no existing wetland 2604 acreage at the mitigation sites and we did not apply a strict three-parameter test. More 2605 acres on non-wetland waters were also created than impacted. Seventeen percent of 2606 individual files resulted in net acreage gains and 46% resulted in net losses. Thus, for 2607 both jurisdictional wetlands and non-wetland waters, our results indicate that there has 2608 been a net gain in acreage overall. However, a quarter to a half of all individual files still 2609 failed to replace fully the acres lost.
2610 This study confirms the findings of Ambrose and Lee (2004) that overall, the 2611 cumulative acreage of compensatory mitigation projects exceed the cumulative impacts. 2612 However, within the Los Angeles/Ventura Region, our previous study found that over 2613 half the mitigation acreage consisted of drier riparian and upland habitats that were 2614 outside jurisdictional waters of the U.S. In this study, we found that, while there was 2615 substantial non-waters mitigation acreage, this was over and above the net gains of 2616 jurisdictional acreage that were obtained.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 62 2617 Although acreage is an important component of the goal to have “no net loss” of 2618 wetlands, the goal also encompasses wetland functions. The achievement of “no net 2619 loss” of wetlands is discussed further in Section 5.6. 2620 5.4. Differences among regions
2621 We found no significant differences in permit compliance among SWRCB 2622 Regions. There was a hint in the data that Regions 8 and 9 might have slightly higher 2623 average 401 compliance scores, and Regions 2 and 3 slightly lower, but these differences 2624 were not significant.
2625 We discovered that some Regional Boards (e.g., Regions 4 and 9) considered 2626 shading for bridge/crossing projects to be a permanent impact, while others (e.g., Region 2627 5) considered only the actual bridge footings as permanent impacts with no mitigation 2628 required for shading except for bridges that were very low relative to the 2629 stream/floodplain elevation.
2630 With respect to mitigation wetland condition, some regional differences were 2631 apparent. There was little difference in Total CRAM scores among the regions with large 2632 sample sizes, except that Region 2 had a slightly lower mean score than some of the other 2633 regions. Differences in proportions of mitigation files in optimal, suboptimal, or 2634 marginal/poor condition were more distinct. The underlying cause(s) of the regional 2635 differences in mitigation wetland conditions are not clear. There was a slight (non2636 significant) indication that Regions 2 and 3 had lower compliance scores, but this seems 2637 unlikely to explain the differences since Region 3 was typical in its distribution of 2638 wetland conditions and overall there was no relationship between compliance and 2639 wetland condition. Differences in the geographic distribution of different wetland types 2640 might explain at least part of the difference. Region 2 had more depressional and 2641 estuarine wetlands, which had the lowest mean CRAM scores, than other regions. In 2642 addition, Region 2 includes a major urban area, which seems likely to constrain many of 2643 its mitigation projects. Region 4 also includes a major urban area; although its proportion 2644 of optimal sites was higher than Region 2’s and its proportion of marginal/poor sites was 2645 not as high, Region 4 did have more marginal/poor sites than some of the other regions. 2646 In contrast to the slightly lower scores we found, previous work by Breaux et al. (2005) 2647 for 20 mitigation sites in Region 2 found relatively high condition scores using the WEA 2648 method. Differences in the two studies could be due to differences in the sites sampled (5 2649 of the 20 sites studied by Breaux et al. were selected nonrandomly, whereas all of our 2650 sites were selected randomly) or methodology (e.g., WEA appears to result in 2651 consistently higher scores than CRAM).
2652 There were regional patterns in mitigation acreage requirements. While most 2653 regions experienced net gains in acreage, sub-Regions 5F and 6T had net losses, though 2654 both of these had relatively few permit file evaluations. The acreage for just two regions 2655 (Regions 2 and 8) exceeded the cumulative mitigation requirements, while the remaining 2656 regions fell short of their respective requirements. Compared to other regions, Regions 7 2657 and 8 stood out as having relatively high cumulative impact acreages given the number of 2658 permits involved. Region 7 had one file involving particularly large impacts. This result 2659 for Region 8 is especially noteworthy since that Regional Board had required the lowest Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 63 2660 cumulative mitigation ratio (1.15:1). Regions 2, 5S, and 7 had required the greatest 2661 cumulative mitigation ratios.
2662 Interestingly, the results for Region 4 were consistent with the Ambrose and Lee 2663 (2004) study, in that over half that region’s mitigation acreage (53%) consisted of non2664 jurisdictional riparian and upland habitats. While Region 4 had a small net gain in 2665 acreage overall, there was a net loss in jurisdictional acreage (14.6 acres lost, or 40% of 2666 the acreage not replaced). Region 8 and Sub-Regions 5F, 6T and 6V also experienced 2667 net losses of jurisdictional acreage. Region 4, 8, and 9 were the only regions requiring 2668 fewer jurisdictional acres of mitigation than impacted. Sub-Region 5S was similar to 2669 Region 4 in that approximately 50% of the gains were non-jurisdictional, though in this 2670 case, it was over and above a net gain in jurisdictional acreage. For Region 3 and sub2671 Region 6V, the proportion of non-jurisdictional habitat was around 31% and 38%, 2672 respectively, of the total obtained mitigation acreage, and for all other Regions and sub2673 Regions the non-jurisdictional acreage was 30% or less. 2674 5.5. Mitigation banks
2675 Our results indicate that compensation at mitigation banks yielded slightly higher, 2676 though non-significant average CRAM scores than project-specific mitigation (see 2677 Appendix 9). The lack of statistical significance could be due to differences in sample 2678 size between mitigation types (formal banks, informal banks and project-specific 2679 mitigation) and the wide range of habitat types which increased variation within each 2680 mitigation type, as well as any natural variation in these responses. For CRAM, the 2681 largest differences between banks and project-specific mitigation projects were in the 2682 hydrology and buffer/landscape context attributes. There were no differences in physical 2683 and biotic structure attributes between banks and project-specific mitigation. Given the 2684 importance of hydrology for mitigation wetlands, as noted above, our results indicate that 2685 banks should continue to be evaluated as a potential improvement to the mitigation 2686 process. There are a number of likely benefits associated with the consolidation of 2687 habitats in mitigation banks, and while our results do not show a strong difference in 2688 CRAM scores, the trends are informative.
2689 Ideally, a more focused evaluation of banks should be designed to compare a 2690 similar number of bank and file-specific projects of similar habitat classes within a 2691 particular region. This would reduce outside variation in CRAM scores, or other 2692 functional measures, and provide a more definitive comparison of the relative 2693 effectiveness of mitigation banks. However, given the actual distribution of mitigation 2694 bank projects within the state this could be difficult. We found that most banks were 2695 clustered in the Central Valley, with a small number of banks being developed in the 2696 Santa Rosa area, and others found sporadically around the state. A focused study within 2697 the Central Valley is most likely to yield high sample sizes. Similarly, banks vary in 2698 terms of habitat types, with most focusing on depressional, vernal pool, and riparian 2699 wetlands. There has not been clear distinction in some banks to differentiate vernal pool 2700 mitigation from other depressional wetlands. More consistent classification in this regard 2701 would be useful for future assessments of banks and other mitigation projects.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 64 2702 Although CRAM scores include aspects of biogeochemical functions, suggesting 2703 that mitigation banks are performing these functions adequately, they do not consider the 2704 geographic distribution of those functions. Mitigation policy has traditionally prioritized 2705 on-site mitigation over off-site mitigation, but the putative benefits of mitigation banks 2706 have many agencies reconsidering this prioritization. However, some wetland functions 2707 may not be replaced on a regional basis as effectively as others. In particular, water 2708 quality improvement, such as nutrient recycling or pollutant removal, provide an 2709 important service to a local watershed, and creating a similar function in a distant 2710 watershed does not seem the same. This may be especially relevant for mitigation banks 2711 in relatively undeveloped areas. In those cases, there will be relatively little gain in water 2712 quality improvement because water quality will already be good. In contrast, the loss of 2713 water quality improvement services at the impact site could be substantial from some 2714 developments (such as a residential development). When focusing on this one service, 2715 other mitigation strategies in the same watershed as the impact, such as removal of 2716 concrete lining from a channelized stream, might provide a better balance to the loss of 2717 water quality improvement services while maintaining geographic proximity to the 2718 impact (see Recommendations 6.1.2 and 6.1.5).
2719 5.6. Evaluating “no net loss”
2720 Our results indicate that, statewide, the overall acreage of compensatory 2721 mitigation projects has exceeded the wetland and other jurisdictional acreages impacted 2722 (see Section 5.3). Although the overall mitigation acreage exceeded the overall impacted 2723 acreage, a substantial portion of the files resulted in net acreage losses. In addition, the 2724 wetter jurisdictional areas lost were frequently replaced by drier riparian and upland 2725 habitats.
2726 A separate question is whether wetland functions are being replaced. Despite the 2727 obvious importance of assessing compensatory mitigation in terms of wetland functions, 2728 there have been remarkably few functional assessments in a regulatory context. In part, 2729 this may be due to the lack of a standard method for such assessments. There is a long 2730 history of wetland evaluation methods being developed for regulatory purposes, but most 2731 methods have had severe limitations. The Hydrogeomorphic (HGM) Assessment 2732 Method was developed specifically to address many of these limitations, and it is well 2733 suited for functional assessments in a regulatory context. In fact, Sudol (1996) used an 2734 early version of the HGM Assessment Method to evaluate Section 404 mitigation sites in 2735 Orange County. However, HGM requires regional models for each wetland type, and 2736 many compensatory mitigation projects in California would not have had an appropriate 2737 model available to assessment them. The California Rapid Assessment Method (CRAM) 2738 is being developed to fill the need for a simple method to assess wetland condition (as a 2739 proxy for function) at a wide range of wetland types in California. In this study, we used 2740 CRAM as an indication of the functioning of wetland mitigation sites. 2741 A more fundamental problem with assessing no net loss of wetland function is the 2742 study designs available for use. Functional assessments conducted at a mitigation site 2743 years after the mitigation was completed, such as we had to do, cannot indicate whether 2744 the policy of “no net loss” has been achieved. Determining the change in function 2745 requires measuring function at the impact site before and after impact to assess loss of Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 65 2746 functions, and at the mitigation site before and after mitigation to assess gain. Such an 2747 approach is not possible in an after-the-fact assessment such as the present study; in fact, 2748 we know of no large-scale survey that has been able to adopt this approach. 2749 Although our assessments of the current condition of the mitigation sites indicate 2750 whether the ultimate outcome of mitigation actions resulted in a high quality/functioning 2751 wetland, our data cannot address how much of the quality/function was caused by the 2752 mitigation action. It is likely that all current “function” was not attributable to the 2753 mitigation activities; in many cases, this was certainly the case. For example, many 2754 mitigation actions consisted of simple vegetative enhancements to pre-existing stream 2755 habitats and other “creation” projects involved slight enlargements of existing wetlands. 2756 Had comparative CRAM evaluations been done at these mitigation sites prior to the 2757 mitigation actions, many of the resulting scores might have been no different than our 2758 post-mitigation assessments. This would be especially true for hydrological and 2759 biogeochemical function, since most mitigation efforts focused on improving vegetation. 2760 In addition, in our decision about how to score sites that were adjacent to existing streams 2761 but did not include any actual stream habitat, we decided to give the mitigation site credit 2762 for the existing channel; although these sites were physically and hydrologically 2763 connected to the channel, in no way did they “create” those functions the CRAM scores 2764 credited them with. Despite the many cases where it was clear the mitigation actions did 2765 not create all of the wetland functions at the site, we could not assess how much gain in 2766 function might have occurred due to the mitigation actions because we had no 2767 comparable data on the pre-existing functions at each mitigation site. Similarly, we had 2768 no information on the loss in function caused by the impact site. With neither an 2769 assessment of gain nor an assessment of loss, a rigorous evaluation of no net loss is not 2770 possible.
2771 In our study of mitigation success for the Los Angeles/Ventura region, we tried to 2772 evaluate “no net loss” directly by assessing the beneficial wetland services lost due to 2773 project impacts and gained through the mitigation actions (Ambrose and Lee 2004). 2774 Through site visits and careful review of files, we gained insights as to the nature of the 2775 functional losses and gains. Through our resulting structured qualitative assessment, we 2776 determined that over half of the mitigation projects (66%) failed to compensate 2777 adequately for the full suite of beneficial services lost through the project impacts. 2778 Unfortunately, time constraints prevented us from performing a similar assessment in the 2779 present study. However, our anecdotal observations suggest that the results would have 2780 been similar if we had performed the same qualitative assessment. 2781 Although a rigorous assessment of no net loss is not possible in this study, the 2782 relatively low CRAM scores do suggest that the mitigation sites are not functioning as 2783 well as they could be. Since our reference sites were representative of the types of 2784 habitats that would have been impacted by the Section 401 projects, and the condition of 2785 the mitigation sites was considerably lower than the condition of the mitigation sites, it 2786 seems likely that the mitigation actions were not fully compensating for the functions lost 2787 at the impact sites. Our understanding of the 401 projects we evaluated is consistent with 2788 this conclusion. However, this conclusion remains unconfirmed pending a study using 2789 the proper study design.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 66 2790 6. Recommended Administrative and Regulatory Changes 2791 The recommendations are separated into five main categories (Table 21). First, 2792 we present recommendations aimed at improving mitigation requirements. These 2793 recommendations concern mainly permit conditions, but also issues of the location of 2794 mitigation projects and how gains and losses associated with a project are tracked by 2795 habitat. Second, we present recommendations under the general heading of Information 2796 Management. These recommendations concern improvements to the database (either the 2797 existing database, or a modified database), improvements to permit archiving, and 2798 improvements to tracking the progress of mitigation projects. Third, we present 2799 recommendations to improve the clarity of permits. Fourth, we recommend that the goal 2800 of “no net loss” be assessed in a more effective manner. Finally, we present 2801 recommendations concerning coordination with other agencies. 2802 To the extent possible, we have tried to ensure that the recommendations included in this section stem directly from the work done under contract to the SWRCB9 2803 . 2804 However, our previous study for the Los Angeles Regional Water Quality Control Board 2805 (Ambrose and Lee 2004) had a similar goal, and we produced an extensive series of 2806 recommendations in a Guidance Document to the LA Board (Ambrose and Lee 2004b); 2807 there are inevitably many similarities between those recommendations and the 2808 recommendations presented here. In addition, we acknowledge the influence of many 2809 other studies of mitigation effectiveness (e.g., Allen and Feddema 1996, Breaux and 2810 Martindale 2003, Breaux and Serefiddin 1999, Breaux et al. 2005, DeWeese and Gould 2811 1994, Kentula et al. 1992, Race 1985, Sudol 1996, Zedler 1996), as well as comments by 2812 State and Regional Board staff.
2813 Although the recommendations presented below are based on work done during 2814 this project, early results and recommendations were discussed with State Board staff. In 2815 addition, there are other ongoing efforts to improve processes associated with the 401 2816 Program. Thus, a number of these recommendations are already being implemented or 2817 are planned for implementation in the near future. For example, two database efforts, the 2818 California Integrated Water Quality System Project (CIWQS) and Wetland Tracker, 2819 would incorporate some of the information requested in some of these recommendations. 2820 6.1. Improving Mitigation Requirements
2821 The success of compensatory mitigation depends fundamentally on the mitigation 2822 requirements specified by the regulatory agencies. Our study found relatively high levels 2823 of compliance with mitigation permit conditions. In addition, there was no relationship 2824 between compliance with permit conditions and the condition of wetland mitigation sites. 2825 It appears that compliance with permit conditions is no guarantee that a mitigation 2826 wetland will have high condition or function. Perhaps the most effective way to improve 2827 the success of compensatory mitigation would be to include permit conditions that lead to 2828 better mitigation projects.

9 Thus, this is not an exhaustive list of how we think mitigation practice could be improved, but rather represents recommendations addressing issues we encountered during the present study.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 67 2829 6.1.1. Permit conditions should ensure complete compensation for the full 2830 suite of wetland functions and services lost.
2831 Wetland functions include a broad range of physical and biological processes. 2832 Many of these functions, such as flood water attenuation, groundwater recharge, water 2833 quality improvement (i.e., pollutant removal), and support of wildlife, provide valuable 2834 services for humans. To ensure that compensatory mitigation provides full compensation 2835 for lost wetland functions and services (also called values), discussion of project impacts 2836 and mitigation should be framed in terms of functions and services. 2837 Note: in this section, “wetland” is used in the broad, non-regulatory sense as a 2838 shortcut to the regulatory terms “waters of the United States and adjacent wetlands.” 2839 6.1.1.1. Permit conditions should place more emphasis on performance 2840 standards
2841 401 permits include conditions addressing various aspects of compensatory 2842 mitigation projects, one of which concerns the performance of the mitigation project. We 2843 found that the number of success and performance standard conditions included in most 2844 401 permits was relatively limited; only 15% of all permit conditions related to 2845 mitigation addressed success or performance standards. Thus, the basis for determining 2846 whether the mitigation project is successful is mostly not specified in the 401 permit; 2847 instead, performance standards are contained in other permits (e.g., 404 or 1600 permits) 2848 or the mitigation plan.
2849 In many cases, other permits or, especially, the mitigation plan may be an 2850 appropriate location for performance standards. For example, the details about a 2851 particular mitigation project are often not known until the mitigation plan is produced. 2852 However, the absence of particular success criteria or performance standards in the 401 2853 permit leaves the Regional Boards with less explicit input into the nature of the 2854 mitigation project. If the Regional Boards want to emphasize particular elements of the 2855 mitigation project (for example, see Recommendation 6.1.2), the 401 permit is the most 2856 effective place to require these.
2857 6.1.1.2. Performance standards should include hydrological and 2858 biogeochemical conditions as well as vegetation
2859 When performance standards are included in 401 permits, they often focus on 2860 aspects of vegetation or invasive plants. We do not recommend that fewer performance 2861 standards be required concerning native vegetation or invasive plants. In fact, the current 2862 attention on vegetation and invasive plants is well-founded on scientific studies of 2863 mitigation success.
2864 Despite the importance of vegetation and invasive plants, there are other 2865 important wetland functions that should be included as performance standards (see 2866 Section 2.2). General summaries of wetland functions, as well as functional assessments 2867 such as the Hydrogeomorphological (HGM) assessment method, include hydrology, Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 68 biogeochemistry10 2868 , and ecological functions. Permit conditions, however, rarely focus on 2869 hydrology or biogeochemistry. In addition, performance standards should include 2870 conditions that cover different ecological scales, such as population, community, and 2871 ecosystem conditions (Ambrose 1995). For example, at the population level performance 2872 standards could require successful reproduction for key species (especially habitat2873 forming species such as trees) to ensure sustainable populations. 2874 Although we found that, in general, hydrological and biogeochemical functions of 2875 wetlands were not addressed as completely as they should be in permit conditions, the 2876 necessary focus depends on the specific circumstances. In some cases, vegetation 2877 standards may need greater emphasis. Some trends were apparent for different wetland 2878 types. For example, “riparian” mitigation tended to be focused too heavily on vegetative 2879 plantings without appropriate hydrological improvements, while “seasonal/depressional” 2880 mitigation tended to involve excavation and seeding without enough plantings. 2881 6.1.2. Ensure that mitigation projects compensate for losses in water quality 2882 (pollution) improvement services
2883 Wetlands can remove pollutants, including excess nutrients, metals and bacteria, 2884 from water flowing through the wetland. This service is frequently cited as a key benefit 2885 of wetlands. Given the focus of Section 401 of the Clean Water Act on water quality, the 2886 pollutant removal capabilities of wetlands should be considered explicitly in 401 permits. 2887 This may best be achieved by having a separate analysis for impacts to water quality and 2888 how each of those impacts would be mitigated. (We use “water quality” here in the 2889 general sense relating to pollutants in water, rather than in the broader regulatory sense.) 2890 Water quality services provided by natural wetlands may be replaced incidentally 2891 by the compensatory mitigation projects that are typically required by 404 and 401 2892 permits. However, without a specific consideration of these services, it is impossible to 2893 know if these services are replaced fully. Systematic consideration of the effects of 2894 different mitigation alternatives on water quality may lead to a shift in priorities for 2895 mitigation for the Regional Boards. For example, treatment wetlands are often 2896 discouraged as a form of mitigation because ostensibly pristine wetlands could be 2897 replaced by urbanized wetlands with high pollutant loads. This may be a valid point from 2898 the perspective of ecological function, and a high-quality wetland may be required to 2899 mitigate impacts to ecological functions. But from the perspective of pollutant removal, 2900 treatment wetlands may be ideal for compensating for impacts to water quality. 2901 We discuss three examples where water quality services are especially likely to be 2902 overlooked.
2903 First, the compensatory mitigation projects we studied focused largely on the 2904 provision of habitat, and the upper, drier riparian habitat that is commonly a part of 2905 compensatory mitigation projects (see Section 4.4.1, Figure 54) provide relatively little

10 Wetland biogeochemical functions include processes that transport or transform different materials (see Section 2.2.2 for more detail). The breakdown of organic material and nitrogen cycling are two common biogeochemical functions. These functions support important services such as removal of nutrients or contaminants from water.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 69 2906 water quality benefit. While such habitats may replace many of the lost functions in the 2907 broader regulatory sense of “water quality,” they may not replace the functions that 2908 remove pollutants. To ensure the replacement of lost water quality improvement 2909 services, it may be necessary to add elements to mitigation projects in addition to the 2910 normal conditions focusing on habitat replacement. For example, a portion of the 2911 mitigation wetland near the water inflow point(s) might incorporate design features used 2912 in such as treatment wetlands, or treatment wetlands might be required outside the 2913 boundaries of the wetland used for general mitigation. It may be appropriate for the 2914 Water Board to require treatment wetlands for all large development projects to ensure 2915 that the permitted projects do not result in water quality impairment (i.e., pollution). 2916 Second, a specific analysis of water quality aspects might alter the mitigation 2917 required for some projects concerning “low quality habitat.” The term “low quality 2918 habitat” may be appropriate when considering the value of a habitat for plants or animals. 2919 However, from the perspective of water quality, such habitats may have significant water 2920 quality improvement function. For example, channels surrounded by development can 2921 have high potential for water quality remediation. Mitigation for impacts to “low quality 2922 habitat” tends to be limited because of the focus on habitat, but such mitigation may not 2923 adequately replace the water quality improvement functions performed by the original 2924 habitat. The Water Board should be careful to ensure that all functions performed by 2925 “low quality habitats,” especially water quality improvement functions, are fully 2926 replaced.
2927 Third, mitigation banks may be effective tools for replacing lost habitat functions, 2928 but, as currently designed, they may not provide adequate compensation for water quality 2929 impacts, particularly for services such as floodwater attenuation and pollutant removal. 2930 For many wetland functions, maintaining the function in the same region may be 2931 appropriate. The loss of water quality improvement functions or floodwater attenuation 2932 in a local reach may have far-reaching local consequences, however, which would not be 2933 compensated by a mitigation bank in a different location (see Section 6.1.5). 2934 6.1.2.1. Projects involving channelization, the installation of concrete 2935 linings, and cut and fill operations resulting in large scale drainage 2936 modification/culvert installation should be discouraged. 2937 When a stream segment is channelized, lined, or culverted, the hydrological, 2938 biogeochemical, and ecological functions and services lost are very difficult to mitigate. 2939 While this has been widely recognized and stream "improvements" are now discouraged, 2940 such projects are still occuring, often because the surrounding area is already urbanized 2941 and the stream is considered degraded and consisting of "low value habitat." This may be 2942 an accurate assessment with respect to ecological functions and services, however, such 2943 streams can be extremely beneficial with respect to water quality improvement (notably 2944 water pollution remediation). Large scale development projects with drainage 2945 modification can have particularly high net water quality impacts because the loss of 2946 water quality function is coupled with increased runoff and pollution input.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 70 2947 6.1.2.2. Promote channel daylighting and complete channel restoration 2948 projects (concrete removal) as compensation for biogeochemical 2949 impacts.
2950 One reason that losses of stream function are difficult to mitigate is that one 2951 cannot easily create stream systems in previously upland habitats. Most projects that 2952 called for riparian creation were, in fact, riparian vegetation projects within upland areas 2953 with little or no alteration of site hydrology. Some mitigation projects have attempted to 2954 create stream function by widening existing streams, or by creating side channels in 2955 upland areas that are fed by water diversions. Such projects can result in limited 2956 functional gains.
2957 In our previous study (Ambrose and Lee 2004), and again in the present study, we 2958 found that complete channel relocation and/or restoration projects, especially those 2959 involving the removal of concrete linings, can result in significant gains in hydrological, 2960 biogeochemical, and ecological functions and services. In urban setting (where concrete2961 lined channels often occur), habitat values can be limited due to buffer landscape context 2962 issues. Nonetheless, channel relocation/restoration projects can still provide substantial 2963 ecological functions and services, as well as providing mitigation opportunities in a 2964 setting where such opportunities can be limited.
2965 Although channel daylighting or complete channel restoration could open up new 2966 opportunities for replacing lost stream functions, such projects could be quite expensive 2967 and thus might not be feasible for all permittees. Large developers might be able to 2968 undertake projects such as these on an individual basis. In addition, mitigation banks 2969 could be developed to enable the benefits of channel daylighting or complete channel 2970 restoration to be realized even for relatively small individual projects. Mitigation banks 2971 have many advantages over permit-specific mitigation, but most existing bank projects 2972 have been focused on ecological functions and services, namely habitat for threatened 2973 and endangered species. Because the benefits they can impart to water quality 2974 improvement, and "no net loss" in general, the SWRCB should promote the development 2975 of mitigation banks involving full channel restoration (including dayighting and the 2976 removal of concrete linings). Channel daylighting and complete channel restoration 2977 might have relatively limited benefit if conducted in only small areas; mitigation banks 2978 would provide a mechanism for pooling efforts to achieve a more meaningful project. 2979 6.1.3. There should be a better accounting of the habitat types lost and gained. 2980 Permit documents should use a standardized habitat classification. Currently, the 2981 Section 401 Draft Guidance document indicates that five different waterbody types 2982 should be used in the Project Information Sheet: wetland, riparian, streambed, lake, and 2983 ocean. (For each waterbody type, the Guidance document indicates that acres of 2984 permanent and temporary impacts should be recorded.) Although these are all generally 2985 recognized waterbody types, our review of impact and mitigation projects suggests that a 2986 somewhat different classification could make it easier to track mitigation of impacts to 2987 jurisdictional habitats, which is an important step towards determining whether the goal 2988 of no net loss of wetland area and function has been achieved.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 71 2989 “Riparian” is a particularly problematic term. Impacts and mitigation concerning 2990 riparian habitats need to be more clearly defined to ensure that non-jurisdictional areas 2991 are not used to mitigate for jurisdictional impacts. The 401 Draft Guidance document 2992 defines riparian as “stream or lakeside jurisdictional water (below line of normal high 2993 water), vegetated, but not jurisdictional wetland (may be either wet or dry most of the 2994 time).” This definition seems to clearly restrict the use of “riparian” to jurisdictional 2995 waters, as is appropriate for regulatory use with respect to 401 and 404 permits. Impacts 2996 are generally delineated according to this definition, although occasionally we found that 2997 the entire jurisdictional area, including the stream itself, was termed “riparian.” 2998 However, mitigation planners have regularly applied a more ecological definition of 2999 “riparian” that includes both jurisdictional and non-jurisdictional habitat. Permits and 3000 mitigation plans seldom distinguish between these two habitat types. Thus, a non3001 regulatory definition of “riparian” is often being used in a regulatory situation. As a 3002 result, impacts to jurisdictional riparian habitat have often been compensated for by 3003 mitigation within non-jurisdictional riparian or even upland areas, resulting in a net loss 3004 of jurisdictional riparian acreage and values.
3005 A more useful terminology would clearly distinguish between areas classified as 3006 waters of the United States versus areas that are not waters of the United States ( for 3007 example, see Table 22). These main categories are distinguished based on regulatory 3008 considerations. Within each of these main categories, appropriate general habitat 3009 classifications are identified. These categories are based on those currently presented in 3010 the 401 guidance (and, in fact, those exact categories could be used if desired). The 3011 categories presented in Table 22 reflect the types of habitats frequently named in wetland 3012 permit documentation, as well as general types of wetlands recognized by wetland 3013 scientists.
3014 Besides standardizing the way habitats are described in wetland permits, Table 22 3015 provides a structure for tracking the areas of losses due to permitted impacts and gains 3016 from mitigation. The losses and gains (in acres and/or linear feet) should be recorded for 3017 wetland/riparian creation, restoration, enhancement, preservation for each of the habitat 3018 types, including transitional habitat and upland buffer areas. 3019 6.1.4. Mitigation projects should have appropriate landscape context 3020 One of the clearest differences between the CRAM evaluations of compensatory 3021 mitigation wetlands sampled in this study and their reference wetlands was their 3022 landscape context. In CRAM, landscape context contains four metrics, one for 3023 connectivity and three related to the amount and quality of the buffer around the wetland. 3024 The CRAM manual defines these concepts as:
3025 The connectivity of a wetland refers to its potential to interact with other 3026 areas of aquatic resources, such as other wetlands, lakes, streams, lagoons, 3027 etc., and their surrounding environs at the watershed or embayment scale, 3028 and to the likely relative importance of the wetland in the landscape 3029 context. Wetlands within a watershed or in the same embayment are often 3030 functionally connected by the flow of water, such that they have an Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 72 3031 additive influence on the timing and extent of flooding, filtration of 3032 pesticides and other contaminants, and the movement of wildlife. 3033 For the purpose of CRAM, a buffer is a zone of transition between the 3034 immediate margin of a wetland and its larger environment that is likely to 3035 help protect the wetland from anthropogenic stress. Areas adjoining 3036 wetlands that probably do not provide protection are not considered 3037 buffers. Buffers can protect wetlands by filtering pollutants, providing 3038 refuge for wetland wildlife during times of high water levels, acting as 3039 barriers to the disruptive incursions by people and pets into wetlands, and 3040 moderating predation by ground-dwelling terrestrial predators. Buffers 3041 can also reduce the risk of invasion by non-native plants and animals, by 3042 either obstructing terrestrial corridors of invasion or by helping to 3043 maintain the integrity and therefore the resistance of wetland communities 3044 to invasions.
3045 Mitigation wetlands frequently had poorer buffers and/or connectivity to adjacent 3046 wetlands (especially for riparian habitats). Because buffers and connectivity relate to 3047 conditions outside mitigation project boundaries, they may not typically be considered 3048 carefully in mitigation planning. However, poor buffers or low connectivity will 3049 adversely affect the functioning of a mitigation wetland. Mitigation projects should be 3050 planned with adequate buffers and functions.
3051 While adequate buffers and adjacent open space are extremely important for 3052 wildlife and other ecological functions and services, they may be less important when the 3053 purpose of the mitigation site is focused on flood control and water pollution remediation. 3054 6.1.5. Offsite mitigation should be within the same catchment, or at least the 3055 same watershed.
3056 While some functions can be replaced in another watershed, other functions (such 3057 as water quality improvement, floodwater retention, habitat connectivity) cannot. When 3058 mitigation occurs outside the catchment in which the impact occurs, some functionality in 3059 that system is lost. In some cases, mitigating those losses in a nearby catchment in the 3060 same watershed would provide adequate compensation for downstream impacts. For 3061 example, if impacts to a wetland reduces its ability to attenuate floods, then mitigation in 3062 the same catchment would provide the most appropriate compensation, but mitigation 3063 somewhere else in the same watershed would at least provide similar protection against 3064 downstream flooding.
3065 The problem of mitigation occurring outside of the catchment or watershed in 3066 which the impact occurred is especially prevalent with third-party mitigation. As 3067 discussed earlier (Section 5.5), mitigation outside the watershed, as occurs with many 3068 mitigation banks, may be especially problematic because the mitigation may occur in 3069 relatively undisturbed watersheds where these services may be less important.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 73 3070 6.2. Information Management Recommendations
3071 In this section, we discuss recommendations to improve the management of 3072 information associated with 401 permits. The performance of this study revealed the 3073 difficulty of retrieving specific permit files. Of the 429 files we sought, we could locate 3074 only 257. The difficulty in locating files had a variety of causes, ranging from limitations 3075 in the database to the physical management of hardcopy permit files. This section also 3076 includes recommendations designed to improve the ability to track the progress of 3077 mitigation projects.
3078 6.2.1. Improvements to Database
3079 Our review of mitigation projects depended on information from the SWRCB 3080 database for project identification. We used the database to select projects indicating 3081 compensatory mitigation requirements, and using the project information contained 3082 therein, attempted to identify and locate the physical permit files at either the Regional 3083 Boards, or Corps district offices. During the course of our extensive work with the 3084 database, we identified a number of areas that could be improved. 3085 Note: Recommendations 6.2.1.1to 6.2.1.4 can be implemented with the existing 3086 database. Although the existing database contains fields for the most important 3087 information concerning 401 permits, we have identified some areas that could be 3088 improved. These improvements would require that the database be modified, as reflected 3089 in Recommendations 6.2.1.5 to 6.2.1.11.
3090 Also note that, as an early action response to the preliminary findings of this 3091 study, the SWRCB began documenting ACOE file numbers in the database 3092 (Recommendation 6.2.1.2) in May 2005. To enhance data quality, file numbers are being 3093 entered duplicatively, discrepant field values are rechecked (Recommendation 6.2.1.3), 3094 and full project titles are being entered (Recommendation 6.2.1.1). In addition, we 3095 recommend a number of additional fields be added to the database. Many of the fields 3096 recommended are included in the California Integrated Water Quality System (CIWQS), 3097 an agency-wide data management system now being deployed that will store all water 3098 board data, and in “Wetland Tracker,” which Region 2 hopes to begin requiring soon as a 3099 permit condition in a pilot program.
3100 6.2.1.1. Full project titles should be entered into the database 3101 The location of permit files was much more arduous than expected because the 3102 information in the State Board database was not sufficient to identify a unique project in 3103 the Regional Board’s or Corps of Engineers’ respective databases. Generally, the project 3104 title was abbreviated, and therefore, lacked many relevant key words that would have 3105 facilitated cross referencing with other databases.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 74 3106 6.2.1.2. Additional critical information should be included within the “notes” 3107 field
3108 Much additional information is available in the 401 permit that would have been 3109 useful in the cross-referencing and identification of files using the Regional Board’s or 3110 Corps’s respective databases. Information such as the Regional Board’s permit ID 3111 number, the Corps’ 404 number, other agency permit numbers, and the county should be 3112 entered in the “notes” field of the database.
3113 Note: if the database is modified as recommended, it would include this 3114 information as database fields; see Recommendation 6.2.1.6. However, there is no 3115 reason to wait until the database is modified to begin entering this information. The 401 3116 guidance document indicates this information can optionally be included in the “notes” 3117 field.
3118 6.2.1.3. Each permit should be assigned a unique numeric or alpha-numeric 3119 identifier to be used by both the Regional Board and the State Board. 3120 While most Regional Boards assign each project a project identification number, 3121 their numbering formats are not compatible with centralized use by the State Board. 3122 Hence, these identification numbers have not been included in the State Board’s database. 3123 A consistent statewide format should be implemented and the State Board’s database 3124 should include a field for these primary identification numbers. 3125 Note: if a centralized database is developed as recommended (see 3126 Recommendation 6.2.1.5), a single permit identifier would naturally be assigned because 3127 both the Regional and State Boards would use the same database. However, there is no 3128 reason to wait until a centralized database is developed to assign a unique identifier. 3129 6.2.1.4. Database records should be entered using a quality assurance 3130 protocol.
3131 As would be expected in any extensive data entry project, there were a number of 3132 mistakes in the State Board database entries. A quality assurance protocol should be 3133 established to double-check entries. This would included, at a minimum: (1) checking 3134 whether the permit represented a modified or re-issued certification to avoid redundant 3135 data entry, (2) ensuring that all permanent and temporary impact to wetlands and non3136 wetland waters are included and that these are inputted into the correct fields per the 3137 established protocol (see Recommendation 6.2.1.8), and (3) checking entries for 3138 typographical errors. In many quality assurance programs, a certain percent of the entries 3139 (e.g., 10%) are checked independently for accuracy. This protocol would have to be 3140 integrated into any future changes to data entry methods. 3141 Although pure entry errors occurred, some database entry errors were due to 3142 misinterpretations of the permit information caused by ambiguous wording or the 3143 difficulty of having to extract important information that was embedded in the text of the 3144 permit (see Recommendation 6.2.2).

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 75 3145 6.2.1.5. A central database should be developed for use by both RWQCB and 3146 SWRCB to avoid redundant data entry.
3147 Currently, the State Board maintains a database for information from all 401 3148 certifications, and some Regional Boards maintain their own independent databases. 3149 There is a lack of correspondence between the fields in the Regional Boards and State 3150 Board databases. In addition, since much of the information required by the State Board 3151 is the same as required by the Regional Boards, there is unnecessary duplication of effort 3152 to maintain a series of independent databases.
3153 6.2.1.6. Database records should include fields for all critical information 3154 from a permit, and those fields should be adequately populated for 3155 every permit
3156 Within the State Board database, project descriptors were often abridged versions 3157 of the full titles found in the certification letters, and the county and other agency permit 3158 numbers were usually absent. With such limited information, it was difficult to identify 3159 and locate the physical permit files at either the Regional Board or Corps offices using 3160 their respective databases. The Section 401 Draft Guidance document specifies “to 3161 facilitate cross-referencing, include the U.S. Army Corps of Engineers’ (Corps) file 3162 number if it is available (Optional).” In practice, we found few files with the 3163 corresponding Corps number included. The database should included fields for the 404 3164 permit number and the numbers of other agency permits including the Department of 3165 Fish and Game’s 1600 permit and the Fish and Wildlife Service’s Biological Opinion. In 3166 addition, a field should be included for the county and the permittee’s consultant (if 3167 relevant). In the draft 401 guidance document, information such as this is identified as 3168 optional additional information that may be added at the Region’s option; we feel that 3169 critical administrative details, such as county and other agency permits, should be 3170 required fields in the database.
3171 Additional fields could also be useful in the database. For example, information 3172 fields for file attachments for permits, pre- and post- mitigation photos, and so forth 3173 would provide a broader view of the project. This information would be useful for later 3174 compliance evaluations, and might be entered by the permittee if electronic form 3175 submission is adopted (Recommendation 6.2.1.10).
3176 Having full project titles, county of project, and other agency permit numbers 3177 would greatly simplify any future efforts to evaluate the 401 program. Perhaps more 3178 importantly, though, it would ensure that each project is unambiguously identifiable. 3179 Clear identification of projects would be important for any action that needed to check 3180 project characteristics, including enforcement actions and (when the database has such 3181 capabilities) tracking mitigation monitoring or other compliance activities (such as 3182 paying in-lieu fees).

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 76 3183 6.2.1.7. Include GPS locations for the impact and mitigation sites in the 3184 SWRCB database
3185 The 401 draft guidance indicates that latitude and longitude information would be 3186 useful for GIS analysis of impact (discharge) locations; this information is listed as 3187 optional. With the ready availability of inexpensive GPS instruments, latitude and 3188 longitude should be required for all permits, for both the impact and the mitigation sites. 3189 As a minimum requirement, a single point location could be recorded for impact and 3190 mitigation site (or each of the mitigation sites, if more than one). 3191 Ideally, a survey-grade GPS would be used to determine the boundaries of impact 3192 and mitigation sites. Recent technological advances have made survey grade GPS units 3193 relatively affordable, and it would be reasonable to expect all future projects to provide 3194 an electronic GIS shape file with the specific boundaries of the mitigation project. This 3195 information could be submitted for GIS mapping and analysis by Regional or State Board 3196 staff. It would simplify the assessment of compliance with acreage permit conditions. 3197 6.2.1.8. Eliminate ambiguities between permanent and temporary impacts by 3198 including fields for “total impacts,” “permanent impacts,” and 3199 “temporary impacts.”
3200 Currently, the fields for total impacts and the subset of the total impacts that are 3201 temporary are not consistently being applied appropriately. As an example, the fields for 3202 wetland impacts include “wetlands” and “wtemp.” According to the database entry 3203 instructions, the total wetland impacts are to be recorded in the “wetlands” field and the 3204 subset of the impacts that were temporary are to be recorded in the “wtemp” field. In 3205 practice, permanent impacts were often entered into the “wetland” field and the 3206 temporary impacts were entered into the “wtemp” field. Data entry staff should be 3207 adequately trained to ensure that these fields are used appropriately. Alternatively, the 3208 confusion could be eliminated by having one field for total impacts, one for permanent 3209 impact, and one for temporary impacts.
3210 6.2.1.9. Permit conditions should be entered into the database 3211 Tracking the compliance of a compensatory mitigation project would be simpler 3212 if the permit conditions upon which compliance will be judged was recorded in the 3213 database. Having permit conditions in the database would simplify independent studies 3214 of compliance. When the database has capabilities for tracking project compliance, 3215 having the permit conditions specified in the database would reduce the amount of time 3216 needed to understand the crucial permit requirements and determine if they had been met. 3217 Currently, it would be difficult to extract the appropriate permit conditions from 3218 the permit file. However, Recommendation 6.3.2 recommends that permit conditions 3219 should be clearly delineated in the permit.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 77 3220 6.2.1.10.Have permittees submit permit information in electronic form 3221 Clearly, one of the difficulties of maintaining a database is the time required to 3222 enter the appropriate data. If the information needed for the database could be submitted 3223 by the permittee in electronic form, staff time needed to enter information would be 3224 minimized. Having an electronic form for permittees to fill out would also minimize 3225 database entries. Instead of having to enter all information (multiple times when separate 3226 databases are maintained by the State Board and each regional board), the basic 3227 information would need only to be checked, although additional information (such as 3228 permit conditions; see Recommendation 6.2.1.9) might have to be entered by Water 3229 Board staff. The form and database could be designed so the information from the form 3230 would flow simply into the database.
3231 6.2.1.11.The database should contain information to improve management 3232 after a permit is issued
3233 Information management for 401 permits currently seems focused almost 3234 exclusively on activities leading up to the issuance of a permit. However, post-permit 3235 activities are also critical for a successful 401 program. Better information about the 3236 project after the permit is issued would allow Regional Board staff to track the progress 3237 of projects and assist compliance and evaluation efforts. 3238 Post-issuance information that could be useful includes:
3239 • The database should track document submissions
3240 • The database should incorporate flags for overdue documents. 3241 • In concert with the fields for specific permit conditions, there should be fields for 3242 recording satisfactory compliance with conditions. 3243 • The database should track any enforcement actions undertaken on the permit. 3244 This type of information is included in CIWQS and is being proposed for the 3245 Wetland Tracker.
3246 6.2.2. Improve permit archiving
3247 During our previous study of permits at the Los Angeles Regional Board 3248 (Ambrose and Lee 2004), we discovered a number of issues associated with the archival 3249 of office hardcopy file management. Informal surveys of other Regions suggested that 3250 file organization and archiving at the Regional Boards did not support efficient file 3251 retrieval, making it necessary to perform our file reviews at the Corps district offices. 3252 Issues with hardcopy file management were also apparent in this project when we tried to 3253 locate specific files and either had difficulty locating them through the issuing Regional 3254 Board or the Regional Board was never able to provide us with a copy of the files. 3255 File archival is obviously important for a retrospective program evaluation such 3256 as this study, but it is also essential for tracking permit compliance, including compliance 3257 with submissions of monitoring reports. Obviously, it is difficult to establish compliance Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 78 3258 with a permit if the file cannot be located. Therefore, we recommend that permit 3259 archiving systems for each Regional Board be evaluated and improved if necessary. 3260 One particular addition to the database that could help with office hardcopy file 3261 management would be a chain of custody field for recording the location of physical 3262 permit file folder. This could avoid the problem of not knowing where the file is 3263 supposed to be, since sometimes staff keep files they are currently or have been working 3264 on at their desks.
3265 6.2.3. Improve tracking the progress of mitigation projects 3266 Various changes to the database could improve its ability to track the progress of 3267 mitigation projects after a permit has been issued (e.g., Recommendation 6.2.1.11). 3268 However, there are additional activities the Water Boards could undertake to improve 3269 project tracking.
3270 6.2.3.1. Track the submission of monitoring reports 3271 Monitoring reports provide a potentially simple and efficient method for assessing 3272 the progress, and potentially the compliance, of a mitigation project (see 3273 Recommendation 7.3.1). However, our review suggests that this tool is not being used 3274 effectively. Monitoring and submission requirements had among the lowest compliance 3275 rates of all categories we evaluated. Through a tracking field in the database or other 3276 means, monitoring reports (and other submission requirements) should routinely 3277 reviewed.
3278 6.2.3.2. Keep better track of credit purchases
3279 Currently, files for projects requiring mitigation bank or in-lieu fees often lack 3280 information about the payment of the required fees. In our assessments we found several 3281 examples where the evidence of fee purchases was submitted to one agency but not other 3282 agencies (see Recommendation 6.4).
3283 6.2.3.3. Track in-lieu fee payments
3284 We found some examples of in-lieu fee projects in which the money was paid, but 3285 not used (yet) for actual mitigation activities. For instance, several payments to the 3286 Center for Natural Lands Management were not applied to a mitigation site because no 3287 approved site was available at the time of fee payment. Several years had gone by in the 3288 interim and those projects appeared to have been forgotten about; at the very least, there 3289 was an extended period of temporal resource loss. It would be useful if a record could be 3290 made, either in the revised database (see Section 6.2.1.8) or elsewhere, when the payment 3291 was made and when the money was applied to mitigation. 3292 6.3. Improve permit clarity
3293 Permit conditions should be written as clearly assessable criteria, with individual 3294 conditions for each specific criterion to be evaluated. Permit conditions should be written 3295 with a clear and direct method of assessment in mind. Our results suggest that more Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 79 3296 clearly written conditions would improve the chance of compliance. Presently, some 3297 conditions are too vague or may be presented in a way that it is not possible to assess 3298 them.
3299 6.3.1. Important permit information, including impact and mitigation acreage 3300 and permit conditions, should be clearly delineated in tables and not 3301 buried within the permit text.
3302 After comparing the information in the 401 permits and database to the other 3303 regulatory permits, we found many cases where the database errors were the result of 3304 ambiguous language in the 401 permit. For example, the language of a permit may not 3305 have been clear whether two or more distinct impacts were additive or inclusive. 3306 Although these were considered database errors, it was clear that the cause was the 3307 difficulty in understanding the intent of the permit. The likelihood of such errors is 3308 higher when information for the database must be extracted from the text of the permit. 3309 Misinterpretations would be less likely if the key mitigation requirements were listed in 3310 tables.
3311 6.3.2. Permit conditions should be written so that efforts made in a small 3312 portion of the site cannot satisfy the verbatim text of the condition when 3313 the intention of the condition was that the efforts would be made 3314 throughout the site.
3315 In our compliance assessments, we frequently encountered situations where 3316 ambiguous phraseology in the permit requirements required that we assign a high 3317 compliance score to a mitigation project even though only partial mitigation efforts had 3318 been made. As an example, in assessing compliance with a condition that read “must 3319 remove invasive plants prior to planting,” we had to assign a high score even if we found 3320 evidence that invasive plants were removed from only a small portion of the site. When 3321 the intention of a particular condition is that the action or success standard would apply to 3322 the entire site, the condition should include such specifications (“…throughout the entire 3323 site”).
3324 6.3.3. Mitigation plans (and perhaps all permits) should include a table listing 3325 the requirements upon which compliance will be judged. 3326 Prior to permit issuance, all parties should understand and approve the conditions 3327 upon which permit compliance will be judged. These conditions have generally been 3328 scattered diffusely throughout the text of regulatory permits and mitigation plans. 3329 Summarizing these clearly and succinctly would ensure that all parties understand the 3330 permits and simplify future compliance evaluations. 3331 The mitigation plan is the most obvious place for a summary of permit conditions. 3332 The mitigation plan must incorporate the requirements from all permits for the project. 3333 In the plan, the permit requirements should be clearly delineated in table form. The 3334 development of this table should be a collaborative effort with all involved agencies (see 3335 Section 6.4) and not left solely to the permittee or consultant. In monitoring reports, 3336 assessment of compliance should be centered on this table (see Recommendation 7.3.1).

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 80 3337 The table of mitigation requirements should distinguish conditions required by 3338 different agencies. In addition, the conditions should be organized within the following 3339 categories: (1) Permittee-responsible acreage requirements, (2) third party acreage credit 3340 purchases, (3) mitigation site implementation, (4) mitigation site maintenance, (5) site 3341 protective measures, (6) success and performance standards, (7) monitoring and 3342 submission requirements, (8) invocation conditions (e.g., “follow the 404 permit”), and 3343 (9) other/miscellaneous.
3344 Although many of the specific mitigation conditions are not known until the 3345 mitigation plan is developed, and hence often cannot be included in the permit, 3346 understanding of exactly what was being required by the permit would be enhanced if 3347 each permit also included a summary table with an explicit statement for each condition 3348 included in the permit.
3349 6.3.4. Permits should be clear about the meaning of enhancement, restoration 3350 and creation.
3351 Enhancement, restoration and creation can all increase the amount of wetlands 3352 functions in ways that can be appropriate for compensatory mitigation, but the amount 3353 and nature of the increase varies, and the likelihood of success also varies. Thus, the 3354 terms should be useful carefully and consistently. The term “restoration” is often used in 3355 a general sense to encompass all three of these terms, but in permit analyses and language 3356 they should be used strictly.
3357 Enhancement refers to changes made to an existing habitat (e.g., wetland) to 3358 improve its functions or services. Enhancement does not increase the area of a habitat, 3359 which is an important consideration when assessing the goal of no net loss of wetland 3360 acreage. Because many physical processes may already be occurring before 3361 enhancement, enhancement projects may be the easiest to achieve successfully. Because 3362 some functions are typically occurring in the degraded habitat before enhancement, 3363 enhancement generally doesn’t produce as many functions or services (per unit area) as 3364 restoration or creation.
3365 Restoration refers to changes made to an area that was once, at some point in the 3366 past, the desired habitat (e.g., wetland), but has been converted to a different habitat type. 3367 Restoration returns the area to the desired habitat, with the general goal of achieving the 3368 level of ecological functioning found in the original habitat. Restoration increases the 3369 area of a habitat as well as the amount of functions and services provided by that habitat. 3370 Creation refers to the creation of a habitat in an area that had never supported that 3371 habitat. Because none of the physical processes or biological functions characteristic of 3372 the habitat, and required to sustain it, occur at the site before the creation, creation can be 3373 the most difficult type of “restoration.” Whenever wetland creation is required, wetland 3374 delineations, or at least proof of inundation or saturation appropriate for wetland 3375 development, should be included as permit requirements to ensure a wetland was actually 3376 created (see Recommendation 6.3.6).

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 81 3377 In its 2004 Final Mitigation Guidelines and Monitoring Requirements, the Los 3378 Angeles District of the Corps uses similar definitions, and has a similar assessment of 3379 benefits and risks of the different types of “restoration”:
3380 Generally, the physical characteristics of the sites considered determine 3381 whether establishment (i.e., creation), restoration, enhancement, or, more 3382 rarely, preservation are viable compensatory mitigation options. The 3383 categories of compensatory mitigation, as defined by Lewis (1990) are:
3384 Restoration: return to a pre-existing condition.
3385 Creation: conversion of a persistent non-wetland habitat into wetland (or 3386 other aquatic) habitat. Two subdivisions are recognized: Artificial (i.e., 3387 irrigation required) or self-sustaining.
3388 Enhancement: increase in one or more functions due to intentional 3389 activities (e.g., plantings, removal of non-native vegetation). 3390 Passive Re-vegetation: allow a disturbed area to naturally re-vegetate 3391 without plantings.
3392 Regulatory Guidance Letter 01-1 used the term establishment instead of 3393 creation. The former term will be used in this document for consistency 3394 with this Corps Headquarters’ guidance. Establishment projects have the 3395 greatest potential because, in theory, the full suite of functions performed 3396 by that habitat type are established; but they also have the highest risks. 3397 Establishing aquatic habitat in an area where it did not previously exist is a 3398 difficult proposition. Restoration projects have had a higher degree of 3399 success in the Los Angeles District. Despite the uncertainties associated 3400 with establishment projects, the Corps usually recognizes establishment 3401 and restoration equally when it comes to determining compensatory 3402 mitigation credit. Enhancement projects generally receive less 3403 compensatory mitigation credit, because enhancement targets particular 3404 functions instead of the full suite of functions performed by that habitat 3405 type. When enhancement is accepted, the Corps will require that the 3406 enhancement improve as many of the functions as possible. 3407 In common mitigation practice, restoration and creation focus on the addition of 3408 plants (normally facultative riparian or wetland species) to areas where they do not 3409 currently occur. These are not true restoration or creation projects. True creation and 3410 restoration projects add hydrological, biogeochemical and ecological functions to a site, 3411 typically through topographical modifications and/or the establishment or re3412 establishment of appropriate hydrology. Section 6.1.1 discusses the need to include the 3413 full suite of physical and biological processes in mitigation projects. 3414 Note that one other related term, preservation, is sometimes used in a mitigation 3415 context. Preservation occurs when an existing habitat (wetland or other) is protected but 3416 not manipulated. Although preservation may be an appropriate component of a Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 82 3417 mitigation requirement (see LAD ACOE guidelines for an example), preservation does 3418 not increase the amount of wetland acreage to compensate for acreage losses, nor does it 3419 increase the amount of wetland function or services to compensate for losses of those 3420 wetland attributes.
3421 6.3.5. When invasive species removal is required, performance standards 3422 should be clear about the goal of invasive species control 3423 In our evaluations, we found examples where invasive species eradication was an 3424 important goal of the mitigation and specifically required as a permit condition, and 3425 others where invasive removal and maintenance were required so that newly planted 3426 native species would have less competition for resources at establishment. However, in 3427 many instances, the goal of an invasive removal was not clearly defined, and while 3428 eradication may have been the intent, the permit language simply required removal. In 3429 such cases, we were forced to assign high compliance scores for the condition (some 3430 removal had occurred) even though substantial recurrence may have been observed. For 3431 some projects (e.g., site-specific invasive removal projects, or in-lieu fee payments for 3432 Arundo donax eradication), enhancement involving invasive species control was the 3433 entire mitigation project. Permits should be specific for the mitigation goal and the 3434 permit language should accurately reflect that goal. 3435 6.3.6. If a wetland is planned as part of a mitigation project, proof of 3436 inundation or saturation appropriate for wetland development should be 3437 required.
3438 We found several examples where one of the regulatory agencies had required 3439 verification of wetland hydrology or three parameter wetlands as a specific performance 3440 standard. Unfortunately, most wetland mitigation projects did not include such a 3441 condition. This condition should be included as a performance standard in all permits 3442 involving wetland mitigation.
3443 6.4. Improve the assessment of “no net loss”
3444 6.4.1. Pre- and post-construction functional assessments of impact and 3445 mitigation sites should be required to ensure no net loss of wetland 3446 functions
3447 Much of the interest about the success of compensatory wetland mitigation 3448 revolves around the question of whether “no net loss” of wetland area and functions has 3449 been achieved. It is very difficult to answer this question definitively with respect to 3450 functions without suitable data before any impacts have taken place. In our previous 3451 study (Ambrose and Lee 2004), we incorporated a method for assessing the net gain or 3452 loss of services, but quantitative, objective conclusions are difficult without appropriate 3453 “before” data. Conceptually, the correct way to answer this question is to assess wetland 3454 functions at the impact site before and after the impact occurs to estimate the loss of 3455 functions, and to assess functions at the mitigation site before and after mitigation occurs 3456 to estimate the gain of functions. These paired before-and-after functional assessments 3457 would provide the information necessary to assess a net change in wetland functions.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 83 3458 We recommend that functional assessments be conducted before the construction 3459 of any development project or mitigation project to establish the baseline conditions at 3460 those sites. Then, as part of the monitoring requirements, post-construction assessments 3461 should be conducted.
3462 There are a variety of methods that could be used for a functional assessment. 3463 Ideally, the State Board would adopt one particular method so the functional assessments 3464 were consistent across the state and could be easily compared and aggregated for a state3465 wide assessment. Some wetland evaluation methods, such as the Hydrogeomorphic 3466 Assessment Method (Hauer and Smith 1998), have been explicitly designed to 3467 incorporate no-net-loss analyses of mitigation projects. Others, such as the newly 3468 developed California Rapid Assessment Method (CRAM), which we used in our study, 3469 are readily used for this use. The method should be useable in a wide range of wetland 3470 habitats, quick to apply, and provide scientifically rigorous, objective data. 3471 Although paired before-after functional assessments are necessary for a careful 3472 assessment of net change in wetland function, they are rarely if ever undertaken. Besides 3473 the general difficulty of funding such studies, this particular study design carries the 3474 additional logistical difficulty that the “after” samples must be taken some years after the 3475 “before” sample. Despite these difficulties, we feel the paired before-and-after study 3476 design is needed to address the key policy question of whether compensatory mitigation 3477 under the Clean Water Act is accomplishing the goal of no net loss of wetland functions. 3478 There are additional benefits of before and/or after functional assessments, of 3479 course. A pre-construction functional assessment of the mitigation site would inform the 3480 design of the mitigation project, to help the analyst determine whether the proposed 3481 design is likely to result in the desired post-construction functions. A post-construction 3482 functional assessment of the mitigation site, such as we performed for this study, would 3483 show whether the mitigation project actually produced the desired functions. Even for 3484 these purposes, adoption of a standard functional assessment method such as CRAM 3485 would increase the value of the functional assessments by allowing the compilation of 3486 results across the state.
3487 6.5. Coordination with other agencies
3488 Although the Water Board has responsibility for 401 permits, the entire process of 3489 regulating impacts to wetlands and waters of the United States is closely coordinated with 3490 other agencies, especially the U.S. Army Corps of Engineers and the California 3491 Department of Fish and Game. Improved information management might improve this 3492 coordination.
3493 6.5.1. Improve incorporation of final permit information into Water Board 3494 files
3495 Although the 401 process is integral to wetland permitting, we found a significant 3496 number of files where changes to a project (impacts and/or mitigation) that occurred later 3497 in the project planning and permitting were not incorporated into Water Board files or 3498 401 permits (see Section 4.1.1). Our review of permit files suggests that the Regional Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 84 3499 Board staff have not always been included in the planning decisions that occurred after 3500 the 401 permit was issued. The Regional Boards should be active through all phases of 3501 the project planning or should at least insist on being copied on all subsequent changes 3502 that are approved by the other regulatory agencies. Once finalized, the 401 permit should 3503 be updated to reflect the actual impacts and mitigation actions/acreage that occurred, and 3504 then the database should be updated.
3505 Although our review focused on 401 permits and the information included in 3506 them, it is worth noting that 401 conditions should always be explicitly included in the 3507 404 permit.
3508 6.5.2. Consider developing an integrated permit
3509 Coordination with other agencies would be maximized if there was a single 3510 integrated permit required for projects impacting wetlands or waters of the U.S. Since 3511 there must already be significant coordination among the agencies, an integrated permit 3512 might not mean additional work, but it would simplify the permitting process for 3513 permittees, it would ensure that all relevant information was available and included in 3514 Water Board files, and it would eliminate redundant permit conditions. 3515 7. Recommended Compliance Monitoring Program
3516 The SWRCB contract for this work states that this final report shall “provide 3517 recommendations on the necessity, frequency, location, and type of ongoing compliance 3518 monitoring.” Section 7.1 discusses the need for compliance monitoring based on the 3519 results of the present study. The next section discusses whether compliance monitoring 3520 might be focused at particular locations, how often it might be needed, and what type of 3521 monitoring might be required. In addition, we have some specific recommendations 3522 (Section 7.3) concerning monitoring.
3523 Our recommendations about compliance monitoring reflect our own experiences, 3524 the scientific literature, and other guidelines. A particularly relevant guideline was 3525 produced in 2004 by the Los Angeles District of the Army Corps (LAD USACE 2004). 3526 Although directed more at monitoring the progress of mitigation projects, aspects of these 3527 guidelines are relevant to compliance monitoring.
3528 7.1. The need for compliance monitoring
3529 The results of this study clearly indicate the need to evaluate the compliance of 3530 mitigation projects with their permits. Thirteen of the 257 permits we located had to be 3531 excluded because of potential compliance issues. This indicates that up to 5% of the files 3532 we reviewed may have significant compliance problems (such as the impact occurring 3533 but no mitigation being undertaken).
3534 Our analysis of discrepancies between the 401 permit and information in the 3535 permit file identified additional compliance issues. For example, 8% of the 143 files we 3536 evaluated had information indicating that the actual impacts were greater than authorized Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 85 3537 in the 401 permit; overall, there appeared to be compliance issues with 42% of the files 3538 we evaluated.
3539 We found relatively high compliance with third-party mitigation requirements, 3540 but substantial lack of compliance with nearly every other category of permit conditions 3541 we assessed (see Table 7). Only about 65% of acreage requirements were met. Only 3542 about 50% of success criteria/performance standards were met. About 53% of 3543 monitoring and submission requirements were met. Moreover, many of the categories we 3544 assessed had a high fraction of permits for which the conditions could not be assessed; 3545 for example, we could not assess monitoring and submission conditions for more than 3546 half of the permits.
3547 These results indicate a definite need for compliance monitoring. Without a 3548 significant compliance effort, permittees are failing to comply with a wide range of 3549 permit conditions without the Water Board staff knowing about it. 3550 7.2. How should compliance monitoring efforts be focused? 3551 Our observations here are based on inferences gained from reviewing the permit 3552 files as well as data on compliance with permit conditions. Data from our analysis of 3553 compliance might be used to guide decisions about the most effective places to focus 3554 compliance monitoring. However, in considering this information, it is important to 3555 remember that ours was a retrospective analysis, sometimes assessing compliance many 3556 years after the mitigation project was completed, and as a consequence there were many 3557 permit conditions we could not assess. It is possible that there were compliance problems 3558 with the permit conditions that were not assessable for us, but we cannot determine that. 3559 A more complete assessment of compliance (enforcement) problems should focus on 3560 contemporary permits so that all conditions could be assessed. 3561 Our data allow us to identify some areas that seem most likely to have low 3562 compliance. For example, we found some differences in compliance for different types 3563 of permittee. The lowest 401 compliance scores were State/Federal and Municipal 3564 agencies. For mitigation plan compliance, Caltrans and private permittees (individual 3565 land owners or commercial entities with small “one-time” projects) joined these two as 3566 having the lowest compliance. Industry (corporation-owned factories, landfills, etc.) had 3567 the highest compliance scores for the mitigation plan compliance. 3568 We also found some regional differences in compliance. Among the different 3569 Water Board regions, Region 2 had relatively low 401 compliance and Region 8 had 3570 lower mitigation plan compliance. The low 401 compliance in Region 2 appears to be 3571 the result of higher expectations and more specific permit conditions in Region 2 3572 compared to other regions rather than the permittees in Region 2 being less diligent. For 3573 this reason, compliance numbers alone do not reflect the quality of the mitigation 3574 undertaken, since better compliance could be achieved by having fewer permit conditions 3575 and less demanding conditions. Among the Water Board regions, Regions 8 and 5F had 3576 among the fewest specific conditions in the 401 and among the highest proportion of 3577 redundant conditions.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 86 3578 The mean 401 compliance differed somewhat among the different wetland types 3579 (Figure 66). High gradient riverine habitats had the highest compliance rate. Low 3580 gradient riverine, depressional, and lagoon (the latter with only a single example) had 3581 intermediate compliance rates. Vernal pools (N=10) and estuarine wetlands (N=1) had 3582 the lowest compliance rates.
3583 Although the preceding results provide some guidance in terms of possible areas 3584 for focusing compliance assessments, in our view it does not provide a very sharp focus. 3585 Compliance issues are spread quite broadly across all aspects of the 401 program, so 3586 compliance monitoring will also need to be spread quite broadly. The areas identified as 3587 having lower compliance might warrant a particular emphasis during compliance 3588 monitoring, but compliance was not so high for most other areas (with the possible 3589 exception of third-party mitigation conditions) that it would be safe to assume high 3590 compliance with them.
3591 Although we have conducted a detailed assessment of compliance with 401 3592 permits, we have little direct knowledge of the State or Regional Boards’ current 3593 activities for checking compliance. Our review of information in the permit files suggest 3594 that there are substantial compliance issues for which there was no evidence of Regional 3595 Board response, but we did not follow up on these instances to determine if the Regional 3596 Boards were aware of those issues or had taken actions not evident in the file. Hence, we 3597 cannot comment on how current compliance efforts might be re-directed. However, we 3598 can identify mitigation monitoring reports as a cost-effective vehicle for evaluating a 3599 mitigation project.
3600 Although monitoring requirements were regularly included as 401 permit 3601 conditions, and evaluated for compliance when appropriate, the relative scarcity of 3602 monitoring reports in the permit files we reviewed suggest that compliance with the 3603 monitoring requirement is checked infrequently. Our compliance assessment indicated 3604 that conditions requiring mitigation monitoring were met only about 53% of the time; it 3605 was unclear whether any enforcement actions were undertaken in response to the absence 3606 of monitoring reports. While we were conducting our study for the Los Angeles 3607 Regional Board, that region was compiling lists of permit files without monitoring 3608 reports and contacting permittees to obtain the reports. This seems like a relatively cost3609 effective area on which to focus compliance monitoring efforts. 3610 In addition to reviewing submissions, it would be ideal if Water Board staff could 3611 undertake periodic site visits to confirm the reported monitoring results. However, we 3612 recognize that Water Board staff time is extremely limited, and it may not be feasible for 3613 existing staff to conduct site visits. Recommendation 7.3.2 suggests an organization that 3614 could undertake these site visits.
3615 7.2.1. Frequency of compliance monitoring
3616 There are different phases of a mitigation project, and different types of 3617 compliance monitoring would be required for each phase.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 87 3618 In the early construction phase of a mitigation project, many decisions are being 3619 made and many activities are being undertaken. Compliance monitoring during this 3620 phase would ensure that the mitigation project took shape as envisioned by the 401 staff 3621 and described in the mitigation plan. In addition, many compliance problems identified 3622 during this early phase are more likely to be resolved easily than if they were to be 3623 identified much later.
3624 The best type of compliance monitoring for the early phase would be on-site 3625 inspections. However, as noted above, it is unlikely that existing Regional Board staff 3626 would have the time to conduct on-site inspections, although perhaps this would be 3627 possible for the largest or most complicated projects. (If an independent monitoring 3628 cooperative was established, as recommended in Section 7.3.2, they could conduct some 3629 site inspections.) In the absence of on-site inspections, appropriate monitoring reports, 3630 required frequently during and immediately after construction, could document the 3631 progress of construction. If the permit conditions relating to construction were clearly 3632 established in the permit and/or monitoring plan, then these initial monitoring reports 3633 could focus their information on documenting that the permit conditions had been met. 3634 Extensive photographs would assist in documenting the progress of construction and 3635 compliance with the permit conditions.
3636 After the initial post-construction period, we would expect the mitigation site to 3637 change fairly rapidly as physical processes establish themselves and equilibrate to the 3638 system and plantings begin to grow. Fairly frequent documentation of these changes 3639 would allow Regional Board staff to confirm the appropriate development of the project. 3640 In the first year, quarterly or semi-annual reports would be useful. 3641 After the initial development of the mitigation site, we would expect changes to 3642 occur at a slower rate (e.g., Zedler and Callaway 2000). Annual monitoring would be 3643 appropriate. However, the second year of a mitigation project is a particularly critical 3644 time, so a particular focus on that period would be important. After two years, there has 3645 been time for the site to become established, so any deficiencies should begin to become 3646 apparent. It is important to identify potential problems early; it deficiencies are not 3647 identified until the end of the monitoring period, there will be limited opportunities for 3648 remediation.
3649 In general, on-site inspections would be the best way to confirm that all permit 3650 conditions had been met, but Regional Board staff should be able to assess compliance by 3651 careful review of monitoring reports. The most efficient use of staff resources would be 3652 to rely on annual monitoring reports through the end of the monitoring period, then 3653 confirm the report findings by an on-site inspection. As noted above, the second year is a 3654 particularly critical period, so an on-site inspection after the second year would also be 3655 useful. However, on-site visits are often not possible due to staffing constraints. Office 3656 review of the monitoring reports would be sufficient in most cases, as long as the 3657 monitoring reports were focused and informative. Because we feel that good monitoring 3658 reports are essential for an efficient evaluation of permit compliance, we have included a 3659 specific recommendation on this topic (Recommendation 7.3.1).

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 88 3660 7.3. Specific monitoring recommendations
3661 7.3.1. Mitigation monitoring reports should be streamlined and focused 3662 around demonstrating compliance with an established list of permit 3663 conditions.
3664 Mitigation monitoring reports tend to be large tedious documents that restate 3665 much of the background project-related information and only diffusely and often 3666 ambiguously address compliance related issues. These documents often include highly 3667 detailed descriptions of the monitoring methods and detailed results of vegetation 3668 monitoring data. Such information can be useful and should be documented, perhaps in 3669 quarterly reports, but annual monitoring reports should focus on the success-related 3670 issues and should clearly document compliance with an established list of permit 3671 conditions (see Recommendation 6.3.3). Because agency permit files are often 3672 incomplete and lack key documents (such as the mitigation plan), we do not feel that all 3673 background information (such as the restating of project impacts and expected mitigation 3674 strategies) should be eliminated from monitoring reports. However, such information 3675 should be well organized and succinct. The extraneous nature of existing monitoring 3676 report has been an impediment to the regulatory review of these documents. 3677 7.3.2. Form a multi-agency cooperative for compliance monitoring and project 3678 tracking.
3679 In California there are typically three to five regulatory agencies involved in the 3680 wetland regulatory process: the Corps, the Regional Board, the DFG if the project 3681 involves stream or lakebed impacts or State-listed endangered species, the FWS if there 3682 are federally-listed endangered species issues, and the Coastal Commission if the project 3683 occurs within the Coastal Zone. Each agency is responsible for independently 3684 monitoring compliance with its own permits, including compliance with compensatory 3685 mitigation requirements. Compliance monitoring is complicated by the fact that not all 3686 agencies receive all required documents (e.g., final mitigation plans, monitoring reports, 3687 deeds, proof of payment/credit purchases, and documents describing planning changes) 3688 from the permittee. Permittees frequently submit documents to a single agency that they 3689 view as the “lead” agency for their project.
3690 Following up on permit compliance includes the time consuming reorientation to 3691 the various projects, keeping track of document submissions and other communications, 3692 the careful review of mitigation monitoring reports, and site visits, plus maintaining the 3693 files and updating the database. Yet each agency suffers from perennial understaffing 3694 and limited resources. The result is that little monitoring of compliance is done by any 3695 agency.
3696 To help address this problem, we recommend that regulatory agencies establish a 3697 multi-agency cooperative to monitor compliance and track wetland losses and mitigation 3698 success across the State. This cooperative could report the results of its evaluation to 3699 each of the regulatory agencies and serve as a central repository for permit-related 3700 information. This could improve compliance monitoring and free-up staff resources.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 89 3701 Costs would be distributed and redundancy would be eliminated, thus maximizing the 3702 efficient use of limited resources.
3703 In our study, we reviewed 200-300 permit files and thoroughly assessed almost 3704 150 files within one year with a limited staff. With limited funding from each agency, a 3705 small staff could receive and manage copies of documents from across the state, visit a 3706 significant percentage of sites as agents of all agencies, and report their findings to each 3707 agency. After issuing their permits, project managers would be freer to concentrate on 3708 new projects instead of simultaneously tracking multiple existing projects. Such a 3709 cooperative would ensure that compliance monitoring would actually get accomplished, 3710 while avoiding substantial redundancy of effort and promoting the centralization of 3711 permit file information and tracking.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 90 3712 8. Literature Cited
3713 Allen, A.O. and J.J. Feddema. 1996. Wetland loss and substitution by the Section 404 3714 permit program in southern California, USA. Environmental Management 3715 20(2):263-274.
3716 Ambrose, R.F. 1995. Performance standards for coastal wetland restoration. In: M.C. 3717 Landin, ed. Proceedings of the National Interagency Workshop on Wetlands:
3718 Technology Advances for Wetlands Science. Technical Report, Wetlands Research 3719 and Technology Center, U.S. Army Engineers Waterways Experiment Station, 3720 Vicksburg, Mississippi. pp. 395-397.
3721 Ambrose, R.F. 2000. Wetland mitigation in the United States: Assessing the success of 3722 mitigation policies. Wetlands (Australia) 19: 1-27. 3723 Ambrose, R.F., S.F. Lee and S.P. Bergquist. 2003. Environmental Monitoring and 3724 Bioassessment of Coastal Watersheds in Ventura and Los Angeles Counties. Final 3725 Report to: Los Angeles Regional Water Quality Control Board. 3726 Ambrose, R.F. and S.F. Lee. 2004. An evaluation of compensatory mitigation projects 3727 permitted under Clean Water Act Section 401 by the Los Angeles Regional Quality 3728 Control Board, 1991-2002. Report prepared for the Los Angeles Regional Water 3729 Quality Control Board. University of California, Los Angeles, December, 2004. 3730 Ambrose, R.F. and S.F. Lee. 2004b. An evaluation of compensatory mitigation projects 3731 permitted under Clean Water Act Section 401 by the Los Angeles Regional Quality 3732 Control Board, 1991-2002. Guidance Document. Prepared for the Los Angeles 3733 Regional Water Quality Control Board. University of California, Los Angeles, 3734 December, 2004.
3735 Breaux, A. and F. Serefiddin. 1999. Validity of performance criteria and a tentative 3736 model for regulatory use in compensatory wetland mitigation permitting. 3737 Environmental Management 24(3):327-336.
3738 Breaux, A. and M. Martindale. 2003. Wetland Ecological and Compliance Assessments 3739 in the San Francisco Bay Region, California. Draft Final Report to the San 3740 Francisco Bay Regional Water Quality Control Board, California State Water 3741 Resources Control Board, California Coastal Conservancy, and US Army Corps of 3742 Engineers, San Francisco District. July 31, 2003.
3743 Breaux, A., S. Cochrane, J. Evens, M. Martindale, B. Pavlik, L. Suer, and D. Benner. 3744 2005. Wetland ecological and compliance assessments in the San Francisco Bay 3745 Region, California, USA. Journal of Environmental Management 74:217-237. 3746 Brinson, M.M. 1993. A Hydrogeomorphic Classification for Wetlands. U.S. Army 3747 Corps of Engineers, Waterways Experiment Station, Vicksburg, MS, USA. 3748 Technical Report WRP-DE-4.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 91 3749 Brown, P., and C.Lant. 1999. The effect of wetland mitigation banking on the 3750 achievement of no-net-loss. Environmental Management. 23(3): 333-345. 3751 Brown, S.C. and P.L.M. Veneman. 2001. Effectiveness of compensatory wetland 3752 mitigation in Massachusetts, USA. Wetlands 21:508-518. 3753 Collins, J.N., E.D. Stein, and M. Sutula. 2004. California Rapid Assessment Methods 3754 for Wetlands v2.0: User’s Manual and Scoring Forms. 3755 Collins, J.N., E.D. Stein, and M. Sutula. 2005. California Rapid Assessment Methods 3756 for Wetlands v3.0: User’s Manual and Scoring Forms. 3757 Craft, C., P. Megonigal, S. Broome, J. Stevenson, R. Freese, J. Cornell, L. Zheng, and J. 3758 Sacco. 2003. The Pace of Ecosystem Development of Constructed Spartina 3759 Alterniflora Marshes. Ecological Applications 13: 1417-1432. 3760 Dahl, T. E. 1990. Wetlands losses in the United States, 1780's to 1980's. U.S. Fish and 3761 Wildlife Service, Washington, DC.
3762 DeWeese, J. and C. Gould. 1994. An evaluation of selected wetland creation projects 3763 authorized through the Corps of Engineers Section 404 Program. US Department 3764 of the Interior, Fish and Wildlife Service, Sacramento Field Office, Sacramento, 3765 California, 90 pp.
3766 Gibbons, J.W. 2003. Terrestrial habitat: a vital component for herpetofauna of isolated 3767 wetlands. Wetlands 23: 630-635.
3768 Gilman, E.L. 1998. Nationwide permit program: unknown adverse impacts on the 3769 commonwealth of the Northern Mariana Islands’ wetlands. Coastal Management 3770 26:253-277.
3771 Gwin S.E., Kentula M.E. & Shaffer P.W. (1999) Evaluating the effects of wetland 3772 regulation through hydrogeomorphic classification and landscape profiles. 3773 Wetlands 19, 477-489.
3774 Hauer, F.R. and R.D. Smith. 1998. The hydrogeomorphic approach to functional 3775 assessment of riparian wetlands: evaluating impacts and mitigation on river 3776 floodplains in the U.S.A. Freshwater Biology 40: 517-530. 3777 Hawke, D.F. 1989. Everyday life in early America. HarperCollins, New York, USA. 3778 Holland, C.C., and M.E. Kentula. 1992. Impacts of Section 404 permits requiring 3779 compensatory mitigation on wetlands in California. Wetlands Ecology and 3780 Management 2:157-69.
3781 Kelly, N.M. 2001. Changes to the landscape patterns of coastal North Carolina wetlands 3782 under the Clean Water Act. Landscape Ecology 16:3-16.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 92 3783 Kentula, M.E., J.C. Sifneos, J.W. Good, M. Rylko, and K. Kunz. 1992. Trends and 3784 patterns in Section 404 permitting requiring compensatory mitigation in Oregon and 3785 Washington, USA. Environmental Management 16:109-19. 3786 Knighton, D. 1998. Fluvial forms and processes: a new perspective. Oxford University 3787 Press, England.
3788 Leibowitz, S.G. 2003. Isolated wetlands and their functions: an ecological perspective. 3789 Wetlands 23: 517-531.
3790 State Water Resources Control Board [SWRCB]. California Environmental Protection 3791 Agency. http://www.swrcb.ca.gov/
3792 Los Angeles District, U.S. Army Corps of Engineers. 2004. Final Mitigation Guidelines 3793 and Monitoring Requirements. Public Notice No.: 970031200-RRS. April 19, 3794 2004.
3795 Lyons, J., S.W. Trimble, and L.K. Paine. 2000. Grass versus trees: managing riparian 3796 areas to benefit streams of central North America. Journal of the American Water 3797 Resources Association 36: 919-930.
3798 Miller, J. 1995. Man-made wetlands’ future starts to dry up. Los Angeles Times, East 3799 Ventura County Edition, 9 July 1995, p. B1.
3800 Miller, R.E., Jr. and B.E. Gunsalus. 1997. Florida Wetland Rapid Assessment 3801 Procedure. Updated 2nd edition. Technical Publication REG-001. Natural 3802 Resource Management District, West Palm Beach, FL. 3803 Mitsch, W.J. and R.F. Wilson. 1996. Improving the success of wetland creation and 3804 restoration with know-how, time, and self-design. Ecological Applications 6:77-83. 3805 National Research Council [NRC]. 1995. Wetlands: characteristics and boundaries. 3806 National Research Committee on Mitigating Wetland Losses. National Research 3807 Committee on Characterization of Wetlands. National Academy Press, Washington 3808 DC, USA.
3809 National Research Council [NRC]. 2001. Compensating for wetland losses under the 3810 Clean Water Act. National Research Committee on Mitigating Wetland Losses. 3811 National Academy Press, Washington DC, USA.
3812 National Research Council [NRC]. 2002. Riparian areas: functions and strategies for 3813 management. National Research Committee on Riparian Zone Functioning and 3814 Strategies for Management. National Academy Press, Washington DC, USA. 3815 Race, M.S. 1985. Critique of present wetlands mitigation policies in the United States 3816 based on an analysis of past restoration projects in San Francisco Bay. 3817 Environmental Management 9:71-82.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 93 3818 Race, M.S. and M.S. Fonseca. 1996. Fixing compensatory mitigation: what will it take? 3819 Ecological Applications 6:94-101.
3820 Rundel, P.W. 2002. Preserving Biodiversity. Southern California Environmental Report 3821 Card 2002. UCLA Institute of the Environment, Los Angeles:4-13. 3822 Quigley M. K. Ranke, and D. Miller. 2006. Evaluation of Federal Clean Water Act 3823 Section 401 Certification Wetland and Stream Mitigation Sites in the Santa 3824 Margarita Watershed. Final report by the State Water Resources Control Board, 3825 San Diego Region. January 2006.
3826 Sifneos, J.C., E.W. Cake, J.R., and M.E. Kentula. 1992a. Effects of Section 404 3827 Permitting on Freshwater Wetlands in Louisiana, Alabama, and Mississippi. 3828 Wetlands 12(1):28-36.
3829 Sifneos, Jean C., Mary E. Kentula, and Paul Price. 1992b. Impacts of Section 404 Permits 3830 Requiring Compensatory Mitigation of Freshwater Wetlands in Texas and 3831 Arkansas. The Texas Journal of Science 44(4):475-485. 3832 Sudol, M.F. 1996. Success of Riparian Mitigation as Compensation for Impacts due to 3833 Permits Issued Through Section 404 of the Clean Water Act in Orange County, 3834 California. D.Env. Dissertation, University of California, Los Angeles. 3835 Sudol, M.F., and R.F. Ambrose. 2002. The U.S. Clean Water Act and habitat 3836 replacement: evaluation of mitigation sites in Orange County, California, USA. 3837 Environmental Management 30: 727-734.
3838 Zedler, J.B. 1996. Tidal wetland restoration: a scientific perspective and southern 3839 California focus. Report T-038. California Sea Grant College System, University of 3840 California, La Jolla, California.
3841 Zedler, J.B. and J.C. Callaway. 2000. Evaluating the progress of engineered tidal 3842 wetlands. Ecological Engineering 15: 211-225.
3843 Zentner, J. J. 1988. Wetland restoration success in coastal California. Increasing our 3844 wetland resource: Proceedings of a Conference. Eds. J. Zelazny, and J. S. 3845 Feierabend, Washington, D.C.: National Wildlife Federation. pp. 216-219. 3846
Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 94 9. Tables
Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 95 Table 1. Reference Site information
SiteID Name Region Latitude Longitude
Research
Group*
Wetland
Type
WCAP99-R026 Coldwater Creek 1 41.84611 124.02750 CCG Riverine Low WCAP99-R029 Clark's Creek 1 41.80861 124.11667 CCG Riverine High WCAP99-RO92 Prairie Creek State Park 1 41.40000 124.05806 CCG Riverine High BC-Y Blue Creek 1 41.20000 123.54000 CCG Riverine High
WCAP99-R037 Horse Linto 1 41.00893 123.60197 CCG Riverine High 11921 Grove's Prarie 1 40.95667 123.48528 CCG Riverine Low WCAP99-R077 Canoe creek 1 40.29490 123.90290 CCG Riverine Low FREE 11130 Freeman Meadow 5R 39.67333 120.62075 SFEI Riverine Low WCAP99-R003 Trout Creek 1 39.53852 122.86077 SFEI Riverine High WCAP99-R008 Rattlesnake Creek 1 39.49388 122.86368 SFEI Riverine High WCAP99-0614 Austin Creek East 1 38.53603 123.07221 SFEI Riverine Low Ref. 16 Ashbury Creek Lo 1 38.35028 122.53793 UCLA Riverine Low Ref. 17 Ashbury Creek Tributary 1 38.34976 122.53352 UCLA Riverine High CA02-0604 Upper Petaluma 2 38.20767 122.56683 SFEI Estuarine CA02-0608 Point Edith 2 38.04353 122.07233 SFEI Estuarine CA02-0612 China Camp 2 38.01475 122.49280 SFEI Estuarine Ref. 22 Briones Regional Park 2 37.92129 122.16454 USF Riverine High Ref. 5 Walker 6V 37.90109 119.12983 UCLA Riverine Low
Ref. 4 McGill Trail Head 6V 37.54992 118.80384 UCLA Riverine High Ref. 3 Fish Slough 6V 37.48043 118.40321 UCLA Seep & Spring Ref. 9 TNC Vernal Pool Reserve 5F 37.39987 120.45229 UCLA Vernal Pool Ref. 10 Chowchilla 5F 37.17623 120.07051 UCLA Riverine Low 101 Upper Scott's Creek 3 37.07404 122.23793 CCG Riverine Low 106(a) East of Seal Bend 3 36.82000 121.77000 CCG Estuarine 12339 Carmel Valley River 3 36.52243 121.81748 CCG Riverine Low 12330 San Antonio River 3 35.89417 121.07361 CCG Riverine Low 310-ADC Arroyo de la Cruz Creek 3 35.70833 121.30035 CCG Riverine Low 310-SSU Upper San Simeon creek 3 35.60921 121.07393 CCG Riverine Low 310-SSC Lower San Simeon creek 3 35.59448 120.12112 CCG Riverine Low CA02-0031 Chorro Creek, marina 3 35.34553 120.83629 CCG Estuarine CA02-0021 Chorro Creek, flats 3 35.34430 120.83168 CCG Estuarine CA02-0002 Los Osos creek 3 35.33418 120.83638 CCG Estuarine Ref. 12 Coon Creek 3 35.25498 120.88692 UCLA Riverine Low 310-COO Coon creek 3 35.25476 120.88549 CCG Riverine Low Ref. 1 Pismo Beach Ecological Reserve 3 35.13359 120.62396 UCLA Lacustrine Ref. 15 Sedwick Reserve 3 34.73013 120.02692 UCLA Depressional Ref. 13 Sedwick Reserve 3 34.72113 120.03613 UCLA Riverine Low Ref. 14 Sedwick Reserve 3 34.68298 120.04469 UCLA Vernal Pool Ref. 2 Los Padres National Forest 4 34.51467 119.26867 UCLA Riverine Low Ref. 20 Arroyo Hondo Canyon 3 34.48702 120.14222 UCLA Riverine Low Ref. 21 El Capitan Canyon 3 34.48049 120.01888 UCLA Riverine High Ref. 18 Santa Paula Creek 4 34.44172 119.07551 UCLA Riverine Low Ref. 11 Upper Santa Clara River 4 34.44020 118.31349 UCLA Riverine Low Ref. 7 City Creek Rte 330 8 34.17385 117.18515 UCLA Riverine High Ref. 19 Solstice Cyn 4 34.03935 118.75321 UCLA Riverine Low Ref. 8 Upper Santa Margarita River 9 33.40826 117.23828 UCLA Riverine Low Ref. 6 Cibola Lake (NWR) 7 33.22461 114.67300 UCLA Lacustrine * CCG = Central Coast Group
Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 96 Table 2. Jurisdictional habitat hierarchy.

Every mitigation site was apportioned into its component habitat types according to this hierarchy. First, the evaluator determined which proportion of the sites consisted of “waters” and which proportion was outside of waters (e.g. 60:40). Next, the wetland and non-wetland waters percentages would be determined (e.g. 50:10), as would any non-waters riparian and upland habitats (e.g. 20:20), and so forth. The sum of the equivalent habitat percentages would equal the above percentage in the hierarchy. These percentages were multiplied by the overall site acreage to determine the individual jurisdictional habitat acreages.

Waters of the United States
Wetland
Non-Wetland Waters
Non-Streambed Open Water
Streambed
Open Water Stream
Unvegetated Streambed
Vegetated Streambed
Riparian Waters
Non-Specified Riparian
Non-waters of the United States
Non-waters Riparian
Upland
Table 3. Overall summary of the permit file selection results by region.

This table includes the 429 permit files that were randomly selected from the SWRCB database, and pursued at either the Corps or Regional Board offices, or both. Two files were initially pursued, but later excluded because they had 401 permits that were issued directly by the State Board (SB).

Region Pursued for
review
Not
located
Removed
during
review
Removed
after field
visit
Not visited
or assessed
Assessed for
compliance only
Assessed
fully
1 32 15 5 0 1 2 9
2 75 29 20 0 0 1 25
3 43 16 4 7 1 2 13
4 44 6 10 9 0 4 15
5F 18 10 0 2 0 2 4
5R 27 17 2 0 2 0 6
5S 54 13 10 2 4 1 24
6T 23 14 4 1 2 0 2
6V 10 4 2 2 0 0 2
7 11 7 1 0 0 1 2
8 25 7 3 2 0 0 13
9 65 33 12 5 0 1 14
SB 2 1 1 0 0 0 0
Total 429 172 74 30 10 14 129
Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 97 Table 4. Number of onsite and offsite mitigation sites for file specific mitigation actions, formal mitigation banks, informal mitigation banks, and in lieu fees.

N FileSpecific
Formal
Mitigation Bank
Informal
Mitigation
Bank
In-Lieu Fee
On Site Mitigation 127 125 1 1 0
Off Site Mitigation 77 29 31 14 3
Total 204 154 32 15 3
Table 5. Summary of the discrepancies between the impact and required mitigation acreage values obtained through our detailed permit reviews and the corresponding values in the State Board’s permit tracking database. Multiple discrepancy categories may apply to a particular file.

Source of Impact and/or Mitigation Acreage Discrepancy Number of Files
% of Total Files
(N=143)
Discrepancy due to minor rounding issues in 401 permit or in SWRCB database 9 6.2 Data entry issue in SWRCB database (typographical error or misinterpretation of information in 401 permit, often due to ambiguous wording). 26 18.2 Issues with the 401 permit itself, including transcriptional and typographical errors, misinterpretations, or a lack of critical information in the 401 permit text 24 16.8 Discrepancy due to accounting difference (e.g., permanent vs. temporary impacts, or wetlands vs. non-wetland waters) between reported values and 401 permit 27 18.9 Other agency required more mitigation than RB, but 401 permit not outdated 19 13.4 Mitigation planning modified after 401 permit issuance, permit outdated 12 8.4 Impacts reduced after 401 issuance, mitigation same, 401 permit outdated 3 2.1 Impacts reduced after 401 issuance, mitigation different, 401 permit outdated 13 9.1 401 outdated, impacts greater than 401 approved, mitigation same or different 12 8.4 Revised 401 permit entered separately into SWRCB database resulting in multiple entries and redundant acreage values 7 5.0
Summaries
Discrepancies between reported values and the SWRCB database 101 70.6 Discrepancies between our reported values and the 401 permits themselves 86 60.1 Regulatory/compliance issues with files from an acreage perspective 60 42.0 Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 98 Table 6. Summary of compliance scores based on 401 and mitigation plan evaluations including average scores and scores for the percentage of conditions met to 100% satisfaction.

Successful included files with compliance scores greater than 75%, partially successful included files with scores between 25% and 75%, and failure included files with scores less than 25%.

N Score Successful Partially Successful Failure
Average 401 84.3% 76% 20% 4%
Average 401 percent-met 124 73.3% 57% 40% 13%
Average mitigation-plan 80.7% 68% 32% 0%
Average mitigation plan percent-met 81 67.6% 48% 35% 6% Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 99 Table 7. Compliance breakdowns for 401 and Mitigation Plan compliance grouped by compliance condition category (N=143 files). See Methods for details on condition categories. 401 Mitigation Plan
Condition Code
Condition Category
Total # Conditions
Average # Conditions
Average # ND
Average Score
Average % Met
Average % ND
Total # Conditions
Average # Conditions
Average # ND
Average Score
Average % Met
Average % ND
1 Third Party 58 1.5 0.1 99.3 99.3 8.8 26 1.6 0.1 90.0 90.0 6.3 2 Acreage 158 1.8 0.2 81.5 64.4 6.9 132 2.0 0.2 83.0 66.8 9.5 3 Site Implementation 411 6.0 2.7 84.8 71.9 45.1 546 7.9 3.1 84.3 72.4 40.44 Site Maintenance 49 1.6 0.8 76.0 56.7 45.6 93 2.2 0.7 80.7 68.1 34.35 Site Protection 66 1.5 0.6 81.3 72.6 42.5 58 1.6 0.4 77.9 72.4 25.66 Success & Performance
Standards 199 3.9 1.5 76.4 49.7 31.0 298 4.4 1.3 76.0 52.9 26.3 7 Monitoring & Submission 254 3.6 2.0 59.5 52.3 54.3 220 3.2 1.4 60.9 53.7 45.78 Invocation of Other Agency
Permits 126 1.7 1.1 N/A N/A 69.3 5 2.5 1.0 N/A N/A 100
9 Other 35 1.3 0.6 96.1 94.4 46.8 13 1.3 0.3 93.8 93.8 20.0 3 - 6
Site Implementation,
Maintenance, Protection,
Success/Performance Standards 725 3.2 1.4 79.6 62.7 41.0 995 4.0 1.4 79.7 66.4 31.6 Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 100
Table 8. Summary statistics of mitigation CRAM scores (N=129) and reference site CRAM scores (N=47) for Total-CRAM scores and the four attributes, along with the percentage of files within each success category.

Reference Sites Filewide CRAM Scores
Median Mean ± SE Median Mean ± SE Optimal Sub Optimal Marginal to Poor
Overall 82.06 79.13 ± 1.36 60.77 58.61 ± 1.10 19.38 56.59 24.03 Landscape Context 90.28 87.10 ± 1.06 72.32 65.57 ± 1.78 47.29 24.81 27.91 Hydrology 90.74 86.67 ± 1.58 62.96 62.67 ± 1.64 27.13 42.64 30.23 Physical Structure 79.17 76.06 ± 2.48 52.79 53.81 ± 1.61 49.61 27.13 23.26 Biotic Structure 68.33 66.68 ± 2.24 51.78 52.63 ± 1.28 62.02 25.58 12.40 Table 9. Summary statistics and success breakdowns of Total-CRAM scores by SWRCB region (N=129 files).

Total-CRAM Scores (Overall Filewide CRAM Scores)
Region N Mean ± SE Median % Optimal
% SubOptimal % Marginal
/ Poor
1 9 57.12 ± 4.76 50.93 22.22 55.56 22.22
2 25 51.08 ± 2.07 48.40 4.00 44.00 52.00
3 13 55.61 ± 3.81 58.74 15.38 61.54 23.08
4 15 57.67 ± 3.40 57.99 20.00 46.67 33.33
5F 4 61.73 ± 5.26 64.86 25.00 50.00 25.00
5R 6 61.57 ± 2.98 61.33 16.67 83.33 0.00
5S 24 64.40 ± 1.43 64.33 16.67 79.17 4.17
6T 2 74.43 ± 3.83 74.43 100.00 0.00 0.00
6V 2 42.52 ± 14.4 42.52 0.00 50.00 50.00
7 2 56.22 ± 8.17 56.22 0.00 50.00 50.00
8 13 64.25 ± 2.79 67.50 23.08 69.23 7.69
9 14 60.44 ± 4.38 65.63 42.86 35.71 21.43
Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 101
Table 10. Summary statistics and success breakdowns of landscape context metrics CRAM scores by SWRCB region (N=129 files).

Landscape Context CRAM Scores
Region N Mean ± SE Median % Optimal
% SubOptimal % Marginal
/ Poor
1 9 55.43 ± 6.60 50.86 22.22 22.22 55.56
2 25 57.84 ± 3.80 57.33 28.00 32.00 40.00
3 13 57.52 ± 6.86 53.30 38.46 15.38 46.15
4 15 64.75 ± 3.79 64.25 33.33 40.00 26.67
5F 4 68.40 ± 14.20 81.78 75.00 0.00 25.00
5R 6 76.92 ± 2.90 74.91 66.67 33.33 0.00
5S 24 82.55 ± 1.95 86.65 83.33 16.67 0.00
6T 2 84.44 ± 3.70 84.44 100.00 0.00 0.00
6V 2 34.97 ± 9.30 34.97 0.00 0.00 100.00
7 2 81.83 ± 4.08 81.83 100.00 0.00 0.00
8 13 61.88 ± 5.64 62.69 38.46 30.77 30.77
9 14 62.29 ± 5.50 70.49 42.86 28.57 28.57
Table 11. Summary statistics and success breakdowns of hydrology metrics CRAM scores by SWRCB region (N=129 files).

Hydrology CRAM Scores
Region N Mean ± SE Median % Optimal
% SubOptimal % Marginal
/ Poor
1 9 65.90 ± 7.77 52.50 44.44 0.00 55.56
2 25 61.39 ± 3.84 58.71 28.00 40.00 32.00
3 13 58.20 ± 5.11 64.82 0.00 76.92 23.08
4 15 59.15 ± 4.66 54.63 20.00 40.00 40.00
5F 4 71.79 ± 9.11 74.58 50.00 25.00 25.00
5R 6 73.00 ± 4.66 72.87 50.00 50.00 0.00
5S 24 62.65 ± 4.15 65.16 29.17 37.50 33.33
6T 2 81.20 ± 1.20 81.20 100.00 0.00 0.00
6V 2 35.51 ± 16.3 35.51 0.00 0.00 100.00
7 2 63.75 ± 27.90 63.75 50.00 0.00 50.00
8 13 63.58 ± 4.37 60.83 30.77 38.46 30.77
9 14 64.04 ± 3.79 64.27 14.29 78.57 7.14
Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 102
Table 12. Summary statistics and success breakdowns of physical structure metrics CRAM scores by SWRCB region (N=129 files).

Physical Structure CRAM Scores
Region N Mean ± SE Median Optimal
% SubOptimal % Marginal
/ Poor
1 9 52.90 ± 4.95 50.00 44.44 33.33 22.22
2 25 40.44 ± 3.52 39.83 24.00 28.00 48.00
3 13 55.55 ± 4.81 58.33 61.54 15.38 23.08
4 15 58.87 ± 5.29 66.67 60.00 26.67 13.33
5F 4 47.18 ± 7.58 45.42 25.00 50.00 25.00
5R 6 50.90 ± 5.32 47.23 33.33 50.00 16.67
5S 24 55.17 ± 2.68 59.56 58.33 25.00 16.67
6T 2 68.75 ± 18.8 68.75 50.00 50.00 0.00
6V 2 52.08 ± 2.08 52.08 50.00 50.00 0.00
7 2 50.69 ± 0.69 50.69 0.00 100.00 0.00
8 13 67.40 ± 3.73 70.83 76.92 23.08 0.00
9 14 57.99 ± 6.49 65.98 57.14 7.14 35.71
Table 13. Summary statistics and success breakdowns of biotic structure metrics CRAM scores by SWRCB region (N=129 files).

Biotic Structure CRAM Scores
Region N Mean ± SE Median Optimal
% SubOptimal % Marginal
/ Poor
1 9 54.24 ± 4.91 54.85 66.67 22.22 11.11
2 25 44.66 ± 2.36 45.00 40.00 36.00 24.00
3 13 51.18 ± 3.39 48.33 61.54 23.08 15.38
4 15 47.89 ± 2.82 45.23 40.00 53.33 6.67
5F 4 59.57 ± 5.32 60.07 75.00 25.00 0.00
5R 6 45.46 ± 4.29 44.55 50.00 33.33 16.67
5S 24 57.23 ± 1.89 60.07 83.33 16.67 0.00
6T 2 63.33 ± 8.33 63.33 100.00 0.00 0.00
6V 2 47.50 ± 30.00 47.50 50.00 0.00 50.00
7 2 28.61 ± 1.39 28.61 0.00 0.00 100.00
8 13 64.14 ± 3.53 65.00 84.62 15.38 0.00
9 14 57.43 ± 5.35 56.04 71.43 14.29 14.29
Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 103
Table 14. Summary statistics and success breakdowns of CRAM scores by individual CRAM metric (N=204 mitigation sites).

Metric N Mean ± SE Median
Buffer and Landscape Context
Connectivity 204 68.2 ± 1.8 77.8
% of AA with Buffer 204 81.6 ± 1.4 91.7
Avg. Width of Buffer 204 61.9 ± 1.9 66.7
Buffer Condition 204 60.6 ± 1.4 66.7
Hydrology
Water Source 204 59.5 ± 1.5 58.3
Hydroperiod 204 64.7 ± 2.0 73.3
Hydrologic Connectivity 117 64.6 ± 2.0 66.7
Physical Structure
Physical Patch Richness 204 43.5 ± 1.8 41.7
Topographic Complexity 204 63.5 ± 1.4 66.7
Organic Matter Accumulation 204 69.3 ± 1.4 68.9
Biotic Structure
Biotic Patch Richness 204 45.7 ± 1.4 41.7
Vertical Biotic Structure 190 39.1 ± 1.5 41.7
Interspersion / Zonation 204 58.6 ± 1.5 58.3
% Non-native Plant Species 204 60.5 ± 2.3 52.8
Native Plant Species Richness 204 49.3 ± 2.0 41.7
Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 104
Table 15. Total impacted and obtained acreage for all files (overall), waters of U.S. and Non waters of U.S., wetland, and non wetland waters.

Overall acreage includes waters of the U.S. plus non-waters areas. The breakdown for wetlands/non-wetland waters does not include 5 permit files for which the jurisdictional impacts could not be distinguished.

Table 16. Permanent impacts and created mitigation acreage, waters of U.S. and non waters of U.S., and wetland, non wetland waters.

Permanent
Impact Created Acreage Proportion
Obtained Net Acreage Gain Gained /Loss
Ratio
Overall Acreage 165.8 270.9 NA 105.1 1.6

Waters of U.S. 162.7 223.1 82.4 60.4 1.4
Non Waters of U.S. 3 47.8 17.6 44.8 NA

Waters of U.S.:

Wetlands 106.3 146.7 66.4 40.4 1.4
Non Wetland Waters 54.9 74.2 33.6 19.3 1.4
Total Impact Total Obtained Proportion Obtained Net Acreage Gain Gained /Loss Ratio Overall Acreage 216.8 417.0 NA 200.2 1.9
Waters of U.S. 212.4 303.2 72.7 90.8 1.4
Non Waters of U.S. 4.4 113.8 27.3 109.4 NA
Waters of U.S.:

Wetlands 121.2 180.5 63.2 59.3 1.5
Non Wetland Waters 74.5 105.2 36.8 30.7 1.4
Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 105
Table 17. Total impacted and obtained acreage for all files (overall), waters of U.S. and Non waters of U.S., wetland, and non wetland waters.
% Files with
Gains
% Files where
Gained = Lost
% Files with
Losses
Overall Acreage 64 17 20

Waters of U.S. 54 13 33
Non Waters of U.S. 45 55 0
Wetlands 58 19 22
Non Wetland Waters 24 34 42
Table 18. Permanent impacts and created mitigation acreage, waters of U.S. and non waters of U.S., and wetland, non wetland waters.
% Files w/Gains % Files
Gained=Lost % Files w/Loss
Overall Acreage 41 20 39

Waters of U.S. 36 17 47
Non Waters of U.S. 24 76 1
Wetlands 40 32 28
Non Wetland Waters 17 37 46
Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 106
Table 19. Mitigation success by permit file for each evaluation category: acreage requirement, 401 conditions, mitigation plan conditions, and wetland condition.

Data shown for acreage and compliance are percentages out of a total number of 143 permit files. Wetland condition data are percentages of a total number of 129 files. Numbers in parentheses are the actual number of sites within each category. For the acreage requirements, success was considered 100 percent, partial success was considered 75 to 100 percent (lower and upper bounds not inclusive), and failure was 75 percent and below. For the 401 and MP compliance evaluation, success was considered 75 to 100 percent, partial success was considered 25 to 75 percent (lower and upper bounds not inclusive), and failure was 25 percent and below. For the CRAM evaluation of wetland condition, success was considered 70 to 100 percent, partial success was 50 to 70 percent (lower and upper bounds not inclusive), and failure was 50 percent and below.

Category Percent
Success (N)
Percent
Partial
Success (N)
Percent
Failure (N)
Cannot Be
Determined
(N)
Acreage Requirement 72 (101) 11 (16) 17 (24) (2)
401 Conditions 76 (94) 20 (25) 4 (5) (19)
Mitigation Plan
Conditions 68 (55) 32 (26) 0 (0) (62)
Wetland Condition 19 (25) 55 (71) 26 (33) Not a category Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 107
Table 20. Acreage, compliance, and CRAM summaries by permittee type. These permittee type categories were taken directly from the 401 Permit Files.

Developer
Industry Caltrans
Municipal
Private
State/Federal
Number of Files 66 9 13 34 13 8
Average Impact Acreage
(Total Impact Acreage)

1.17 (76.96)

1.73 (15.54)

2.35 (30.55)

1.75 (59.55)

0.63 (8.19)

3.26 (26.05)
Average Required Acreage for Mitigation
(Total Required Acreage)

2.30 (151.80)

7.12 (64.11)

5.22 (67.80)

2.36 (80.30)

0.97 (12.65)

8.57 (68.59)
Average Obtained Acreage
(Total Obtained Acreage)

2.15 (141.75)

6.44 (57.95)

4.79 (62.25)

2.28 (77.63)

0.83 (10.84)

8.33 (66.60)
Average Acreage Gained
(Total Acreage Gained)

0.98 (64.80)

4.71 (42.41)

2.44 (31.71)

0.53 (18.08)

0.20 (2.66)

5.07 (40.55)
Mitigation Ratio (Required) 3.22:1 16.91:1 1.51:1 2.32:1 1.67:1 1.63:1 Mitigation Ratio (Obtained) 3.13:1 17.36:1 1.38:1 2.40:1 1.89:1 1.33:1 Average 401 Compliance Score 85.93 84.06 87.60 79.77 87.87 76.20 Average Mitigation Plan Compliance Score 81.70 89.96 73.94 80.56 76.98 79.20 Average Total-CRAM Score 57.42 56.71 61.24 59.81 58.03 63.53 Average CRAM-Adjusted Acreage
(Total CRAM-Adjusted Acreage)

1.35 (81.18)

3.55 (31.91)

3.58 (35.79)

1.24 (38.38)

0.44 (4.82)

4.09 (32.71)
Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 108
Table 21. Summary of administrative and regulatory recommendations. Improving mitigation requirements Information management Improve permit clarity Assessment of “no net loss” Coordination with other agencies Permit conditions should ensure complete compensation for the full suite of wetland functions and services lost
X
Ensure that mitigation projects compensate for losses in water quality (pollution) improvement services
X
There should be a better accounting of the habitat types lost and gained
X
Mitigation projects should have appropriate landscape context X Offsite mitigation should be within the same catchment, or at least the same watershed
X
Improvements to Database X
Improve permit archiving X
Improve tracking the progress of mitigation projects X
Important permit information should be clearly delineated in tables
X
Permit conditions should be written so that the extent of efforts must match the intent of the condition to be in compliance X
Every mitigation plan and permit should include a table of requirements upon which compliance will be judged
X
Permits should be clear about the meaning of enhancement, restoration and creation
X
Performance standards should be clear about the goal of invasive species control
X
Proof of inundation or saturation appropriate for wetland development should be required for mitigation wetlands
X
Pre- and post-construction functional assessments of impact and mitigation sites should be required
X
Improve incorporation of final permit information into Water Board files
X
Consider developing an integrated permit X
Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 109
Table 22. Suggested jurisdictional and non-jurisdictional habitat hierarchy, with structure for tracking losses and gains.

Impacted Required
Impact/Mitigation Acreage Accounting
Total Permanent Temporary Total Creation Restoration
Habitat Enhancement
Preservation
Waters of the United States. Wetland (Total) Riverine Estuarine/Lagoon Seasonal/Depressional Vernal Pool Seep/Spring/Wet Meadow Lacustrine Fringe Other Non-Wetland Waters Non-Streambed Open Water Streambed (Total) Open Water Unvegetated Streambed Vegetated Streambed Other (Ex: Riparian Waters)
Non-waters of the United States. Riparian Upland
Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 110

10.Figures
Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 111
618
6%
557
6%
1872
19%
237
2% 82
1%
137
1%
2118
21%
952
10%
1199
12%
236
2%
807
8% 1088
11%
1 5R
5S
5F 6V
7 9
8 2
6T
3 4
Figure 1. Map of California state board regions with breakdown of number of permit files.

The total number of files listed in the SWRCB database by region from 1991-2002 (N=9924 files) and the percentage of files by region of the total number of files in the SWRCB database from 1991-2002.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 112
Figure 2. Statewide distribution of reference sites.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 113
Figure 3. Statewide distribution of the assessed mitigation sites associated with the 143 permit files.

Several of these sites, especially those in the central valley (Region 5) involved a collection of shared mitigation banks which resulted in fewer than 143 mitigation sites. Points represent each assessed mitigation site rather than multiple sites per file.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 114
Region
1 2 3 4 5F 5R 5S 6T 6V 7 8 9
Number of Files
0 5
10 15
20 25
30 Assessed fully
Compliance only
Figure 4. Files assessed fully and for compliance only by state board region.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 115
Figure 5. Statewide distribution of the impact and mitigation sites associated with the 143 permit files assessed.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 116
Certification Year
1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 Percentage of Files
0 5
10 15
20 SWRCB - All Files
SWRCB - 143 Assessed Files
Figure 6. Percentage of applications per certification year listed in the SWRCB database from 1991 to 2002 compared with the percentage of files per year in our sample of files assessed fully and for compliance only (N for files assessed=143, N for SWRCB database=9924).

Permittee Type
Caltrans Developer Industry Municipal Private State/Federal Percentage of Files
0 10
20 30
40 50
N= 13 66 9 34 13 8
Figure 7. Percentage of files assessed by permittee type (N=143 files).

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 117
Number of Files
0 10 20 30 40 50 60 70
Habitat Type Impacted
Wetland
Riparian
Streambed
Wetland and Riparian
Wetland and Streambed
Wetland and Lake or Ocean
Riparian and Streambed
Wetland, Riparian, and Streambed
Wetland, Riparian, and Lake or Ocean
Unspecified Waters of the US
No impacts specified
Information from SWRCB database
Information from Project Analyses
Figure 8. Breakdown of the 143 assessed files by habitat type impacted as reflected by the SWRCB database, and by our detailed permit reviews.

Some files had impacts to a single habitat type while others impacted multiple habitat types. The individual wetland types are not included here as such information is not consistently available in the SWRCB database.

Number of Files
0 10 20 30 40 50 60 70
Acreage
0 >0-0.001
>0.001-0.01
>0.01-0.1
>0.1-1
>1-10
>10-100
Impacted
Required
Figure 9. Acres impacted and acres of mitigation required displayed by acreage-size categories using data from project analyses for files assessed (N=143).

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 118
Impact Type
Temporary Permanent
Acreage
0 20
40 60
80 100
120
140
160
180
Figure 10. Breakdown of the 143 assessed permit files by permanent and temporary impacts as reflected by the SWRCB database, and by our detailed permit reviews.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 119
Certification Year
1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Acreage
0 20
40 60
80 100
Impacted
Required
Figure 11. Acres impacted and acres of mitigation required displayed by certification year from the project analyses for files assessed (N=143).

Certification year
1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Mean Mitigation Ratio (Required:Impacted)
0 5
10 15
20 25
30 N= 22 5 10 17 25 19 16 18 15 10 1
Figure 12. Average mitigation ratios required by certification year as determined from our detailed permit file review (N=143).

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 120
Region
1 2 3 4 5F 5R 5S 6T 6V 7 8 9
Acreage
0 20
40 60
80 100
Impacted
Required
Figure 13. Acres impacted and acres of mitigation required displayed by state board region from the project analyses for files assessed (N=143).

Region
1 2 3 4 5F 5R 5S 6T 6V 7 8 9
Mean Mitigation Ratio (Required:Impacted)
0 2
4 6
8 10
12 N= 11 26 14 19 5 6 24 2 2 3 13 15
Figure 14. Mitigation ratios required by region (N=143).

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 121
Acreage from SWRCB database - Acreage from Project Analyses <-10 to -100
<-1 to -10
<-0.1 to -1
<-0.01 to -0.1
<-0.001 to -0.01
<0 to -0.001
0 >0 to 0.001
>0.001 to 0.01
>0.01 to 0.1
>0.1 to 1
>1 to 10
>10 to 100
Number of Files
0 20
40 60
80 Impacts
Required
Figure 15. Plot of the differences between the impacted and required acreage values obtained through our detailed file review, and the corresponding values recorded in the SWRCB database.

A logarithmic scale was used for the data bins due to the wide range of acreage values involved. Negative values indicate that a lower value of acreage required was recorded in the SWRCB database compared to the acreage calculated during project analyses.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 122
Percentage
0 10 20 30 40 50 60 70 80 90 100
Number of Files
0 20
40 60
80 Average 401 Score
401 Percent-Met Score
Figure 16. Distribution of files according to the average 401 permit compliance score and 401 percent-met score (N=124 files with assessable 401 permit conditions).

Percentage
0 10 20 30 40 50 60 70 80 90 100
Number of Files
0 20
40 60
80 Average Mitigation Plan Score
Mitigation Plan Percent-Met Score
Figure 17. Distribution of files according to the average mitigation plan compliance score and mitigation plan percent-met score (N=81 files with assessable mitigation plan conditions).

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 123
Percentage of Conditions that could not be determined
0 10 20 30 40 50 60 70 80 90 100
Number of Files
0 10
20 30
40 50
Figure 18. Distribution of files according to the percentage of 401 permit compliance conditions that could not be determined (N=124 files with assessable 401 permit conditions).

Percentage of Conditions that could not be determined
0 10 20 30 40 50 60 70 80 90 100
Number of Files
0 10
20 30
40 50
Figure 19. Distribution of files according to the percentage of mitigation plan compliance conditions that could not be determined (N=81 files with assessable mitigation plan conditions).

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 124
401 Certification Year
1992 1994 1996 1998 2000 2002
Average 401 Score
0 20
40 60
80 100
Figure 20. Relationship between 401 certification year and average 401 permit compliance score (N= 124 files with assessable 401 permit conditions; p=0.845, r²=0.000). 401 Certifcation Year
1992 1994 1996 1998 2000 2002
Average Mitigation Plan Score
0 20
40 60
80 100
Figure 21. Relationship between 401 certification year and average mitigation plan compliance score (N= 81 files with assessable mitigation plan conditions; p=0.119, r²=0.030).

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 125
Region
123456789
Average 401 Score
0 20
40 60
80 100
N= 11 25 13 19 33 4 2 13 4
Figure 22. Average percentage score for 401 permit compliance by state board region (N=124 files with assessable 401 permit conditions).

Region
123456789
Average Mitigation Plan Score
0 20
40 60
80 100
N= 3 22 11 13 16 * 2 4 10
Figure 23. Average percentage score for mitigation plan compliance by state board region (N=81 files with assessable mitigation plan conditions).
*None of the four files from Region 6 included mitigation plans.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 126
Certification Type
Certification
Conditional Certification
Waiver
Conditional Waiver
Average 401 Score
0 20
40 60
80 100
N= 15 32 71 25
Figure 24. Average 401 score by certification type (N=143 files).

The categories used in this analysis correspond to the categories in the SWRCB database as follows:

Certification=CERT, STDCERT, WDR; Conditional Certification=CONDCERT; Waiver=WAIVE, WDRWV; Conditional Waiver=CNDWV, WDRCNDWV. Several files were listed as certifications and as waivers of waste discharge requirements; these files were categorized as certifications for the purposes of this figure. File #0 was not listed in any of these categories in the SWRCB database, so we determined from the 401 permit that it was a certification and waiver of waste discharge requirements. Therefore, it is listed as a certification for this analysis.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 127
Third Party
Acreage
Site Implementation
Site Maintenance
Site Protection
Success & Performance Standards
Monitoring & Submission
Other
Percentage
0 20
40 60
80 100
Avg 401 Score
Avg % of non-determinable conditions
Figure 25. Average scores for 401 permit compliance and average percentage of conditions that could not be determined grouped by the type of permit condition (N=124 files with assessable 401 permit conditions).

Third Party
Acreage
Site Implementation
Site Maintenance
Site Protection
Success & Performance Standards
Monitoring & Submission
Other
Percentage
0 20
40 60
80 100
Avg MP Score
Avg % of non-determinable conditions
Figure 26. Average scores for mitigation plan compliance and average percentage of conditions that could not be determined grouped by the type of permit condition. (N = 81 files with assessable mitigation plan conditions).

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 128
No
Number of 401 Conditions 401 permit
0 12
3 45
67 8
9 10
11 12
Number of 401 Permit Orders
0 10
20 30
40 Figure 27. Breakdown of the number of mitigation-related permit requirements (conditions) in each 401 permit order (N=143).

Region
1 2 3 4 5F 5R 5S 6T 6V 7 8 9
Mean Number of Conditions
0 2
4 6
8 10
12 14
N= 11 26 15 19 6 6 25 2 2 2 13 5
Figure 28. Mean number of mitigation-related 401 conditions per order within each SWRCB Region, including standard error bars (N=132). Eleven files for which no 401 permit was obtained were excluded.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 129
Third Party
Acreage
Site Implementation
Site Maintenance
Site Protection
Success and Performance Standards
Monitoring and Submission
Other
Number of Conditions
0 20
40 60
80 100
120
Figure 29. Breakdown of all mitigation-related 401 permit conditions by condition category (N=132).

The conditions from all permit orders were combined into a single list prior to categorization. Eleven files for which no 401 permit was obtained were excluded.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 130
Third Party
Acreage
Site Implementation
Site Maintenance
Site Protection
Success and Performance Standards
Monitoring and Submission
Other
Mean Number of Conditions per Permit Order

0.0 0.5

1.0 1.5

2.0 2.5
Figure 30. Mean number of mitigation-related 401 permit conditions per permit order (N=132).

Number of 401 Conditions Per File
1234
Percentage of Conditions Per 401 Order
0 10
20 30
40 50
60 Third Party
Acreage
Site Implementation
Site Maintenance
Site Protection
Success and Performance Standards
Monitoring and Submission
Other
Figure 31. Frequency of occurrence for the eight permit condition categories when the 401 order includes just a single mitigation-related condition (N=36, 23, 18, 14, respectively).

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 131
Unique to 401 Redundant Invoking
Percentage of Conditions
0 10
20 30
40 50
Figure 32. Percentage of mitigation-related conditions found in 401 permit orders that were unique to the 401, redundant with equivalent conditions required by other regulatory agencies, or invoking those other agency permits or the common mitigation plan (i.e., “must follow the 404”) (N=115).

Region
1 2 3 4 5F 5R 5S 6T 6V 7 8 9
Percentage of Conditions
0 20
40 60
80 100
% Redundant
% Invoking
N= 11 13 16 5 6 22 2 1 11 3 23 2
Figure 33. Percentage of redundant and invoking 401 conditions by Region (N=115).

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 132
Mean 401 Score
ND 0 10 20 30 40 50 60 70 80 90 100
Number of Files
0 20
40 60
80 100
ND
Figure 34. Distribution of files according to the average 401 permit compliance score including only those mitigation conditions explicitly specified in the 401 permit order (N=143).

Wetland Class
Depressional
Estuarine
Lacustrine
Lagoon
Riverine High
Riverine Low
Seep and Spring
Vernal Pool
Number of Mitigation Sites
0 20
40 60
80 100
Figure 35. Breakdown of wetland hydrogeomorphic classes as defined and assessed by the CRAM evaluations for all 204 mitigation sites representing 129 files evaluated using CRAM.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 133
Figure 36. Distribution of assessed mitigation sites by wetland class across the state.

Symbols indicate individual mitigation actions; multiple points may be indicated for individual projects with multiple mitigation actions, and some points may represent multiple projects, e.g., mitigation banks.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 134
Percentage Score
0 10 20 30 40 50 60 70 80 90 100
Percentage of Files / Reference Sites
0 10
20 30
40 50
Reference Site Data
Filewide Data
Poor Marginal Sub-Optimal Optimal
Figure 37. All CRAM data combined into a single overall wetland condition success score for each of the 129 files and 47 reference sites evaluated using CRAM.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 135
Figure 38. Map of California showing location of mitigation sites color coded by condition score.

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 136
401 Certification Year
1992 1994 1996 1998 2000 2002
Mean Overall CRAM Score
0 20
40 60
80 100
Figure 39. Relationship between 401 certification year and filewide mean overall CRAM percentage scores grouped by certification year (N=129 files, r2
=0.005, p=0.415).

Region
1 2 3 4 5F 5R 5S 6V 6T 7 8 9
Mean Overall CRAM Score
0 20
40 60
80 100
N= 9 25 13 15 4 6 24 2 2 2 13 14
Figure 40. Filewide mean Total-CRAM percentage scores by SB region (N=129 files).

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 137
Region
1 2 3 4 5F 5R 5S 6T 6V 7 8 9
Percentage of Files
0 20
40 60
80 100
Marginal/Poor
Sub-Optimal
Optimal
N= 9 25 13 15 4 6 24 2 2 2 13 14
Figure 41. Percentage of files in CRAM success categories by state board region (N=129 files).

Evaluation of California Compensatory Wetland Mitigation Final Report (Review Copy) 138
Percentage Score
0 10 20 30 40 50 60 70 80 90 100
Percentage of Files / Reference Sites
0 10
20 30
40 50
60 Reference Site Data
Filewide Data
Poor Marginal Sub-Optimal Optimal
Figure 42. Landscape context metric CRAM scores compared to reference-site data.