St. Helena City Council
February 10, 2026

CEQA Biological Resources Analysis
Spring Grove Project
St. Helena, California
31 15. LITERATURE CITED
Baldwin D.H, Goldman D.H., Keil D.J., Patterson R, Rosatti T.J., Wilken D.H. (ed.). 2012. The Jepson Manual Vascular Plants of California: Second Edition. University of California Press, Berkeley. 1568 pps.

Beier, P. and S. Loe. 1992. "In my experience.." a checklist for evaluating impacts to wildlife movement corridors. Wildlife Society Bulletin Vol. 20(4): 6.

CDFW (California Department of Fish and Wildlife). 2016. Complete list of amphibian, reptile, bird, and mammal species in California. Published September 2008; updated May 2016.

CNDDB (California Natural Diversity Data Base). 2024. RareFind 6. Computer printout for special-status species within a 5-mile radius of the project site. California Natural Heritage Division, California Department of Fish and Wildlife, Sacramento, CA.

Holland, V.L. & D.J. Keil. 1989. California vegetation. Biological Sciences Department/ California Polytechnic State University. San Luis Obispo, California.

Holland, V.L. & D.J. Keil. 1995. California vegetation. Kendall/Hunt Publishing Company.

Agenda Item #11.1.

Agenda Item #11.1.

Agenda Item #11.1.

Agenda Item #11.1.

Table 1
Plant Species Observed on the Spring Grove Project Site monk & associates
Gymnosperms
Cupressaceae
Sequoia sempervirens Redwood
Angiosperms - Dicots
Apocynaceae
*Nerium oleander Oleander
Fagaceae
Quercus agrifolia var. agrifolia Coast live oak
Quercus lobata Valley oak
Lythraceae
*Punica granatum Pomegranate
Moraceae
*Ficus carica Fig
Myrtaceae
*Eucalyptus sp. Eucalyptus
Oleaceae
*Ligustrum japonicum Japanese privet
*Ligustrum lucidum Glossy privet
*Olea europaea Olive
Rosaceae
*Prunus cerasifera Cherry plum
*Pyracantha coccinea Scarlet firethorn
Sapindaceae
*Acer palmatum Japanese maple
*Acer saccharinum Silver maple
Angiosperms -Monocots
Poaceae
*Avena barbata Slender wild oat
*Bromus diandrus Ripgut grass
*Bromus hordeaceus Soft chess
*Phyllostachys aurea Fishpole bamboo
* Indicates a non-native species Page 1 of 1
Agenda Item #11.1.

Table 2
Wildlife Species Observed on the Spring Grove Project Site

MONK & ASSOCIATES
Reptiles
Western fence lizard Sceloporus occidentalis
Birds
Northern Flicker Colaptes auratus
Turkey Vulture Streptopelia decaocto
California Quail Callipepla californica
Anna's Hummingbird Calypte anna
Acorn Woodpecker Melanerpes formicivorus
Nuttall's Woodpecker Picoides nuttallii
California Scrub Jay Aphelocoma californica
Oak Titmouse Baeolophus inornatus
Bushtit Psaltriparus minimus
Ruby-crowned Kinglet Regulus calendula
American Robin Turdus migratorius
Yellow-rumped Warbler Setophaga coronata
California Towhee Melozone crissalis
Dark-eyed Junco Junco hyemalis
House Finch Haemorhous mexicanus
Lesser Goldfinch Spinus psaltria
Mammals
Fox squirrel Sciurus niger

Agenda Item #11.1.

TREES:

Acacia cultriformis Knife Acacia 24” Box L
Acer ssp. Autumn Blaze Red Maple 36” Box M
Araucarua heterophylla Norfolk Island Pine 36” Box M
Arbutus ‘Marina’ Marina Strawberry Tree 36” Box L
Cassia leptophylla Gold Medallion Tree 36” Box L
Cercidium ‘Desert Museum’ Desert Museum Palo Verde 36” Box VL Cercis canadensis Eastern Redbud 24” Box M
Cercis occidentalis Western Redbud 24” Box VL
X Chitalpa tashkentensis ‘Pink Dawn’ Chitalpa Chitalpa 36” Box L Cornus ssp. Dogwood 24” Box M
Ginkgo biloba ‘Autumn Gold’ Autumn Gold Maidenhair Tree 36” Box M Jacaranda mimosifolia Jacaranda 36” Box M
Lagerstroemia spp. Crape Myrtle 24” Box L
Laurus nobilis Bay Laurel 24” Box L
Liriodendon tulipifera Tulip Tree 36” Box M
Magnolia grandiflora Southern Magnolia 36” Box M
Magnolia stellata Star Magnolia 24” Box M
Melalueca quinquervia Paperbark Tree 24” Box L
Olea europaea ‘Swan Hill’ Olive 36” Box VL
Pinus halepensis Aleppo pine 24” Box L
Pinus pinea Italian Stone Pine 36” Box L
Pistacia chinensis Chinese Pistache 36” Box L
Platanus x acerfolia ‘Columbia’ London Plane Tree 36” Box M Platanus racemosa Western Sycamore 36” Box M
Podocarpus gracilior African Fern Pine 36” Box M
Podocarpus elongatus ‘Monmal’ Icee Blue Yellow-wood 24” Box M Prunus ssp. Cherry Plum 24” Box L/M
Punica granatum Pomegranate 24” Box L
Pyrus ssp. Pear 36” Box M
Quercus ssp. Valley Oak 36” Box L/M
Rhus lancea African Sumac 36” Box L
Tillia ssp. Linden 24” Box M
Ulmus parvifolia ‘True Green’ True Green Chinese Elm 36” Box M Umbellularia californica California Bay 24” Box M

SCIENTIFIC NAME COMMON NAME SIZE WUCOLS

S H R U B A N D
GROUNDCOVER:

Acacia redolens Prostrate Acacia 1 gal. VL
Agave spp. Agave 5 gal. VL
Aloe spp. Aloe 5 gal. L
Alogyne huegelli Blue Hibiscus 15 gal. L
Arctostaphylos spp. Manzanita 5 gal. L
Artemisia ssp. Sagebrush 1 gal. L
Asparagus densiflorus Asparagus Fern 5 gal. M
Bouteloea gracilis Blue Grama 1 gal. L
Buxus japonica ‘Green Beauty’ Japanese Boxwood 5 gal. M Callistemon spp. Bottle Brush 5 gal. L
Carex tumulicola Foothill Sedge 1 gal. L
Chondropetalum tectorum Small Cape Rush 5 gal. L
Cistus ‘Sunset’ Magenta Rockrose 5 gal. L
Cotoneaster spp. Cotoneaster 5 gal. L
Crassula spp. Jade Plant 5 gal. L
Dasyliron spp. Desert Spoon 5 gal. VL
Dianella ssp. Flax Lily 5 gal. L
Dietes bicolor Fortnight Lily 5 gal. L
Festuca mairei Atlas Fescue 5 gal. L
Hesperaloe spp. Yucca 5 gal. L
Heteromeles arbutifolia Toyon 15 gal. VL
Ilex spp. Holly 5 gal. M
Juncus patens California Gray Rush 5 gal. M
Lavandula ssp. Lavender 5 gal. L
Leucophyllum spp. Barometer Bush 5 gal. L
Leymus condensatus ‘Canyon Prince’ Canyon Prince Wild Rye 5 gal. L Ligustrum japonicum ‘Texanum’ Waxleaf Privet 5 gal. L
Lomandra spp. Dwarf Mat Rush 5 gal. L
Myoporum p. ‘Putah Creek’ Creeping Myoporum 1 gal. L
Myrtus communis Common Myrtle 5 gal. L

SCIENTIFIC NAME COMMON NAME SIZE WUCOLS

VINES AND
Bougainvillea spp.

Calliandra haematocephala
Distictis ssp.

Eriobotrya deflexa
Jasminum polyanthum
Lonicera japonica
Macfadyena unguis-cati
Magnolia grandiflora ‘Little Gem’
Trachelospermum jasmnoides
Rosa banksiae
Vitis ssp.

Olea europaea ‘Montra’
Pittosporum tobira variegata
Phormium spp.

Prunus caroliniana ‘Bright n Tight’
Rhaphiolepis spp.

Rosmarinus officinalis
Rosa ssp.

Salvia ‘Bees Bliss’
Trachelospermum asiaticum
Westringia spp.

SCIENTIFIC NAME

SCIENTIFIC NAME

SPRING STREET
LEGEND
Focal Tre
Large Ac
Medium
Basin Tre
Flowerin
Fruiting T
Small Pa
Large Ac
0 15’ 30’ 60’
Scale: 1”= 30’-0”

CONCEPTUAL PLANT PALETTE

ILLUSTRATIVE SITE PLAN
FDS
Planning | Design | Collaboration

FOURNIER DESIGN STUDIO
[email protected] (562) 719-4650
Santa Rosa, CA
Brittain Townhomes FDS

FOURNIER DESIGN STUDIO
Sebastopol and Brittain Rd. City Ventures 2nd City Submittal 2022#12345
CONCEPTUAL PERIMETER FENCE ‘A’ - 8’-0”H
SCALE: NTS
CONCEPTUAL PERIMETER FENCE ‘B’ - 6’-0”H
SCALE: NTS
Spring Grove
1447~1505~1515 Spring Street
St. Helena, CA
APN: 009-590-003, -005, -010
FDS #24010
September 30th, 2024
G-5
Agenda Item #11.1.

SPRING STREET
BLDG 6 BLDG 1
BLDG 3 BLDG 2
BLDG 4
BLDG 5 STREET A STREET C
UNIT 1
UNIT 1
UNIT 1
UNIT 1 UNIT 3 UNIT 3 UNIT 4 UNIT 4 UNIT 3 UNIT 3 UNIT 4 UNIT 4 UNIT 4 UNIT 4 UNIT 4 UNIT 4
UNIT 4 UNIT 4
UNIT 4
UNIT 4
UNIT 4 UNIT 2
UNIT 3
UNIT 4 UNIT 2
UNIT 3 UNIT 3 UNIT 3 UNIT 3 UNIT 3 UNIT 2
UNIT 4 UNIT 2
UNIT 3
UNIT 3
UNIT 4
UNIT 3
UNIT 4
UNIT 3
UNIT 2
UNIT 3
S
S
S
S

ADA ADA ADA ADA ADA

STREET B SITE LAYOUT
C1.1 TENTATIVE MAP SPRING GROVE adobe associates, inc. 1220 N. Dutton Ave., Santa Rosa, CA 95401 P. (707) 541-2300 F. (707) 541-2301 Website: www.adobeinc.com "A Service You Can Count On!" civil engineering I land surveying I wastewater

SITE LAYOUT LEGEND
HATCHING LEGEND: Agenda Item #11.1.

1185MASON STREET SAN FRANCISCO,CALIFORNIA 94123
November 19, 2025
Mr. Mike Janusek, AICP
Senior Planner
City of St. Helena
Community Development Department
1088 College Avenue
St. Helena, CA 94574
Mike,
We appreciate your thorough review of the Spring Grove project. Please find below our response to how we have addressed public input and Planning Commission early feedback since the Planning Commission meeting of January 21, 2025.

Thank you,
Kelso G. Barnett
cc. Maya DeRosa
Xinia Gamero
Michael Brewer
Agenda Item #11.1.

1185MASON STREET SAN FRANCISCO,CALIFORNIA 94123
Project Screening
The entire community will be enclosed by a wood perimeter fence designed to provide both privacy and a consistent visual edge. The fence height will be 6 feet in most areas; however, within the 50 feet of the property line/right of way on Spring Street, fencing will be reduced to 3 feet to maintain open sightlines in order to comply with the city’s vision triangle and applicable design standards.

Beyond fencing, the landscape plan incorporates significant new planting to create a natural, attractive, and effective vegetative screen. Along the eastern boundary—where Spring Grove is adjacent to St. James Drive—the project will install six 36-inch-box Carolina Cherry Laurels. These evergreen trees are fastgrowing, typically reaching heights of 20–35 feet and widths of 15–20 feet, with a dense pyramidal canopy that provides year-round privacy.

The southern property line will be softened and screened with a combination of four Western Sycamores and three Western Redbuds. The Western Sycamore is a fast-growing deciduous tree, commonly growing 2–3 feet per year. At maturity it can reach 40–80 feet in height with a broad 30–50-foot spread. While it drops leaves in winter, this location is adjacent to the open-space interface on the St. Helena Park development, where seasonal leaf-off is appropriate and helps maintain a more open feel toward the shared open area. The Western Redbud is a moderate-growing deciduous tree, typically reaching 15–20 feet in height and 15–20 feet in width. It grows roughly 1–2 feet per year and is valued for its dense branching, spring blossoms, and summer privacy. Although deciduous, its branching structure still provides some winter screening.

Along Spring Street, the project will be screened primarily through the preservation of existing mature redwoods and sycamores, supplemented with additional understory and mid-story plantings, including Vine Maples, Rocky Mountain Maples, Coast Live Oaks, and Engelmann Oaks. The Vine Maple is a moderate-growing deciduous tree or large shrub, typically growing 1–2 feet per year. Mature size ranges from 10–20 feet tall and wide, forming layered, multi-stemmed screening. The Coast Live Oak is a slowto-moderate growing evergreen oak, adding roughly 12–24 inches of growth per year. At maturity it can reach 20–70 feet tall with a broad 20–30-foot canopy. Its year-round foliage provides consistent evergreen screening. Also in the oak family, the Engelmann Oak is a slow-growing semi-evergreen oak, usually adding less than 12 inches of growth per year. Mature height ranges from 20–60 feet with a 20–40-foot spread. Though technically semi-deciduous, it generally retains most of its leaves year-round in our climate, making it effective for long-term screening.

Additional shrub layers will reinforce the project’s planting design, including Woodleaf Privet and Toyon.

These shrubs are fast-growing and evergreen, typically adding 2 feet per year, maturing at 8-12 feet tall and 6-10 feet wide.

Taken together, this combination of fencing, preserved mature trees, new evergreen and deciduous plantings, and layered shrub material will provide a robust, attractive, and context-appropriate screening strategy for Spring Grove on all sides.

Agenda Item #11.1.

1185MASON STREET SAN FRANCISCO,CALIFORNIA 94123
Feedback: Neighborhood Outreach, Providing Smaller Units & Community Gardens We received extensive community input during the January 21, 2025 Planning Commission study session.

We took that feedback seriously and immediately followed up with a dedicated Open House on February 20, 2025 at the Native Sons Hall on Spring Street, just steps from the project site. This informal setting allowed neighbors to speak with us one-on-one, walk through the plans in detail, and share both their support and their concerns directly.

Across all of these engagements; emails, in-person conversations, comments at public meetings, and comments on Next Door, the community’s feedback has been mixed. Many residents expressed support for increasing “missing-middle” homeownership opportunities in St. Helena, especially for young families and first-time homebuyers. Others shared concerns about the project’s density, height, and architectural expression. We listened carefully and committed to making meaningful improvements in the areas where adjustments were feasible.

Density is set by the municipal code and the Housing Element designation for this site, which has a direct effect on our overall building height and our required unit sizes; as such, those parameters could not be reduced without undermining both the City’s housing obligations and the core purpose of this project.

However, within those constraints, we made a series of targeted design refinements aimed at softening the buildings, improving compatibility with nearby homes, and enhancing the visual character of the neighborhood. These enhancements include:

Architectural and Aesthetic Improvements
• Increased the primary roof pitch from 8:12 to 11:12 to create a more traditional residential profile and reduce the appearance of massing.
• Removed the top parapet wall to improve roof articulation and achieve a cleaner, more refined ridgeline.
• Introduced undulating ridgelines with gable peaks raised approximately 3 feet to create variation and visual interest.
• Added accent metal roof sheds at key elevations to break up facades and add high-quality material expression.
• Added second- and third-floor balcony decks to create depth, shadow lines, and more humanscaled architecture. • Increased the level of siding detail throughout the elevations to enrich texture and avoid flat or monolithic façades.
• Added vine-trellis systems along garage walls to introduce greenery and soften the appearance of auto-oriented elements.
• Added enhanced architectural fixtures at gable ends and other focal points for additional craftsmanship and detail.
• Provided varied garage-door styles to better individualize units and reduce visual repetition.

Agenda Item #11.1.

1185MASON STREET SAN FRANCISCO,CALIFORNIA 94123
• Engaged a professional colorist to refine and diversify the color palette in a way that complements the existing neighborhood fabric.

After considering the feedback received during the January study session, we evaluated the idea of incorporating community garden beds into the common open space. While community gardens can be successful in some developments, our experience across multiple City Ventures communities shows that they tend to be highly variable; some are heavily utilized, while others are quickly abandoned, becoming maintenance challenges and detracting from the quality of shared spaces. Rather than installing garden beds that may not reflect the preferences or lifestyles of the future residents, we believe the most responsible and community-oriented approach is to allow the future HOA to make this decision once the neighborhood is established. This ensures that any garden areas created will reflect genuine resident interest, be maintained appropriately, and become an amenity that truly adds value to the community.

While we could not change every aspect raised by neighbors, we worked within the boundaries of the development code and Housing Element requirements to make improvements that materially enhance the project’s compatibility with its surroundings. Our goal has always been to design a neighborhood that feels at home in St. Helena—respectful of the community’s character, aligned with the City’s long-range housing policies, and accessible to the households this project is intended to serve.

Circulation & Fire Access
The project’s circulation system and fire access design have been reviewed and approved by the St.

Helena Fire Department, including the City’s Fire Inspector. The site plan meets all applicable requirements of the California Fire Code, including turning radii, street widths, hydrant placement, emergency access routes, and unobstructed fire-lane clearances. While feedback was received adding a secondary point of egress (EVA/fire access), state fire code allows an exemption for up to 200 units if all units are equipped with fire sprinklers. The project will be designed with an NFPA-13 system and all units will have fire sprinklers.

In addition, we commissioned a Transportation Impact Study to evaluate on-site circulation, emergency vehicle operations, driveway performance, and interaction with the public street network. Technical review confirm that the project is safe, functional, and capable of supporting timely emergency response:
o While private streets are not subject to city standards, the internal streets are 28 feet wide, which exceeds City standards for public streets and accommodating two-way travel and emergency access.
o While no on-street parking is permitted on these internal streets, ensuring clear and unobstructed fire lanes, the project will exceed parking requirements. Two side-by-side parking spaces within garages will be provided for each unit, exceeding the 1.5 parking spaces required by code. Guest parking also exceeds code requirements by providing 11 spaces. o A 200-foot driveway connects the site to Spring Street. Although the internal circulation pattern resembles a parking-lot drive aisle rather than a public street, both vehicle speeds and vehicle Agenda Item #11.1.

1185MASON STREET SAN FRANCISCO,CALIFORNIA 94123
volumes will be low and drivers are expected to proceed cautiously—conditions that reduce conflict potential and improve overall safety.
o The perpendicular guest parking spaces create typical interactions found in commercial and residential lots throughout the Bay Area; the Transportation Impact Study found no unique or unacceptable safety concerns.
o Wayfinding and delivery-parking signage will be installed to prevent delivery vehicles from obstructing garages or internal circulation routes, further supporting safe operations and unimpaired access for fire responders.

Both the Fire Department and the transportation study confirm that the project is fully compliant, safe, and operationally sound. Emergency access meets or exceeds all City and California Fire Code standards; internal circulation is appropriate for a residential environment with low speeds and low volumes; and no unique hazards or conflicts were identified.

Taken together, the evidence demonstrates that the development will function safely from both a circulation and fire-response perspective and will not adversely affect emergency response times.

Lighting / Off-Site Glare
The project’s lighting plan was carefully designed to prevent off-site glare and minimize light spillover onto neighboring residential properties along Spring Street, St. James Drive, and the surrounding areas.

The photometric layout shows that all illumination levels taper down to near-zero at the property boundaries. This is achieved through the use of fully shielded and house-side-shielded fixtures, including cutoff post-top luminaires, wall-mounted LED fixtures with controlled distribution patterns, and lowheight bollards located away from sensitive edges. The placement of fixtures is oriented inward toward internal drive aisles, walkways, and common areas, ensuring light is directed only where needed for safety and not toward adjacent homes.

In addition to shielding and directional control, the lighting plan incorporates appropriately warmtemperature LEDs, and precise photometric modeling to ensure compliance with modern lighting standards and to maintain a residential character. The combination of controlled optics, house-side shielding, and low-intensity fixture selection results in extremely limited light trespass. Together, these measures ensure that the project provides safe, adequate illumination for residents and visitors while protecting the privacy, comfort, and nighttime environment of the surrounding neighborhood.

Agenda Item #11.1.

1185MASON STREET SAN FRANCISCO,CALIFORNIA 94123

SPRING GROVE

WATER NEUTRALITY REPORT
Updated
November 10, 2025
A. BACKGROUND
To ensure compliance with the City of St. Helena's Water Neutral Policy for Development, City Ventures is submitting a water neutrality analysis for the proposed Spring Grove Subdivision.

This analysis outlines the anticipated water consumption for the development (per St. Helena’s water neutral policy) and presents potential options for water off-set mitigation.

B. CURRENT THEORETICAL DOMESTIC WATER DEMAND
There are currently three homes on the property: one 4-bedroom, one 3-bedroom, and one 2-bedroom.

According to the St. Helena Water Neutral Policy for Development section on water demand analysis for existing water use, the analysis should use 150 gpd/bedroom. This equates to a theoretical domestic water demand to be 1,350 gallons per day for the current homes.

Address Bedrooms Gal/Bdrm
1447 Spring 4 150
1505 Spring 2 150
1515 Spring 3 150
Gal/Day 1,350
Gal/Year 492,750
* Current landscape irrigation not supplied by city water but by well onsite C. PROPOSED THEORETICAL DOMESTIC WATER DEMAND
The proposed development includes 41 lots with side-by-side townhomes, consisting of 41 3+ bedroom units. Water demand for the project is calculated based on the water usage of fixtures and appliances, determined by the number of occupants, in accordance with the City of St. Helena's Water Neutral Policy for Development.

Agenda Item #11.1.

1185MASON STREET SAN FRANCISCO,CALIFORNIA 94123
The following outlines the water performance requirements for fixtures and appliances.

Fixture # Unit Days Occupant
Toilet 3 Flushes 1 1
Sink Faucet 0.25 Minute 3 1
Showerhead 8 Minute 1 1
Kitchen Faucet 4 Minute 1 1
Clothes Washer 0.37 Load 1 1
Dish Washer 0.1 Load 1 1
Water usage from fixtures and appliances will comply with the minimum standards set by the 2010 California Green Building Standards Code (Title 24, Part 11) and the City of St. Helena's Water Neutral Policy.

Fixture Gal Unit
Toilet 1.3 Flushes
Sink Faucet 1.5 Minute
Showerhead 2 Minute
Kitchen Faucet 1.5 Minute
Clothes Washer 19.7 Load
Dish Washer 6.3 Load
In accordance with the City of St. Helena's policy, the analysis assumes two occupants per bedroom (including a den/office/optional bedroom). Following the City's Water Neutral Policy for Development, the estimated domestic water demand for the proposed project is 11,182 gallons per day. The calculations are included below.

Number of Rooms Number of Occupants
Unit Bdrm Other Total
Per
Rm.

Per
Unit
#
Units Total
Plan 1 3 0 3 2 6 4 24
Plan 2 3 1 4 2 8 5 40
Plan 3 3 1 4 2 8 15 120
Plan 4 3 1 4 2 8 17 136
Total Number of Occupants* 320
* Total number of occupants per St. Helena Water Neutral Policy; even if the total number of residents per home exceeds St. Helena’s average household size, the projected number of actual occupants in a real world scenario is projected to be significantly lower.

Agenda Item #11.1.

1185MASON STREET SAN FRANCISCO,CALIFORNIA 94123
Fixture Unit Times/Day Gal/Unit Occupants Gal/Day
Toilet 1 3 1 320 1,248
Sink Faucet 0 3 2 320 360
Showerhead 8 1 2 320 5,120
Kitchen Faucet 4 1 2 320 1,920
Clothes Washer 0 1 20 320 2,332
Dish Washer 0 1 6 320 202
Total Gal/Day 11,182
Total Gal/Year 4,081,459
D. THEORETICAL LANDSCAPE WATER DEMAND
Landscape irrigation will be provided by the City of St. Helena and the current well will be abandoned.

Per Landscape Sheet L-7, the estimated total water usage per year (ETWU) is 285,008 gallons, or 781 gallons per day.

E. PROPOSED WATER NEUTRALITY MITIGATION
The existing theoretical water usage on the property is 1,350 gallons per day, and the proposed developments theoretical water usage is 11,963 gallons per day. Therefore, to comply with water neutrality the development must offset 10,613 gallons per day.

Spring Grove Theoretical Water Usage Gal/Day
Homes 11,182
Landscape 781
Total 11,963
Existing Theoretical Water Usage 1,350
Gal/Day to Offset 10,613
This can be done by either retrofitting residential and non-residential properties with low flow devices or by paying an in-lieu fee that is equivalent to the cost of retrofitting enough homes, as provided by Municipal Code 13.12.020. The calculation of the in-lieu fee is attached below.

F. IN-LIEU RETROFIT FEES
The proposed project will increase water usage by 10,613 gallons per day. Therefore, per city policy, the project proposes to pay an in-lieu fee equivalent to the cost of retrofitting showerheads, faucets, and toilets in existing homes to comply with the required water savings.

Agenda Item #11.1.

1185MASON STREET SAN FRANCISCO,CALIFORNIA 94123
The theoretical water use for homes with retrofitted low flow fixtures are summarized in the table below:

Theoretical Water use with retrofitted low flow fixtures per residence Flow Occupants
Fixture Rate Duration Daily Uses Number* Gal/Day
Toilet 2 8 1 7.8 125
Sink Faucet 1.5 0.25 3 7.8 9
Showerhead 1.5 4 1 7.8 47
Kitchen Faucet 1.3 1 3 7.8 30
Clothes Washer 19.7 1 0.37 7.8 57
Dish Washer 6.3 1 0.1 7.8 5
Total Gal/Day - 1 Home (Rooms - Spring Grove Blended Avg.) 273 Total Gal/Day - 41 Homes (Rooms - Spring Grove Blended Avg.) 11,182 *Number of occupants (7.8) is a blended average of the proposed project’s number of residents per unit Theoretical Water Use Saving per Residence
Gal/Rm Rooms* Total
Theoretical Water Use - Old Fixtures 150 3.9 585
Theoretical Water Use - New Fixtures 273
Savings 312
*Number of rooms (3.9) is a blended average of the proposed project’s number of rooms per unit To offset the anticipated water demand of the proposed project, 34 residences will need retrofitting with low flow fixtures.

Required Residences needing retrofitting
Gal/Day
New Water Demand 10,613
Water Savings per Home (Retrofit) 312
Residences Needing Retrofit 34
Agenda Item #11.1.

1185MASON STREET SAN FRANCISCO,CALIFORNIA 94123
The expected cost of retrofitting existing homes with new low flow fixtures are summarized in the table below:

Retrofit Material Cost per residence
Fixture Qty Cost/Unit Total
Showerhead 2 $139.06 $278.12
Sink Faucet 2 $247.98 $495.96
Kitchen Faucet 1 $341.08 $341.08
Toilet 2 $114.69 $229.38
Clothes Washer 1 $698.00 $698.00
Dish Washer 1 $449.00 $449.00
Total Estimated Material Cost Per Home $2,491.54
Retrofit Labor Cost Per residence
Fixture Hours Qty Hourly Rate Total Cost
Showerhead 1 2 $110.00 $220.00
Sink Faucet 1 2 $110.00 $220.00
Kitchen Faucet 1 1 $110.00 $110.00
Toilet 2 2 $110.00 $440.00
Clothes Washer 2 1 $110.00 $220.00
Dish Washer 2 1 $110.00 $220.00
Total Estimated Labor Cost per Residence $1,430.00
The total estimated cost is the sum of the estimated material and labor cost per residence.

Total Retrofit Cost
Material Cost $2,492
Labor Cost $1,430
Total Cost $3,922
Residences Needing Retrofit 34
Total Estimated In-Lieu Fee $133,348
Agenda Item #11.1.

1185MASON STREET SAN FRANCISCO,CALIFORNIA 94123
Proposed low flow fixture costs are listed below. Please find links below; prices are current as of November 10, 2025
o Showerhead- Delta, Kayra, Monitor 14 Series Shower Trim in Chrome- Model#: T14233 o Sink Faucet- Delta, Kayra, Two Handle Widespread Bathroom Faucet in Chrome- Model#:
3533LF-MPU
o Kitchen Faucet- Delta, Coranto, Single Handle Pull-Down Kitchen Faucet in ChromeModel#: 9179-DST o Toilet- Western Pottery- 1.28G Elongated Bowl Easy Height- Model#: 832-T8ULF-HP o Clothes Washer - Samsung - 4.5 Cu. Ft. High-Efficiency Stackable Smart Front Load Washer with Steam and Super Speed Wash - Platinum
o Dish Washer- LG - 24 in. in Stainless Steel Front Control Dishwasher- Model #: LDFN3432T o Labor Cost cited in table 6 are estimates for contractors in the St. Helena area found on https://www.taskrabbit.com
G. CONCLUSION
The proposed project theoretical water usage is 11,963 gallons per day.

The existing theoretical water usage on site is 1,350 gallons per day.

Therefore, the project must save 10,613 gallons per day or pay an in-lieu fee to comply with St. Helena water neutrality policy for development.

The project will pay an in-lieu fee of $133,348, equal to the cost of retrofitting enough residences with low flow fixtures to offset the proposed theoretical water use.

Agenda Item #11.1.

Response to Public Comments from the January 13, 2026 City Council Public Hearing:

Spring Grove Vesting Tentative Map and Water Neutrality In-Lieu Retrofit Fee
Public Comments are grouped by theme and a single response prepared for related comments.

Concern that the project has not been processed in compliance with the St.

Helena Municipal Code.

The development application was submitted on November 22, 2024 and routed for review to Planning, Fire, Building, Public Works, City Arborist, and applicable outside agencies. The applicant was deemed to have submitted a complete application on May 13, 2025 and a subsequent consistency review commenced in compliance with SB 330.

The project underwent multiple rounds of consistency analysis before the project was determined to be consistent with all applicable state and local standards. Comment letters were posted to the project webpage on the City website. City staff have applied the State and Municipal Codes where applicable. If commentors wish to disagree with this assertion, they need to specify which Code they believe to have been violated before Council may consider their claim.

Concern that the Applicant has not satisfied all requirements of the Municipal Code.

Some provisions of the Municipal Code become applicable to the project at Building Permit and Final Inspection stages of the review process. The Public Works Department concurs that there are outstanding issues that will need to be addressed during subsequent phases of the development process. However, the outstanding issues are not prerequisites to Council approval of the Tentative Map or Water Neutrality In-Lieu Retrofit Fee. All outstanding issues will be resolved either before Building Permit issuance or Final Inspection, as appropriate.

For example, during the consideration of a tentative map application, Public Works staff will only review a preliminary stormwater management plan, and more detailed analysis of construction drawings is performed by Public Works staff during the building permit review. Post-construction storm water management remains an outstanding issue, but find there are multiple design options available to the applicant to mitigate the issue.

Staff will ensure that the project incorporates a reasonable solution in the plan prior to approving the Building Permit. As conditioned, resolution of this matter is not needed for Council to approve the Vesting Tentative Map or Water Neutrality In-Lieu Fee Retrofit request.

Site Layout and Irregular Lot
SHMC Section 16.32.050 requires new lots created by subdivisions to meet minimum lot dimensions. SHMC Table 17.16.030 (Development Standards) establishes the Agenda Item #11.1.

standards for residential lots. The project proposes a 2.43-acre (105,851 square feet) lot for development, which exceeds the required minimum lot area of 7,000 square feet.

In addition, the subdivision consolidates three existing parcels to create a 66-foot-wide public frontage, improving site access, and exceeding the required minimum lot width of 45 feet. While the resultant parcel is irregularly shaped, it exceeds the minimum dimensional requirements for lot area and lot width as required by the City Code, is buildable, and provides adequate access.

Fire Access
The project plans, dated November 10, 2025 (Attachment 5 of the January 13 City Council staff report) were reviewed for compliance with applicable state and local Fire codes by the Fire Department. The Vesting Tentative Map proposes a minimum 26- foot-wide Emergency Vehicle Access Easement (28-feet-wide in most places), which meets the minimum 26-foot-wide fire access width standard pursuant to California Fire Code Appendix D Section D105.2. In addition, the Minor Design Review application for the project proposes a 40-foot-tall building height, which requires aerial fire apparatus pursuant to California Fire Code Appendix D Section D105.2. Fire has verified aerial equipment can access the third story windows and turning radii are adequate. Lastly, the project is conditioned to install NFPA13 fire sprinklers in each unit prior to building permit issuance. Per the 2025 California Fire Code, Appendix D, Section D107.1, developments where the number of units exceeds 30 shall be provided with two separate fire access roads, except that where all units are equipped with an approved automatic fire sprinkler system, access from two directions shall not be required.

Therefore, as designed and conditioned, the single means of access from Spring Street fully complies with the 2025 California Fire Code, and a secondary means of access is not required.

Height
Proposed building elevations are not under purview of the City Council and were reviewed and found to be in compliance with all applicable zoning standards as part of the Minor Design Review. The project proposes a building height of 40 feet, which is less than the required maximum building height of 42 feet per SHMC Table

17.16.030(B).

Number of Affordable Units
Pursuant to the City’s Inclusionary Housing Ordinance, SHMC Section 17.30.030, the Project will provide at least 20% affordable units. Per SHMC Section 17.30.030.G.1, for owner-occupied affordable units, at least half of the required affordable units shall be available to households earning less than 100% of annual median income (AMI). The Project proposes a 41-unit, 19% affordable, multifamily development for moderate Agenda Item #11.1.

income households earning 120% or less of AMI, with four units available to households earning less than 100% AMI and four units available to households earning less than 120% AMI. Pursuant to SHMC Section 17.30.030(A), Vesting Tentative Map condition of approval #26 requires the developer to pay the housing in-lieu fee for the remaining

0.2 fractional units of affordable to achieve the 20% minimum requirement of the Inclusionary Housing Ordinance. Please note, the City Council will consider the affordable housing agreement for the project at a future date.

Architecture
Project architecture is not under City Council review and was reviewed and approved with the Minor Design Review. Pursuant to SHMC Section 17.05.080.B.2, new multifamily development consistent with the applicable development and design standards provided in SHMC Section 17.16.040 (objective design standards) are subject to Minor Design Review by the Community Development Director. The Minor Design Review for site plan layout, architecture, parking, and landscaping has been reviewed and found to be consistent with the objective design standards and ministerially approved by the Community Development Director. The administrative Notice of Action is included in Attachment 8 of the January 13 City Council staff report. No appeal was filed and the Minor Design Review approval is final.

Also it is important to note that as part of the last Housing Element cycle, the State required the City to streamline the review of multi-family housing and eliminate the zoning requirement of a discretionary Conditional Use Permit for multi-family housing.

Occupancy of Existing Homes
Some commenters have indicated that the applicant must replace the three existing homes on the property, indicating that those units provide existing affordable housing.

State law requires that if a new housing development requires the demolition of existing residential units, the new development must produce at least as many units as are being demolished. This project meets that requirement, since it will demolish 3 units, but produce 41. State law also requires that any “protected units” that are demolished must be replaced with an equal number of “protected units”. Protected units are units that are either (1) subject to local rent control requirements, (2) are subject to a recorded covenant, ordinance or law restricting rents to levels affordable to low or very low income household (i.e., households earning 80% or 50% of the Napa County area median income, adjusted for household size), or (3) were actually rented to a low or very-low income household within the past 5 years. One of the homes (at 1447) was used by the owners as a weekend home, and has not been rented. Further, the City has not obtained any information indicating that either of the remaining homes has be Agenda Item #11.1.

rented to a low or very low income household within the past five years. Nevertheless, in order to ensure compliance with this requirement, a modification to Vesting Tentative Map condition of approval #25 is included in the Vesting Tentative Map resolution to make clear that the affordable housing agreement must comply with all applicable laws, including those related to demolition of existing housing units. In the event that the City determines that one of the existing units was rented to a low or very low income household, the City will require as part of the Affordable Housing Agreement that one of the affordable housing units included as part of the project be sold or rented to a household at an equivalent income level.

Construction Impacts to Neighbors
Vesting Tentative Map condition of approval #38 limits construction hours to 8:00 a.m. to 5:00 p.m. Monday through Saturday to reduce the disturbance to residents in the vicinity. The project would retain 35 existing trees, most of which are mature trees located along the perimeter of the site that would provide screening to adjacent properties. In addition, the project would provide 59 new trees, mostly along the perimeter of the project site to provide enhanced screening in combination with continuous 6-foot-tall wood fencing.

Play Area
Common outdoor recreation spaces are not under review of the City Council and were reviewed for compliance with SHMC Section 17.16.040 as part of the Minor Design Review. The project proposes 18,548 square feet of common open space, which exceeds the 6,150 square foot common open space requirement. Passive and active recreational amenities within the common open space area include walking paths, community seating, a picnic area, and a vineyard trellis with fruiting vines. Per SHMC Section 17.16.040.A.10.f, the play area requirement does not apply to any development that is age-restricted to senior citizens or located within 300 feet of a public park. The project is located less than 300 feet from Mary Fryer Park. Per Vesting Tentative Map condition of approval #13, the project is required to pay all development impact fees, including park impact fees and parkland fees, prior to building permit issuance.

Entry Design
Entry design is not under purview of the City Council and was reviewed for compliance with SHMC Section 17.16.040.C.3 with the Minor Design Review. Each unit contains a separate, primary entry. Per SHMC Section 17.16.040.C.3.b.i, individual ground floor entrances to units within a multifamily development must be visible and directly accessible from the right of way or a shared pathway. This standard is not met for two of the end units in Buildings 1 and 6 where direct pedestrian access is not provided from a walkway, but is accessible from the private street. The City acknowledges strict Agenda Item #11.1.

compliance with this standard would physically preclude development of the project at the densities allowed under State Density Bonus Law and has accepted a Density Bonus waiver requested by the applicant as documented in the Density Bonus Eligibility Letter (Attachment 8 of the January 13, 2026 City Council staff report).

General Plan Consistency
The project implements 2023-2031 Housing Element Site #9. Because the Housing Element is a part of the General Plan, the project is consistent with the General Plan. In addition, while strict compliance with all General Plan policies is not required for development projects, an analysis of consistency with other General Plan policies is included in Attachment 7 of the January 13, 2026 City Council staff report. It should also be noted that the application has been submitted under SB 330, which limits the ability of the City to apply subjective standards to the project. As stated in Chapter 1.6 of the General Plan, the role of the General Plan “sets forth the City’s policies to guide future land use decisions and provides the needed framework to preserve the character and quality of development that the community desires.” Policies that guide development based on character and quality reflecting community desires are commonly subjective in nature.

Concerns that the Project has not demonstrated compliance with the Water Neutrality Policy.

Staff have reviewed the water neutrality calculation and find the most recent version (11/19/2025) to comply with the Policy. The estimated water use, volume to be mitigated and the in-lieu fee to be paid all conform to the Policy guidelines.

The developer has requested use of the in-lieu fee process and has demonstrated that strict compliance with the Policy presents an "unusual hardship". Public Works has reviewed this claim and finds it to be valid for this project. As a result of this finding, the Project has not been required to provide a list of parcels to retrofit, or fixtures lists since these items are not relevant should Council allow payment of an in-lieu fee. Please note that this is not a waiver from these Policy requirements. The Developer will still have to comply with these requirements should Council disallow payment of an in-lieu fee.

Council retains the discretionary right to grant or deny the applicant use of the In-Lieu fee option.

Concerns that the in-lieu fee does not ensure there is adequate water capacity to serve the Project.

The In-Lieu fee is paid by developers when the Council determines that directly offset water use is impractical or poses an unusual hardship. Compliance with the water neutrality policy is separate from an evaluation of existing infrastructure. Developers are subject to separate requirements to prepare calculations showing that they will not Agenda Item #11.1.

exceed the capacity of the service mains to which they will connect. This evaluation is not Citywide nor intended to be. Citywide infrastructure evaluation is performed by the City through periodic studies and development of the Capital Improvement Program (CIP). These studies are made available to all residents as a reference tool for planning their projects big and small.

It is important to note that the Bell Canyon Forecast, periodically presented to Council, is not a measure of "safe yield". The Bell Canyon Forecast is a snapshot of short-term water reserves only and not a long-term planning document. The City has separate "safe yield" studies, the latest of which was completed 2024. Staff began referencing the latest Safe Yield study in decision making following its initial presentation to the ATSC at the July, 2024 meeting.

Spring Grove has adequately demonstrated that the project will not negatively impact public health and safety either on or off the property. Post-Construction storm water management will need to be addressed prior to issuance of building permits, but Staff has determined that there are multiple design options available to the applicant to mitigate this outstanding issue. Staff will ensure that the project incorporates a reasonable solution in the plan prior to approving the Building Permit. Resolution of this matter is not needed for Council to approve the Vesting Tentative Subdivision Map or the In-Lieu Fee request.

Third-Party review of the Spring Grove water neutrality report.

Public Works is responsible for reviewing water neutrality calculations. The water neutrality calculation is not complex, requiring only basic math skills to prepare and review. Third-Party review of this calculation is not a requirement of the municipal code and would not add benefit to the review process.

In-lieu fee only reflects water use for a single year, and should be collected annually.

The in-lieu fee is not based on water use. It is a fee that may be charged in lieu of providing retrofitted toilets or other fixtures to reduce water use, and as a consequence, the fee amount is based on the cost to the developer of conducting such retrofits, were they able to do so. Given that the fee is structured as an alternative to a one time obligation, the fee also should not be recurring. Actual water usage by Spring Grove residents will be paid through water bills like all other water customers.

Concern that the Project will exacerbate water quality issues the City currently experiences.

Taste, odor and discoloration issues with City water have persisted for over 100 years.

These issues are caused by naturally occurring iron and manganese in the City's water sources. The presence of these metals is independent of water demand. Similarly, issues with HAA5 compounds are the result of the interactions between the chemistry of our water sources and the disinfection process. The presence or absence of HAA5 in our water is independent of water demand. The anticipated water demand of the Spring Agenda Item #11.1.

Grove project has no influence on the chemistry of the City's water sources, the treatment processes, or the formation of HAA5 compounds.

Concern of a connection between wastewater treatment cost and Spring Grove.

The wastewater treatment plant has nearly completed installation of a new treatment process. Part of this new process is the continuous removal of solids from wastewater and weekly disposal. The new process does not represent a reduced level of treatment or limit on discharge volume. The level of treatment is substantially higher and creates water safe for irrigating food crops, in part because of new solids removal process. In addition, the treated water meets the requirements for direct discharge to the Napa River and, if discharged directly (not stored in a pond), does not require disinfection with chlorine. Approval of the Spring Grove project, or any project, has no influence on the treatment process or the need to dispose of solids. Cost of disposal is based on volume and additional volume from the Spring Grove Project will be paid by the new residents via their water bill like all other residents.

Fines paid by the City for wastewater discharge violations.

Fines paid by the City are the result of the old treatment process combined with severe weather events and delays in delivery of the new treatment process. Now that the new treatment process is fully operational, future violations should be significantly reduced or eliminated. The Spring Grove development has no impact on this issue.

Alternative Water Neutrality and In-Lieu Fee calculations.

Ms. Covell has issued multiple comment letters regarding the accuracy and lawfulness of the Water Neutrality and In-lieu Fee calculations prepared by the Spring Grove development team. Ms. Covell has prepared alternative calculations and cited the calculations prepared by other projects to support her comments. Some of Ms. Covell's comments were erroneously based on early Spring Grove submittals which had already been rejected by Staff. These early comments have been superseded by Ms. Covell's comments dated 1/9/2026. Staff have prepared responses to the 1/9/2026 comments only.

1. The Policy relies on an occupancy/bedroom metric for estimating water usage in residential development. This metric allows comparison of the estimated maximum number of people who could reasonably occupy the development.

Ms. Covell’s comments reference a “per home” metric that does not consider the number of people who could live in the home. This metric is not supported by the policy and is less accurate. Public Works has relied on the Policy for evaluating the Spring Grove project and finds that the calculation prepared by Spring Grove meets the requirements of the Policy.

2. The Policy Guideline includes a “typical calculation” table for use by developers. The table is provided as a formatting reference for preparing their calculation. The table also includes the estimated water usage for each Agenda Item #11.1.

standard fixture or appliance. The water usage figures are based on the 2010 LEED standards in effect at the time of the Policy’s adoption. These Policy standards have not been updated. However, they remain mostly consistent with the 2022 California Green Building Standard applicable to the project today. Public Works accepts the Policy Guideline standards without additional documentation.

3. Some developers achieve water neutrality by specifying hyper-efficient appliances that exceed the Policy standards. Public Works allows this method but requires that the developer document the claimed efficiency by providing the specification sheet for the specific appliance. The appliance is then incorporated into the building permit so that the Building Official or Public Works can verify its installation before final inspection. Spring Grove did not use this option and elected to incorporate conventional water efficient fixtures/appliances into the Water Neutrality and In-Lieu Fee calculations for the project. This approach is in strict conformance with the Policy.

4. Ms. Covell asserts that the developer did not use the calculation method required by the Policy. The Developer did use unique formatting, but their methodology adheres to the Policy. Public Works performed its own calculation, which incorporates all decimal values required by the Policy, during review and found that the developer correctly estimated daily water usage for the anticipated 320 residents at the property. The City's independent calculation also established that Ms. Covell's concerns regarding miscalculation, blended averages, or potential "rounding errors" are immaterial.

5. Ms. Covell asserts that the calculation for “kitchen faucet” should be 1.8 gallons per minute. Ms. Covell's value relies on alternative standards. While these alternative standards are acceptable to Public Works, they remain optional. The Policy allows 1.5 gallons per minute for this fixture, so the developer’s calculation is acceptable.

6. Ms. Covell asserts that the water “savings” calculation, prepared for the Hunter Subdivision is a more accurate method for calculating the in-lieu fee.

Public Works disagrees. The Hunter Subdivision calculation omits dishwasher or clothes washer use. The Spring Grove calculation includes these appliances and is the more complete methodology. The Hunter Subdivision also relies on well water as a compliance method and is significantly larger than the Spring Grove project. Any comparison between these two projects is not apples-to-apples.

Agenda Item #11.1.

7. Ms. Covell asserts that the "baseline" (i.e. "water budget") water calculation should use the average number of bedrooms from the 3 (three) original homes. This assertion has no basis in either the Policy or any outside standard applicable to the Policy. The Policy relies on a formula which allows 150 GPD for each existing bedroom. Public Works confirmed, via County tax records, that the 3 (three) original homes have a combined total of 9 bedrooms which equates to a "baseline" water usage of 1,350 GPD. Public Works finds that the "baseline" calculation prepared by Spring Grove conforms to the Policy.

8. Ms. Covell expressed concern that dens/home offices/optional bedrooms had been omitted from the Spring Grove calculations. Public Works had expressed similar concerns over early drafts of the floor plans and the purpose of an additional room located adjacent to the garage of some units.

The plan was revised to clearly identify these rooms and the Water Neutrality calculations updated. Public Works finds that all bedrooms applicable to the Policy have been correctly identified and occupancy correctly calculated.

Concern that the City has not made the findings needed to approve Spring Grove.

At the December 2, 2025 regularly scheduled Planning Commission meeting, the Planning Commission held a public hearing and adopted findings and a recommendation for approval to City Council pursuant to SHMC Section 16.16.060 (Attachment 3 of the January 13 City Council staff report).

SHMC Section 16.16.080(A) includes the findings that must be made by City Council to approve a Tentative Map. While there are a number of concerns that have been expressed by the community, a Tentative Map may only be denied if it is found to be inconsistent with the findings of SHMC Section 16.16.080(A) of the St Helena Municipal Code. Applicable findings are included in the January 13, 2025 City Council staff report and the draft City Council resolution for the Vesting Tentative Map (Attachment 2 of the January 13 City Council staff report).

Other Comments from Councilmembers at January 13 CC Hearing What will prevent on-street parking on private streets? Agenda Item #11.1.

The project proposes 93 off-street parking spaces, which exceeds the 77 parking space minimum requirement per SHMC Table 17.26.040(A). No on-street parking is proposed within the private streets. The project includes a combination of two-car garages for each unit and 11 perpendicular parking stalls. Per Vesting Tentative Map condition of approval #19, the project must include provide wayfinding signs along the project’s internal roadway and at the sidewalk connections, and identity a short-term parking area for delivery vehicles. In addition, the frequency of driveway curb cuts will discourage any unauthorized on-street parking. The applicant has confirmed the HOA and CC&R’s will prohibit on-street parking within the development.

Can the City impose a condition of approval that would require partial retrofits, as demanded, and partial payment of the in-lieu fee for the balance of the required offset?
Yes, the City Council and applicant may consider a condition of approval to require some retrofits and a partial payment of the in-lieu retrofit fee to the extent the applicant is able to find existing demand for retrofits within the community. However, after discussing this with the applicant, given the challenges of finding demand (e.g. survey) and the uncertainty of how the City would enforce such a condition, the applicant indicated they would not be amenable to a partial retrofit.

Agenda Item #11.1.

1 Spring Grove Key Issues FAQs
Issue Response
Concern that the project has not
been processed in compliance
with the St. Helena Municipal
Code.

The project underwent multiple rounds of consistency
analysis before the project was determined to be
consistent with all applicable state and local standards.

Comment letters are posted to the project webpage on
the City website here:
https://www.cityofsthelena.gov/797/PL24-065-1447-
1505-1515-Spring-St--Sprin
Concern that the Applicant has
not satisfied all requirements of
the Municipal Code.

Some provisions of the Municipal Code become
applicable to the project at Building Permit and Final
Inspection stages of the development review process.

For example, during the consideration of a Tentative
Map application, Public Works staff reviews a
preliminary stormwater management plan, and more
detailed analysis of construction drawings is performed by Public Works staff during the building permit review (see condition of approval #54 in Attachment 2). Any
outstanding issues will be resolved either before
Building Permit issuance or Final Inspection, as
appropriate.

Site Layout and Irregular Lot While the resultant parcel is irregularly shaped, it exceeds the minimum dimensional requirements for lot
area and lot width per SHMC Table 17.16.030 (see
Attachment 5).

Fire Access The Vesting Tentative Map (Attachment 5) proposes a minimum 26-foot-wide Emergency Vehicle Access
Easement (28-feet-wide in most places), which meets
the minimum 26-foot-wide fire access width standard
pursuant to California Fire Code Appendix D Section
D105.2. Fire has verified aerial equipment can access
the third story windows and turning radii are adequate.

Lastly, the project is conditioned to install NFPA13 fire sprinklers in each unit prior to building permit issuance (see condition of approval #72 in Attachment 2).

Height Proposed building elevations are not under purview of the City Council and were reviewed and found to be in
compliance with all applicable zoning standards as part of the Minor Design Review (Attachment 8).

Agenda Item #11.1.

2 Number of Affordable Units The location of the eight affordable units are shown on the plans (see Attachment 5, sheet A-1). The project
is conditioned to pay the housing in-lieu fee for the
remaining 0.2 fractional units of affordable housing to achieve the 20% minimum requirement of the
Inclusionary Housing Ordinance (see condition of
approval #26 in Attachment 2).

Architecture Project architecture is not under City Council review and was reviewed and approved with the Minor Design
Review (Attachment 8).

Occupancy of Existing Homes State law also requires that any “protected units” that are demolished must be replaced with an equal
number of “protected units.” While the City has not
obtained any information indicating that any of the
remaining homes have been rented to a low or very low
income household within the past five years, a
modification to condition of approval #25 is included in Attachment 2 to ensure the affordable housing
agreement complies with all applicable laws, including
those related to demolition of existing units.

Therefore, there is no legal requirement to have the
unit status resolved prior to approval of the
Subdivision Map.

Construction Impacts to
Neighbors
Condition of approval #38 limits construction hours to
8:00 a.m. to 5:00 p.m. Monday through Saturday to
reduce the disturbance to residents in the vicinity
(Attachment 2). The project would propose 59 new
trees and retain 35 existing trees, most of which are
located along the perimeter of the site to screen
adjacent properties (Attachment 5).

Play Area Common outdoor recreation spaces are not under review of the City Council and were reviewed for
compliance with SHMC Section 17.16.040 as part of
the Minor Design Review (Attachment 8).

Entry Design Entry design is not under purview of the City Council and was reviewed for compliance with SHMC Section

17.16.040.C.3 with the Minor Design Review
(Attachment 8).

General Plan Consistency The project implements 2023-2031 Housing Element Site #9. Because the Housing Element is a part of the
General Plan, the project is consistent with the General Plan. In addition, while strict compliance with all
General Plan policies is not required for development
projects, an analysis of consistency with General Plan
policies is included in Attachment 7.

Water Neutrality Policy Public Works has reviewed the water neutrality Agenda Item #11.1.

3 Compliance calculation in Attachment 11 and finds it complies with the Policy. The estimated water use, volume to be
mitigated, and in-lieu fee to be paid all conform to the Policy guidelines.

Concerns that the in-lieu fee
does not ensure there is
adequate water capacity to
serve the Project.

The In-Lieu fee is paid by developers when the Council
determines that directly offset water use is impractical or poses an unusual hardship pursuant to SHMC
Section 13.12.030. Compliance with the water
neutrality policy is separate from an evaluation of
existing infrastructure.

Third-Party review of the Spring
Grove water neutrality report.

Public Works is responsible for reviewing water
neutrality calculations pursuant to SHMC 13.12.030.

Public Works’ analysis is included in the Staff Report, pages 12-13.

In-lieu fee only reflects water
use for a single year, and should
be collected annually.

The in-lieu fee is not based on water use. It is a fee
that may be charged in lieu of providing retrofitted
toilets or other fixtures to reduce water use pursuant to SHMC Section 13.12.030. Therefore, the fee amount
is based on the cost to the developer of conducting
such retrofits, were they able to do so.

Concern that the Project will
exacerbate water quality issues
the City currently experiences.

Taste, odor and discoloration issues with City water
have persisted for over 100 years. These issues are
caused by naturally occurring iron and manganese in
the City's water sources. The presence of these metals
is independent of water demand of individual
developments.

Concern of a connection
between wastewater treatment
cost and Spring Grove.

The wastewater treatment plant has nearly completed
installation of a new treatment process. Part of this new process is the continuous removal of solids from
wastewater and weekly disposal. Approval of the
Spring Grove project, or any project, has no influence
on the treatment process or the need to dispose of
solids. Cost of disposal is based on volume and
additional volume from the Spring Grove Project will be paid by the new residents via their water bill like all other residents.

Fines paid by the City for
wastewater discharge
violations.

Now that the new treatment process is fully operational, future violations should be significantly reduced or
eliminated. The Spring Grove development has no
impact on this issue.

Alternative Water Neutrality and
In-Lieu Fee calculations.

See Attachment 12 for Public Works’ response to
concerns regarding alternative fee calculations.

Concern that the City has not
made the findings needed to
approve Spring Grove.

Applicable findings are included in the Staff Report,
pages 10-11 as well as the draft City Council resolution for the Vesting Tentative Map (Attachment 2).

What will prevent on-street The project proposes 93 off-street parking spaces Agenda Item #11.1.

4 parking on private streets? (Attachment 5), which exceeds the 77 parking space minimum requirement per SHMC Table 17.26.040(A).

The applicant has confirmed the HOA and CC&R’s will
prohibit on-street parking within the development.

How does AB 130 apply to this
project?
If there is no action by the City Council to approve or deny the project by March 10, 2026, the project is
deemed approved.

Agenda Item #11.1.

83578.00132\44619133.1
Ethan J. Walsh
(916) 551-2825
[email protected]
500 Capitol Mall. Suite 2500.Sacramento, CA 95814
Phone: (916) 325-4000 | Fax: (916) 325-4010
bbklaw.com
Memorandum

VIA E-MAIL
To: Mayor and City Council
City of St. Helena
File No.: 83578.00132
From: Ethan J. Walsh
Date: January 27, 2026
Re: Response to Comments from DP&F Law dated January 23, 2026 and January 26, 2026
This memo responds to the comment letter submitted by Joshua Devore of DP&F Law on behalf of Julie Dickson on January 23, 2026 and the follow-up email from Mr. Devore to Councilmember Barak dated 26, 2026 related to the application from City Ventures for a vesting tentative subdivision map and to pay a water neutrality in lieu retrofit fee for the Spring Grove housing development. The City Attorney’s office prepared this memo in collaboration with City staff.

1. Standard of Review
At the outset of the letter, Mr. Devore correctly notes that in order for the City Council to disapprove or reduce the density of the Spring Grove housing development, the City Council must find that the project will have a “specific, adverse impact upon the public health or safety” of the community and there is no feasible method to safisfactorily mitigate or avoid the adverse impact other than to disapprove or reduce the density of the project. (Gov’t Code §65589.5(j)(1).) A “specific adverse impact” is defined as a significant, quantifiable, direct, and unavoidable impact, based on objective, identified written public health or safety standards, policies, or conditions as they existed on the date the application was deemed complete. (Id. §65589.5(j)(1)(A).) However, it is also important to note that a Council decision to disapprove the project based on a specific, adverse impact upon public health and safety would need to be based on writing findings supported by a preponderance of the evidence on the record. (Id. §65589.5(j)(1).) This means that the evidence in the record finding that there is a specific adverse impact upon public health or safety needs to outweigh the evidence in the record to the contrary.

Agenda Item #11.1.

83578.00132\44619133.1
- 2 -
Best Best & Krieger LLP

2. Design Review Application
In both the January 23 and 26 correspondence, Mr. Devore argues that the project does not comply with the City’s objective design standards and should have been subject to a Major Design Review under the City’s code. In support of this assertion, he points to the correspondence from the City confirming density bonus eligibility and the waivers granted by the City pursuant to State Density Bonus Law (“SDBL”). (Staff Report, Attachment 8A.) He asserts that because the project, with the waivers included, does not comply with the City’s objective standards, it should have been subject to Major Design Review.

Staff Response: First, it is important to note that the Design Review application is not under consideration by the Council. The application was reviewed by the Community Development Director in accordance with the City’s Zoning Code, and a Notice of Action was issued. That action was not appealed, and the action is final. That is not a part of the action under consideration by the Council.

Nevertheless, it was appropriate to find that the project is consistent with the City’s objective standards, notwithstanding the request for waivers under the SDBL. A housing development where at least ten percent of the dwelling units are sold to persons and families of moderate income is eligible for a density bonus, incentive and waivers in accordance with the SDBL. (Gov’t Code §65915(b)(1)(E).) This project meets those requirements and is therefore eligible for waivers that may be granted under the SDBL. Further, as noted in the January 13, staff report, the Housing Accountability Act specifically states that State Density Bonus eligibility or density bonus incentives, concessions, or waivers do not provide a valid basis to find that a qualifying housing development project is inconsistent or not in compliance with applicable standards. (Gov’t Code §65589.5(j)(3).) In other words, the City staff must determine whether the project is consistent with the City’s objective standards, except those that have been waived as is permitted under the SDBL. The City staff appropriately made this finding as part of the completed design review process.

Additionally, in the January 26 email, Mr. Devore alleges that the City has decided to waive fire and building code provisions. That is incorrect. The waivers granted are listed in Attachment 8A to the Staff Report, and they are all waivers of requirements set forth in the City’s Zoning Code.

3. Garage Setback
The letter notes that the project does not meet the required setback distance between garage doors and the street and there are not individual unit driveways and states that there is nowhere for a car to stop out of the lane of travel if the garage door is not open. This means that there is zero parking adjacent to any of the units; and not even any stopping without blocking the street.

Staff Response: Under the SDBL, the City is required to grant waivers to development standards that will have the effect of physically precluding construction of the development, unless the waiver would have a specific adverse impact to public health or safety, under the standard described above. (Gov’t Code §65915(e).) For each of the eleven waivers granted as documented in Attachment 8A, the City has provided a response as a basis for granting the waiver which was published both as an attachment to the December 2, 2025 Planning Commission staff report as Agenda Item #11.1.

83578.00132\44619133.1
- 3 -
Best Best & Krieger LLP
well as the City Council staff reports dated January 13th and 27th, 2026. The referenced deviation from the zoning code is one of the granted waivers.

The requirement regarding minimum distance from a drive aisle to a garage (SHMC §17.16.040(A)(7)(e)), this is an urban design standard within the zoning code intended to reduce the visual impacts of “snout houses,” those with protruding garages that visually dominate the street frontage. As noted in Attachment 8A, the project qualifies for a waiver from the minimum distance to interior drive aisle as strict compliance with this standard would physically preclude development of the project at the densities allowed under State Density Bonus Law. Further, as described in Staff Report Attachment 12, the project proposes 93 off-street parking spaces, which exceeds the 77 parking space minimum requirement per SHMC Table 17.26.040(A). No on-street parking is proposed within the private streets. The project includes a combination of two-car garages for each unit and 11 perpendicular off-street, surface parking stalls. Per Vesting Tentative Map condition of approval #19, the project must provide wayfinding signs along the project’s internal roadway and at the sidewalk connections, and identity a short-term parking area for delivery vehicles. In addition, the frequency of driveway curb cuts will discourage any unauthorized on-street parking. The applicant has confirmed the HOA and CC&R’s will prohibit on-street parking within the development.

4. Egress Requirements
The letter asserts that the building code mandates egress routes to the public way, citing California Building Code sections 1022.2.2 and 1028.5, and asserts that the units do not meet this requirement because the homes do not have egress doors faced toward the street. The letter also asserts that for many of the units the only doors are directly adjacent to “potential fire corridors of the creek” adjacent to the project. The letter asserts that this is a critical safety failure.

Staff Response: The project did include a waiver to the required entry design under the Zoning Code Section 17.16.040(C)(3)(b)(i). This is an urban design standard within the zoning code intended to provide convenient pedestrian access while enhancing building aesthetics as viewed from the public right of way. As described in Attachment 8A of the January 27, 2026 City Council Staff Report, this standard is not met for two of the end units in Buildings 1 and 6 where direct pedestrian access is not provided from a walkway. The City acknowledges strict compliance with this standard would physically preclude development of the project at the densities allowed under State Density Bonus Law.

This waiver does not violate the California Building Code. The Building Official reviewed this comment and noted that the code sections cited out of the California Building Code generally apply to commercial buildings where we are expecting high occupant loads. This project is governed by the California Residential Code as it fits into the scope of R101.2 CRC. Even if the City did apply Chapter 10 CBC requirements to this project, only one exit is required for rooms or spaces with 49 persons (or less) and a travel distance of not more than 200’ for sprinklered buildings. It would be reasonable to expect that these dwelling units do not exceed a 200’ travel distance from anywhere within the unit to the exterior, and we do not anticipate more than 5 or 6 persons (probably less) per unit. It is reasonable to conclude that occupants will be able to exit out of the main entry door, the balcony door on the second floor, or a bedroom egress window. Section 1022 &1028.5 CBC does not apply to this project, however, if it did, the City is not aware of any Agenda Item #11.1.

83578.00132\44619133.1
- 4 -
Best Best & Krieger LLP
obstructions that would prevent an individual from accessing the public way. Also, Chapter 2 of the CBC defines “Public Way”; (it is not necessarily a public street such as Spring Street that qualifies as a public way.

Further, in response to the characterization of the adjacent creek as a “fire corridor”, City Public Works staff noted that the creek adjacent to the project is part of the City's hybrid municipal separate storm sewer system (MS4). The portion of the MS4 located immediately upstream of the development is a concrete channel, of which a significant portion is fully underground. Because of the interrupted surface exposure, the MS4 lacks a contiguous fuel corridor to the wildland urban interface. Downstream of the development, the MS4 opens to a short string of manmade ponds located adjacent to Mitchell Drive. These ponds form a natural fuel break from its outfall at Sulphur Creek. Additionally, the Spring Grove parcel is located about 2,600 feet from the nearest Fire Hazard Severity Zone. Therefore the characterization of this area (without supporting evidence) as a fire corridor is inaccurate.

5. Size of Units
The January 26 letter asserts that the developer sought to maximize the size of the units and that the minor design review waived numerous requirements on the basis that such adjustments were necessary to build the number of units proposed; but it cites no evidence that the units could not have been made smaller and met the objective standards without waivers.

Staff Response: The City cannot condition the granting of waivers on a reduction in the size of the proposed residential units. The size of the proposed units is consistent with the Zoning Code.

Under the SDBL, the City is required to grant requested waivers if failure to grant the waivers would physically preclude the construction of the project as it is designed. (Bankers Hill 150 v.

City of San Diego (2022) 74 Cal.App.5th 755, 775; Wollmer v. City of Berkeley (2011) 193 Cal.App.4th 1329, 1347.) The Courts have clearly held that a City cannot condition waivers on adjustments to the design of the project (such as the reduction of unit) size. The City may only deny a waiver request if it (1) would not physically preclude the construction of the project as designed, or (2) it would have a specific adverse impact to public health or safety, and for which there is no feasible method to satisfactorily mitigate or avoid the specific adverse impact or “would have an adverse impact on any real property that is listed in the California Register of Historical Resources, or to grant any waiver or reduction that would be contrary to state or federal law. (Gov’t Code §65915(e)(1).) The City could not require that the units be smaller as a condition of granting a waiver.

6. Water Supply
The January 23 letter asserts that the City has not demonstrated capacity to expand water service to an additional 40+ units, and questions whether there is substantial evidence in the record that the City’s water neutrality policy is valid or that there is sufficient water to serve the project. The letter also asserts that the in-lieu fee would appear to be arbitrary and not supported by evidence in the record. On this same topic, the January 26 email refers to an “acknowledgement that the City is out of compliance with its water supply Safe Yield”, and that “there appear to be unaddressed questions as to whether the City is exceeding the sustainable yield of the Agenda Item #11.1.

83578.00132\44619133.1
- 5 -
Best Best & Krieger LLP
groundwater basin, potentially violating the Sustainable Groundwater Management Act (SGMA) by appropriating groundwater from people who have the rights to it, and creating a larger concern of limiting the groundwater resources over time.” The email also assertsthat there is no Water Availability Analysis provided with the Staff Report. Under Napa County’s current implementation of SGMA for the Napa Valley Subbasin, the parcel is entitled to only 0.3 acrefeet of water per acre under the Groundwater Sustainability Plan. The 2.53 acre parcel is entitled to only 0.76 acre feet of water – but the proposed water use is more than 12.5 acre feet.

Staff Response: First, as noted in the Staff Report, this project is not required to do a Water Availability Analysis due to its size. This project is contemplated in the City’s General Plan and Zoning Code for residential use and as stated previously, City staff does believe that the City has capacity to serve this project. The City is in the process of updating its safe yield estimate and based on the analysis conducted by the City’s engineering consultants, the City is currently within the existing safe yield as project in the City’s draft report.

With regard to the allegations related to the water neutrality requirement, the City is applying the water neutrality requirement in conformance with the City Code, which grants the option of accepting an in-lieu fee rather than retrofitting units. The City’s water neutrality policy provides the method of calculating the in-lieu fee, which was in place at the time the application for this project was received. City staff do not believe that the Water Neutrality Ordinance or Policy, or the methodology for calculating the in-lieu fee are arbitrary. If they were, however, the result would be that the Water Neutrality requirements would be stricken from the Code, and this project would not be subject to water neutrality requirements. The City disagrees with this position and does believe that the project is subject to water neutrality requirements.

Finally, the references to SGMA do not apply to this project. The Sustainable Groundwater Management Act is a statewide law that creates a framework to ensure long-term sustainable management of groundwater resources. SGMA does not include specific limitations on water use. Further, the standards referenced in the email (that this parcel is entitled to only 0.76 acre feet of water) refer to the standards that Napa County uses when issuing permits for a new well or improvements to an existing well. This project is connecting to City water and is not drilling a new well or improving an existing well. These standards therefore do not apply.

7. Water Quality
The letter asserts that the City has “failed in quality of its water”, citing concerns about discolored and malodorous water and the City’s disclosure that haloacetic acid levels in the water supply exceeded state drinking water standards.

Staff Response: City staff does acknowledge that color and odor issues with City water do arise periodically, but these are the result of minerals naturally occurring in some of the City's water sources. The presence of these minerals is entirely independent of water demand and would occur regardless of any development. The proposed Spring Grove development has no effect on the severity of this issue. Further, the haloacetic acid (HAA5) compounds are formed when organic matter reacts with chlorine. The City uses chlorine in its disinfection process. The presence of organic matter in the City's water sources is entirely independent of demand and Agenda Item #11.1.

83578.00132\44619133.1
- 6 -
Best Best & Krieger LLP
would occur regardless of any development. The proposed Spring Grove development has no effect on the severity of this issue.

8. Sewer System
The letter cites recent discharge violations at the City’s wastewater treatment plant and suggests that these violations indicate that the City cannot support additional development.

The letter also asserts that based on a recent budget amendment to increase sludge hauling due to higher than anticipated operational needs, the City wastewater treatment system does not have capacity to support this project.

Staff Response: Discharge exceedances over the last two years are the result of severe storm events which occurred during construction of a new treatment process. Storm severity is entirely independent of sewage generation, and the exceedances would have occurred regardless of any development. The proposed Spring Grove development has no effect on the severity of this issue.

Additionally, the need to haul additional sludge does not indicate a lack of capacity in the treatment facility. Solids disposal is not a new process or expense for the WWTP. The old process retained solids in the treatment ponds, and these would have to be removed with excavation equipment at great, but infrequent, expense. The new treatment process relies, in part, on the continuous removal of solids which occur semi-automatically. This results in a more frequent disposal cycle but requires no excavation or interruption to operations. The recent budget amendments to accommodate the disposal of these solids are a direct result in the change in treatment process and would have occurred regardless of any development.

9. CEQA Concerns
The letter asserts that there is a reasonable possibility that the project will have a significant effect on the environment, and therefore the project is not actually exempt from CEQA, citing CEQA Guidelines Section 15300.2(c).

Staff Response: As detailed in the Staff Report for the project, the City has determined that the project is statutorily exempt from CEQA review under Public Resources Code section 21080.66. This statutory exemption is not subject to the provisions of CEQA Guidelines section 15300.2, including those provisions regarding unusual circumstances. Statutory exemptions, such as Public Resources Code section 21080.66, are enacted to lift the burden of environmental review from specified classes of projects. The sole question is whether the project fits within the language of the exemption. (Napa Valley Wine Train, Inc. v. Public Utilities Commission (1990) 50 Cal.3d 370.) As substantial evidence in the record for the Project confirms that the Project fits within the elements of Public Resource Code section 21080.66, no CEQA review is required.

Agenda Item #11.1.

83578.00132\44619133.1
Ethan J. Walsh
(916) 551-2825
[email protected]
500 Capitol Mall. Suite 2500.Sacramento, CA 95814
Phone: (916) 325-4000 | Fax: (916) 325-4010
bbklaw.com
Memorandum

VIA E-MAIL
To: Mayor and City Council
City of St. Helena
File No.: 83578.00132
From: Ethan J. Walsh
Date: February 5, 2026
Re: Response to Comments from Mark Wolfe dated January 27, 2026 Introduction
On January 27, 2026, the city received a comment letter from Mark Wolfe regarding the Spring Grove development. The letter focused in large part on site access and compliance with the St. Helena Municipal Code, stating that with only one point of access to the site, the Project does not comply with the local implementation of the California Fire Code.

For the reasons outlined below, the local amendments to the 2022 California Fire Code that Mr. Wolfe states the Project does not comply with are no longer in effect. Furthermore, even if these local amendments were still in effect, Mr. Wolfe is misinterpreting the relevant SHMC section and the Project would still be compliant with the 2022 Fire Code as it was amended.

Therefore, staff respectfully disagrees with Mr. Wolfe’s analysis of this issue and is satisfied that the Project complies with all applicable laws by providing a single access route and including fire sprinklers throughout the project.

Background & Analysis

1. Relevant Attributes of the Spring Grove Project
The Project is a 41-unit multi-family development. Through its Vesting Tentative Map, the Project is designed to provide an Emergency Vehicle Access Easement that meets the statemandated minimum width of 26’, though is 28’ wide in most places along the easement.

Furthermore, the project will provide NFPA13 rated fire sprinklers in each unit.

Agenda Item #11.1.

83578.00132\44619133.1
- 2 -
Best Best & Krieger LLP

2. Applicability of the 2025 California Building Standards Code and the 2025 California Fire Code
The California Fire Code, a part of the State Building Standards Code, is a triennially adopted regulatory scheme adopted by the California Building Standards Commission to govern fire safety issues throughout the state. While these codes are adopted by the state, Health and Safety Code § 17960 requires that local jurisdictions enforce the State Building Standards Code.

The California Building Standards Commission most recently updated the State Building Standards Code in 2025. This updated was published by the Commission on July 1, 2025 and—pursuant to Health and Safety Code Section 17958—took effect on January 1, 2026.

While the Spring Grove project is subject to vesting due to the submission of a Preliminary Application by the applicant, that vesting does not apply to the California Building Standards Code because Government Code Section 65589.5(o) limits vesting to local regulations, not state standards. Per Health and Safety Code Section 18938.5(a), only those California Building Standards in effect at the time of building permit application “shall apply to the plans and specifications for, and to the construction performed under, that building permit.”
Therefore, the Project will have to comply with the California Building Standards Code in effect at the time building permits are sought for the project, presumably that will be the 2025 California Building Standards Code.

3. Local Modifications of the California Building Standards Code Health and Safety Code Section 17958(a) allows for local modifications of the California Building Standards Code. Health and Safety Code Section 17958(b) restricts the scope and purposes for which local amendments may be adopted where they apply to residential units.

Critically for purposes of the analysis of which codes apply to the Project, local amendments only apply to the currently adopted California Building Standards Code. In other words, local amendments to the 2022 California Building Standards Code are no longer effective as of the effective date of the 2025 California Building Standards Code.

City Council introduced local modifications to the 2025 California Building Standards Code at the January 13 Council meeting but those local modifications have not yet been adopted by the City. The city’s currently adopted local modifications were applicable to the 2022 California Building Standards Code, which ceased being effective on December 31, 2025. Therefore, none of the City’s local amendments to the California Building Standards Code are currently in effect for any project seeking building permits at this time.

Agenda Item #11.1.

83578.00132\44619133.1
- 3 -
Best Best & Krieger LLP

4. 2025 California Fire Code Sections on Multi-Family Access The 2025 California Fire Code sections relevant to access routes to multi-family developments is found in Appendix D, Section 106, which reads as follows:

D106.1 Projects having more than 100 dwelling units. Multiple-family residential projects having more than 100 dwelling units shall be equipped throughout with two separate and approved fire apparatus access roads.

Exception: Projects having up to 200 dwelling units shall have not fewer than one approved fire apparatus access road where all buildings, including nonresidential occupancies, are equipped throughout with approved automatic sprinkler systems installed in accordance with Section 903.3.1.1 or 903.3.1.2.

D106.2 Projects having more than 200 dwelling units. Multiple-family residential projects having more than 200 dwelling units shall be provided with two separate and approved fire apparatus access roads regardless of whether they are equipped with an approved automatic sprinkler system.

D106.3 Remoteness. Where two fire apparatus access roads are required, they shall be placed a distance apart equal to not less than one-half of the length of the maximum overall diagonal dimension of the property or area to be served, measured in a straight line between accesses.

As discussed above, the 2025 California Fire Code applies to this Project. Since the Project only proposes to develop 41 multi-family units, the 100 dwelling unit threshold for providing “two separate and approved fire apparatus access roads” required by Section D106.1 is not met and the Project is not obligated to provide more than one access route.

Furthermore, Section D106.1 provides an exception to this multiple access road requirement for projects with more than 100 dwelling units but less than 201 dwelling units where those units are equipped with approved fire sprinkler systems. Here, the Project is providing such a sprinkler systems and therefore could have up to 200 units without having to provide a secondary access road.

5. Additional Considerations
For the reasons discussed above, the city’s local amendments to the California Building Standards Code are not currently in effect. However, even if the city’s prior local amendments were still in effect, the project would nevertheless be compliant for several reasons.

Agenda Item #11.1.

83578.00132\44619133.1
- 4 -
Best Best & Krieger LLP
a. Only Modified Language is Included in the SHMC for California Building Standard Code Adoption
First, Mr. Wolfe misreads how the city adopts local amendments to the California Building Standards Code. Mr. Wolfe points to SHMC Section 15.36.100-D, which reads as follows:

Section D107.1, One or two-family dwelling residential developments, California Fire Code 2022, Appendix B, is amended to read as follows:

Developments of one, two-family or multi family dwellings where the number of dwelling units exceeds thirty (30) shall be provided with two separate and approved fire apparatus access roads, and shall meet the requirements of Section D104.3.

Mr. Wolfe makes much of the fact that the Exception that appears in the California Fire Code’s Section D107.1 allowing for fire sprinklers in lieu of a secondary access route is not included in SHMC Section 15.346.100-D. However, City staff has confirmed that it is the city’s routine practice to only list language actually being modified when listing the local amendments. Put another way, the City’s local amendments only expressly include exceptions where the exceptions themselves are being modified in some fashion.

In this instance, the modification to Section D107.1 is the addition of “or multi family” to the list of types of structures subject to Section D107.1. Nothing in the ordinance or staff reports that adopted this local modification as far back as 2007 when it was first included in the municipal code made any mention of a deliberate intention to delete the exception to Section D107.1.
b. Section D106.1 was Not Modified in the SHMC for the 2022 Fire Code Adoption When adopting the 2022 California Fire Code, the City did not modify the provisions of Section D106.1 (regulating access to multifamily developments). Section D106.1 remained functionally unchanged from the 2022 California Fire Code to the 2025 California Fire Code.

Since the project is a multifamily development, it is subject to Section D106.1 for the reasons outlined above. As also discussed above, it complies with the requirements prescribed by D106.1 for larger multifamily developments even though the Project is not large enough to actually trigger the secondary access requirement.

It should also be noted that because Section D106.1 governs multifamily projects, in preparing the local modifications to the 2025 Building Standards Code adoption and local modifications, staff intends to strike the local modification to Section D107.1 adding “or multi family” as unnecessarily duplicative.

Agenda Item #11.1.

83578.00132\44619133.1
- 5 -
Best Best & Krieger LLP
c. Interpreting the Exception to D107.1 as Being Removed Runs Counter to the City’s Housing Element and Zoning
Site constraints preclude a secondary access to the Project without the developer acquiring additional land from other private property owners. The zoning for the site allows for a density between 37 and 68 units on the Site, and the Housing Element contemplated that 48 units would be developed on the site. If the City were to interpret its code to require secondary access for the Project, it would mean that the Site could not be developed at the current minimum density, nor could it be developed consistent with the Housing Element, unless secondary access were acquired. We believe it is appropriate to interpret the City’s local amendments to the Fire Code consistent with the City’s other planning documents, and that staff’s interpretation of the local amendments is appropriate. d. Time to Raise Objections to Objective Standards Has Run The Project is subject to streamlining under the Housing Accountability Act. Under that streamlining, the City must raise objections to a project’s lack of compliance with objective standards within 30 days of the date the application for the Project is determined to be complete. (Gov’t Code §65589.5(j)(2)(A)(i)1.) The City did not notify the applicant of a requirement that secondary access would be required, because City staff believed that the Exception in Section D107.1 applied. The time to raise any objection based on a failure to comply with objective standard expired months ago and there is no longer a basis to reject the project based on an objective standard under the HAA.

Conclusion
For the reasons outlined above, the Project complies with all applicable fire access requirements set forth in the 2025 California Fire Code. While the City’s local amendments are currently not in effect, even if the City’s formerly adopted local amendments to the California Fire Code were in effect, the Project would nevertheless be compliant with the access requirements that were set forth in those amendments. 1 This timeline applies to projects of 150 units or less. For larger projects, the City has 60 days to object.

Agenda Item #11.1.

3 rd PARTY PLAN REVIEW
PHASE 1
Spring Grove
1447, 1505, and 1515 Spring Street St. Helena, CA
January 26, 2026
© 2026ALL RIGHTS RESERVED.

Zari Consulting Group, Inc | Fire Protection Engineers | Building Code Experts 755 Baywood Drive, 2nd Floor | Petaluma, CA | 94954
t +1.888.779.3397 | ZariCode.com
Agenda Item #11.1.

3 rd Party Plan Review
Spring Grove
1January 26, 2026
ZCG No. 0126009

TABLE OF CONTENTS
Contents
INTRODUCTION 2
PROPOSED SITE DESCRIPTION 2
FIRE HAZARD HISTORY 4
ACCESS AND SECONDARY ENTRANCE 5
TURNAROUNDS AND STAGING 5
FIRE HOSE TRAVEL DISTANCE AND REACH 6
AERIAL APPARATUS ACCESS 7
FIRE FLOW AND WATER SUPPLY 7
VEGETATION MANAGEMENT PLAN 7
FIRE SERVICE FEATURES 8
CONCLUSION 9
Agenda Item #11.1.

3 rd Party Plan Review
Spring Grove
2January 26, 2026
ZCG No. 0126009

INTRODUCTION
This report provides a third-party fire protection engineering review of the proposed "Spring Grove" residential subdivision. The analysis, findings, and technical recommendations are based upon the California Fire Code (CFC), relevant NFPA (National Fire Protection Association) general requirements, and established fire protection engineering best practices. Furthermore, this review evaluates the project’s compliance with the St. Helena Municipal Code (SHMC) Chapter 15.36, which adopts and amends the California Fire Code to address the unique challenges of the St.

Helena area.

The objective of this report is to verify that the site layout, access roads, and fire service features provide a level of safety that protects the existing residence, future residents and the emergency responders. Per SHMC 15.36.010, the city recognizes that response time and effective fire mitigation are critical to public health and safety. These municipal and state codes are not arbitrary and are based on recent California wildfires. These highlight the necessity of safe evacuation routes, adequate emergency access, and the ability of firefighters to operate effectively under chaotic, high-stress conditions.

Given the project’s location near a documented high-hazard wildland-urban interface (WUI) zone, the application of these codes is interpreted through a conservative lens focused on life safety. This review serves as a technical check to ensure that the "minimum code" is not treated as the ceiling for safety, but rather the floor. In a region prone to rapid-fire development and wind-driven ember storms, adhering only to the bare minimum standards without considering the physical reality of the site’s constraints can introduce safety risks that compromise the community.

PROPOSED SITE DESCRIPTION
Spring Grove is a high-density residential development. This project is designated as a Housing Development Project per Government Code section 65589.5(h)(2), emphasizing all-electric living.

Site Location: The project site encompasses three distinct parcels totaling approximately 2.52 acres at 1447, 1505, and 1515 Spring Street. Situated on the west side of St. Helena, the site is bordered by Spring Street and St. James Drive. The parcels at 1447 and 1515 Spring create "Site 9" within the 2023-31 Housing Element. While 1505 Spring serves as the primary medium-density access corridor, the main residential footprint is zoned for high-density residential (HR).

This specific configuration creates a "flag-pole" style layout, where the primary access road (Street A) traverses a relatively narrow strip to reach the bulk of the development. This complicates fire apparatus maneuvers and limits the visibility of responders entering the site. The constrained nature of the "flag" portion of the lot means that any incident occurring at the entrance (Street A) would effectively isolate the 41 homes from emergency services.

FD Response: The proposed Spring Grove Subdivision meets minimum lot width dimensional requirements and therefore does not meet the definition of a “flag lot”. Therefore, Fire Department access is not constrained.

Agenda Item #11.1.

3 rd Party Plan Review
Spring Grove
3January 26, 2026
ZCG No. 0126009
Building and Occupancy Details: The development proposes 41 solar-powered townhome-style condominiums distributed across six building groups (Buildings 1–6). • Total Units: 41 residential units (4 P1, 5 P2, 15 P3, and 17 P4 units). • Occupancy Classification: Group R-2 (Multi-family Residential). • Estimated Population: Based on configurations of 3 to 4 bedrooms per unit, the subdivision is expected to house between 112 and 143 residents. This density significantly increases the "life load" on the site's egress infrastructure; during an emergency, over 100 people and dozens of vehicles would be attempting to exit via a single road while fire engines are attempting to stage and deploy.

FD Response: This proposed private street network is two lanes (26’-28’ in width) allowing one lane for egress and one lane for Fire Dept. apparatus deployment. • Construction Type: Type V-B (non-rated wood-frame), utilizing combustible structural members. This construction type is particularly vulnerable to fire spread between buildings if access for suppression is delayed.

FD Response: All units are required to have independent automatic fire sprinkler systems installed in accordance with SHMC 15.36.060 and CFC Chapter 9. • Building Height: 3-story structures reaching a maximum peak height of 40 feet. The height of these buildings relative to the narrow access ways presents challenges for ground-based laddering and aerial operations.

FD Response: The proposed Spring Grove conceptual plan shows the second story windowsill at 21 ft. from grade. Comparing this height to Turley Flats, a comparable building design located at 1105 Pope Street in St. Helen. The windowsill is located at 23 ft. from grade which is 2 ft. higher than the proposed Spring Grove design. The ground ladder used for deployment is a standard 24 ft. extension ladder used on all SHFD type one engines and Ladder Truck.

Photo of ladder deployment at
1105 Pope Street St.

St. Helena
Friday January 30, 2026
Agenda Item #11.1.

3 rd Party Plan Review
Spring Grove
4January 26, 2026
ZCG No. 0126009
Site Tabulation and Assumptions (Sheet G-1): The Site Tabulation Matrix highlights design choices that adhere strictly to absolute minimums:
• Street Width: The primary entrance (Street A) is confirmed to provide a 26-foot clear width.

While internal streets (B and C) provide a 28-foot width, these measurements provide zero margin for error.

FD Response: As stated, the width of each road provides at least 26 ft of clearance (28 ft. in most places), which meets the minimum standard according to the California Fire Code. • Parking: The development is designed without street parking. This "zero-margin" approach to parking often leads to illegal street parking. On a 28-foot road, even one illegally parked vehicle reduces the clear width to roughly 18–20 feet, which falls below the legal minimum for fire access and can prevent a large apparatus from reaching the scene of a fire.

FD Response: An illegally parked vehicle could potentially inhibit a quick response.

However, an illegally parked vehicle still provides the above stated 18-20 feet of road clearance. This is enough clearance for an 8’ 6” wide ladder truck or engine to proceed to the incident. The minimum width requirement of 26 ft. accounts for illegally parked vehicles. Also in the event of an emergency response, first responders are equipped to break vehicle windows, if necessary, put cars in neutral and move them aside, or nudge vehicles out of the way with the fire engine to clear a path. Additionally, the subdivision design with frequent driveway cutouts would prevent street parking in the development.

Agenda Item #11.1.

3 rd Party Plan Review
Spring Grove
5January 26, 2026
ZCG No. 0126009

FIRE HAZARD HISTORY
Overall, St. Helena is categorized as a high hazard area for wildfires. The City of St. Helena "Fire Resilience" website explicitly states that "residents live and work in a fire-prone area." According to Cal Fire Hazard Severity Zone map, the high severity zone is located 0.5 miles from the proposed Spring Grove site.

FD Response: The Spring Grove project site is not located in any state or local fire severity zone. The map below is the St. Helena City Fire Hazard Severity Zone Map.

Currently in St. Helena, 8 structures are in Very High (red) zones. 220 structures are in High (orange) zones. 290 structures are in Moderate (yellow) zones. 2,846 Structures in St. Helena (including the Spring Grove site) are not located in a fire severity zone (grey)
Of the 3,364 structures currently in City limits, 518 (15.3%) are in either Moderate, High or Very High zones.

Spring Grove Site
Agenda Item #11.1.

3 rd Party Plan Review
Spring Grove
6January 26, 2026
ZCG No. 0126009
Major Fire Events the last 10 years:
• 2025 Pickett Fire (not shown above): Napa County. The fire spread rapidly through terrain fueled dry vegetation typical in this area. Reaching a final size of 6,819 acres (10.3 sq. miles) • 2020 Glass Fire: Burned over 56,701 acres (88.7 sq. miles) and destroyed 1,555 structures, jumping the Napa River and Highway 29.
• 2017 Tubbs Fire: Demonstrated the "urban conflagration" effect, where fires starting in wildlands penetrated city limits and destroyed homes in "lower risk" urban areas through massive ember storms in neighboring Sonoma County. 36,807 acres (57.5 sq. miles) Vulnerability: According to CAL FIRE, nearly 99% of properties in St. Helena face wildfire risk over the next 30 years. St. Helena has a higher risk of wildfire than 85% of communities in the United States. The inherent danger of this location requires the site to not only meet minimum requirements, but al Needs deleted replaced
FD Response: The actual burn acreage of the 2020 Glass Fire totaled 67,484 acres according to the CalFire website. Not 56,701 as stated above.

The St. Helena Fire Department reached out to CAL Fire to confirm the following statement:
“According to CAL FIRE, nearly 99% of properties in St. Helena face wildfire risk over the next 30 years”
The following email provides CAL Fire Lake Napa Unit (LNU) Chief Matt Ryan’s response:

Agenda Item #11.1.

3 rd Party Plan Review
Spring Grove
7January 26, 2026
ZCG No. 0126009
Agenda Item #11.1.

3 rd Party Plan Review
Spring Grove
8January 26, 2026
ZCG No. 0126009

ACCESS AND SECONDARY ENTRANCE
Requirement (CFC Appendix D104.3 / SHMC 15.36.100): Section D104.3 requires that multiplefamily residential projects having more than 30 dwelling units shall be equipped with two separate and approved fire apparatus access roads.

Analysis: The project proposes 41 units (36% over the 30-unit threshold). A single point of access is a critical life-safety flaw in a "fire-prone area." Per CFC D106.3, where two fire apparatus access roads are required, they shall be placed a distance apart equal to at least one-half of the diagonal dimension of the property. The current plan does not show a secondary access point. The project submittal shall be updated to include a secondary access road that meets this remoteness requirement, and its specific location and distance must be verified on the revised plans.

FD Response: CFC Appendix D107.1 includes an exception that allows up to 100 one or twofamily dwelling units before requiring secondary access as long as all the units are equipped with automatic fire sprinkler systems. Fire Sprinklers are required for every unit per SHMC & CFC. ▪ It has been the Fire Department’s stance that if any local amendment to the California Fire Code does not specifically negate an exception in the fire code, such exception remains applicable.
▪ SHMC – 15.36.060 B is a clear example of this existing practice:
“B. Chapter 9, subsection 903.2, is amended by repealing the entire subsection 903.2 (with the exception of subsections 903.2.12, 903.2.17 and Tables 903.2.11.6 and 903.2.5.2, which shall remain in effect), and replacing it with the following”
▪ Such language makes it very clear which sections of an existing code remain in effect, and which ones have been “repealed” and/or “replaced”. ▪ Neighboring Municipalities such as Calistoga, Healdsburg and City of Sonoma use the same kind of clear language in their code amendments.

See excerpts below:
• City of Calistoga:

Appendix D Fire Apparatus Access Roads. California Fire Code Appendix D107.1 is amended as follows:

D107.1 One- or two-family dwelling residential developments.

Developments of one- or two-family dwellings where the number of dwelling units exceeds 20 shall be provided with two separate and approved fire apparatus access roads, and shall meet the requirements of Appendix D104.3.

Exceptions 1 and 2 are deleted.
(Ord. 757 § 4 (Exh. A), 2022)
• City of Healdsburg:

15.08.020 California Fire Code (CFC) Amendments
CFC Appendix D, Section D106.2, is deleted.

CFC Appendix D, Section D107.1, Exceptions Nos. 1 and 2, are deleted.

Agenda Item #11.1.

3 rd Party Plan Review
Spring Grove
9January 26, 2026
ZCG No. 0126009
(Ord. 1222 § 3, 2022)
• City of Sonoma:

Amend Appendix D, subsection D106.1 to read as follows:

D106.1 Projects having more than 50 dwelling units. Multiple-family residential projects having more than 50 dwelling units shall be provided with two separate and approved fire apparatus access roads. (The exception to D106.1 is deleted)
▪ No such language exists in SHMC 15.36.100 D. Excerpt below:
• “D. Section D107.1, One or two-family dwelling residential developments, California Fire Code 2022, Appendix B, is amended to read as follows:” For this reason, the Fire Department defaults back to the referenced California Fire Code and all exceptions remain in force. Excerpt below:
• CFC 2025 – D107.1 - Developments of one- or two-family dwellings where the number of dwelling units exceeds 30 shall be provided with two separate and approved fire apparatus access roads.

Exceptions:

1. Where there are more than 30 dwelling units accessed from a single public or private fire apparatus access road and all dwelling units are equipped throughout with an approved automatic sprinkler system in accordance with Section 903.3.1.1, 903.3.1.2 or 903.3.1.3, access from two directions shall not be required.

TURNAROUNDS AND STAGING
Observation: The internal roadway system includes "Street B," which terminates in a dead end to the plan East. This leg extends 171 feet, highlighted in red in the figure below.

Requirement (CFC 503.2.5): Dead-end fire apparatus access roads in excess of 150 feet shall be provided with an approved turnaround (hammerhead or cul-de-sac).

Staging Analysis: The current dead-end layout on Street B creates a significant tactical risk for the Fire Department. If a fire engine hooks to the hydrant at the Eastern terminus of Street B, it effectively blocks the only access path for additional responding engines to reach Buildings 3, 4, 5, and 6. This "choke point" makes bypass impossible without substantial reconfiguration. During an emergency, this lack of staging and bypass capacity prevents the simultaneous operation of multiple fire apparatus, which is essential for a high-density wood-frame community. Additional consideration for a bypass or widened staging area should be considered and approved by the local AHJ.

FD Response: The access drive provides 2 hammerhead turnarounds. During the initial planning phase of this project, the Fire Department had very specific requirements to address the needs for apparatus turnarounds. Both intersections (Streets A & B, streets B & C) were required to meet the Agenda Item #11.1.

3 rd Party Plan Review
Spring Grove
10January 26, 2026
ZCG No. 0126009
width and turning radius standards of a hammerhead turnaround (see diagram below). This design provides 2 separate, and approved hammerhead turnarounds that exceed the minimum requirements.

CFC 2025 - 503.2.5 Dead ends. Dead-end fire apparatus access roads in excess of 150 feet (45 720 mm) in length shall be provided with an approved area for turning around fire apparatus.

The Fire Department measures from the end of the hammerhead return to the dead end rather than the return of the intersecting road as pictured below. Therefore, the design of the access road and turnaround is in compliance.

Agenda Item #11.1.

3 rd Party Plan Review
Spring Grove
11January 26, 2026
ZCG No. 0126009
150’
Agenda Item #11.1.

3 rd Party Plan Review
Spring Grove
12January 26, 2026
ZCG No. 0126009

FIRE HOSE TRAVEL DISTANCE AND REACH
Requirement (CFC Section 503.1.1): Approved fire apparatus access roads shall be provided for every facility, building, or portion of a building hereafter constructed or moved into or within the jurisdiction. The fire apparatus access road shall comply with the requirements of this section and shall extend to within 150 feet of all portions of the facility and all portions of the exterior walls of the first story of the building as measured by an approved route around the exterior of the building or facility.

Analysis: This requirement ensures that every portion of the building's exterior is reachable with a standard 150-foot hose lay from a fire apparatus parked on an approved access road. Based on our review of the current site plan and building configurations:
• Building 2 Accessibility Issues: According to the plans, Building 2 has insufficient fire department access. The travel distance from the nearest point on the approved fire apparatus access road to the most remote portion of the exterior walls for Building 2 is measured at 173 feet. Shown in red in the figure below. This exceeds the 150-foot maximum allowable distance by 23 feet and limits fire department access to portions of the building. • Excessive Reach: Forcing firefighters to extend hose lines beyond standard lengths or navigate complex routes to reach the rear of Building 2 introduces dangerous delays in suppression efforts.

The site layout or building footprint for Building 2 must be revised to ensure that all exterior portions of the structure fall within the mandatory 150-foot hose travel distance from an approved fire apparatus access road. See Attachment A – Hose Reach Plan at the end of this report for details.

Agenda Item #11.1.

3 rd Party Plan Review
Spring Grove
13January 26, 2026
ZCG No. 0126009
FD Response: The fire hose paths of travel are determined by the shortest unobstructed path.

They are not determined by walkways and/or ADA pathways. The Project Developer has removed the exterior fences on both Southern corners of Building 2 to achieve the CFC requirement of 150 ft.

AERIAL APPARATUS ACCESS
Requirement (CFC D105.1 and D105.2): Where the vertical distance between the grade plane and the highest roof surface exceeds 30 feet, approved aerial fire apparatus access roads shall be provided.

Analysis: The proposed buildings are 3 stories and reach a maximum height of 40 feet, thus triggering the requirements of CFC D105. To comply, the following must be met and clearly shown on the plans:

1. Width: Aerial access roads shall have a minimum unobstructed width of 26 feet in the immediate vicinity of the buildings.

2. Proximity: At least one of the required access routes for each building shall be located within a minimum of 15 feet and a maximum of 30 feet from the building. The current site plan must be updated to show the specific positioning of the aerial apparatus (staging area) relative to each 40-foot building to confirm that these reach and width parameters are achievable given the tight site constraints.

Agenda Item #11.1.

3 rd Party Plan Review
Spring Grove
14January 26, 2026
ZCG No. 0126009
FD Response: The minimum width requirement of 26 ft. is achieved on all proposed roads. This requirement ensures that the above stated requirements about proximity are always attainable. The St. Helena Fire Department ladder truck has an 85 ft. long aerial ladder.

FIRE FLOW AND WATER SUPPLY
Requirement: Per SHMC 15.36.090, fire flow requirements for buildings shall be determined in accordance with CFC Appendix B.

Analysis: Given the high-density wood construction and the significant square footage of these residential occupancies, it is possible that the required GPM (gallons per minute) and PSI (water pressure) may exceed the capacity of the current municipal mains.

Recommendation: We recommend a preliminary fire water analysis and formal fire flow test be performed for this site immediately. This analysis should confirm the feasibility of the municipal water supply providing both the necessary volume and the required pressure for the hydrant system and residential sprinkler systems. This must be confirmed before further site development to ensure the infrastructure can support a wood-frame subdivision of this density.

FD Response: Fire flow tests are performed when fire sprinkler plans are drawn and submitted. If the flow requirements are not met with existing infrastructure, a fire pump is required to make up any deficiencies before the project may proceed.

VEGETATION MANAGEMENT PLAN
Requirement (CFC Section 4906): Hazardous vegetation and fuels shall be managed to reduce the risk of ignition and fire spread.

Analysis: According to the California Fire Code, we suggest a formal Vegetation Management Plan be provided at this stage as a condition of approval. This plan must evaluate the existing conditions, particularly the adjacent creek corridor, and detail how the developer proposes to maintain a safe, fire-resilient environment for the high-density community.

FD Response: This property is not located in the Wildland Urban Interface (WUI) and is not rated in the State adopted Fire Hazard Severity Zone map. Therefore, it is not required to have a Hazardous Vegetation and Fuels Plan. It may be “suggested” above, but it is not required.

See excerpt from 2022 California Fire Code:

SECTION 4906

VEGETATION MANAGEMENT
4906.1 General. Planting of vegetation for new landscaping shall be selected to reduce non-fireresistant vegetation in proximity to a structure and to maintain vegetation as it matures. 4906.2 Application. All new plantings of vegetation in State Responsibility Areas (SRA) and Local Responsibility Areas (LRA) designated as a Very High Fire Hazard Severity Zone shall comply with Sections 4906.3 through 4906.5.3.

Excerpt from 2025 California Wildland Urban Interface Code:

SECTION 603 – VEGETATION PLAN
Agenda Item #11.1.

3 rd Party Plan Review
Spring Grove
15January 26, 2026
ZCG No. 0126009
603.2 Application. All new plantings of vegetation in State Responsibility Area (SRA) and Local Responsibility Areas (LRA) designated as a Fire Hazard Severity Zone shall comply with sections 603.3 through 603.4.2.1
Agenda Item #11.1.

3 rd Party Plan Review
Spring Grove
16January 26, 2026
ZCG No. 0126009

FIRE SERVICE FEATURES
Observation: A fire hydrant on Street C appears to encroach into the roadway and minimum road width for a fire apparatus access road. See object highlighted in green below.

Analysis:

1. Hydrant Protection: The fire hydrant on the Eastern end of Street B extends into the roadway and is not located on a sidewalk or path. Per CFC 507.5.6, this hydrant requires bollard protection. This protection must be shown on the plans to ensure it does not further reduce the effective clear width of the roadway.

2. Fire Lane Marking: Because internal roads (Streets B and C) are at the 28-foot minimum, CFC 503.3 requires that they be strictly marked as "Fire Lanes" with "No Parking" signage or red-curb painting on both sides to prevent street parking from reducing the required width.

FD Response: Placement of bollards, red zone painting, no parking and fire lane signage are determined at a later stage in the building permit review process. Such details are not determined on a tentative map. They are also inspected for adherence throughout the lifespan of the project.

Agenda Item #11.1.

3 rd Party Plan Review
Spring Grove
17January 26, 2026
ZCG No. 0126009

ZARI CONSULTING GROUP,

CONCLUSION
Our review has identified fundamental non-code-compliant issues per the CFC and SHMC 15.36 that present safety concerns. The following table summarizes the primary violations:

Topic Description Code Ref.

Secondary
Access Points
The project proposes 41 residential units with only a single point of ingress/egress. Developments exceeding 30 units require two separate and remote access roads to ensure safe evacuation and emergency vehicle access during a fire event.

CFC D104.3;
SHMC 15.36.100
Dead End
Street B extends more than 170 feet to the plan East without an approved hammerhead or turnaround. Dead-end fire access roads exceeding 150 feet must provide a code-compliant turnaround to prevent apparatus from being trapped.

CFC 503.2.5;
Appendix D
Table D103.4
Fire Hose
Reach
Building 2 contains exterior wall portions that are 173 feet from the nearest access road. This exceeds the 150-foot limit, creating dangerous operational delays for fire suppression.

CFC 503.1.1
Water Supply
and Fire Flow
Given the high density of wood-frame construction, a formal analysis is required to confirm that the municipal water supply can provide the necessary volume (GPM) and pressure (PSI) for fire
suppression.

SHMC

15.36.090; CFC
Appendix B
Staging
Constraints
The narrow layout and hydrant placement create significant tactical bottlenecks. Bollard protection for the hydrant and strict "Fire Lane" conditions are required to maintain clear access for multiple apparatus.

CFC 503.3; CFC
507.5.6; CFC
Chapter 5
Aerial
Apparatus
Access
With building heights reaching 40 feet, the site must accommodate aerial ladder trucks. Current plans do not demonstrate the required unobstructed width or vertical clearance. The specific scaled conditions are required to be shown and verified for safe aerial operations.

CFC D105.1;
CFC D105.2
Please don’t hesitate to contact our office if you have any questions or would like to discuss our comments in further detail.

ZARI CONSULTING GROUP, INC. INC.

Duran Muniz Joseph P. Zari, PE
Project Manager Principal
Agenda Item #11.1.

3 rd Party Plan Review
Spring Grove
18January 26, 2026
ZCG No. 0126009
Agenda Item #11.1.

City of St. Helena
City Council
Project: Spring Grove
Vesting Tentative Map & Water Neutrality In-Lieu Fee
Staff: Mike Janusek, Contract Planner
Community Development Department
February 10, 2026
(continued from
January 27, 2026)
File No. PL24-065 and TSM25-001
Agenda Item #11.1.

Overview
• Background
• Required Findings
• Staff Recommendation
• Key Issues
Agenda Item #11.1.

Location
Agenda Item #11.1.

2023-2031
Housing Element
Site #9
Agenda Item #11.1.

General Plan Land Use
• High Density Residential
(1447 and 1515 Spring St)
• Medium Density Residential
(1505 Spring St)
Project Site
Agenda Item #11.1.

Zoning Map
• High Density Residential (HR)
(1447 and 1515 Spring St)
• Medium Density Residential
(MR) (1505 Spring St)
• HR development standards and
density apply to the site Project Site
Agenda Item #11.1.

Required Entitlements
• Vesting Tentative Map (Attachment 2)
• 41 condominium “airspace” lots
• 1 common area lot owned by the HOA
• 1 lettered lot for the riparian corridor not for
development
• Planning Commission recommended City Council
approval by a vote of 4-0 at the December 2 Public
Hearing (Attachment 3)
• Contingent upon City Council authorization of the
Water Neutrality In-Lieu Retrofit Fee (Attachment 1)
Agenda Item #11.1.

Required Entitlements
• Demolition Permit approved by Planning Commission by
a vote of 4-0 on December 2, 2025 (Attachment 4)
• Demolition of 3 existing single-family homes
• No appeals filed
• Minor Design Review with Density Bonus approved
ministerially on November 21, 2025 (Attachment 8)
• No appeals filed
• Future actions by City Council
Agenda Item #11.1.

Regulatory Context
Senate Bill (SB) 330
• Amends the Housing Accountability Act (HAA)
• Vests zoning and objective design standards with a
Preliminary Application
• Limits the number of public hearings to five
• Prohibits a local government from disapproving or
reducing the density of qualifying housing
development project based on subjective
standards
• Requires health and safety finding to deny or
condition the project to reduce density
Agenda Item #11.1.

Regulatory Context
SB 330 (cont.)
• Denial requires written findings that:
• There is a specific adverse impact on public health or safety • Impact is based on objective public health or safety
standards
• Impact is significant, quantifiable, direct and unavoidable • There is no feasible method to mitigate or avoid the impact other than denial
Gov’t Code sec. 65589.5(j)(1)
Agenda Item #11.1.

Regulatory Context
SB 330 (cont.)
• Housing project is consistent with objective standards if there is substantial evidence that would allow a reasonable person to conclude that the project is compliant with
applicable standards (Sec. 65589.5(f)(4).)
• Legislature has directed that this law be interpreted to afford the “fullest possible weight” to the approval of housing. (Sec. 65589.5(a)(2)(L).)
• In the event of a denial, City has the burden of proof to show it complied with applicable law. (Sec. 65589.6.)
Agenda Item #11.1.

Regulatory Context
Assembly Bill (AB) 130
• Effective June 30, 2025, AB 130 made significant
modifications to the California Environmental Quality Act • Creation of a new statutory CEQA exemption, PRC
Section 21080.66, for infill housing developments that
meet certain site conditions (Attachment 9)
• If there is no action by the City Council to approve or deny the project by March 10, 2026, the project is
deemed approved
Agenda Item #11.1.

Project Description:

Vesting Tentative Map
Spring Street
Agenda Item #11.1.

Vesting Tentative Map Findings
(SHMC 16.16.080.A)

1. The proposed subdivision, together with the provision for its design and improvement, is consistent with the general plan, any applicable specific plan, and applicable provisions of this code in that the proposed subdivision is consistent with the allowable density of the General Plan land use designation as well as applicable General Plan policies as demonstrated in Attachment 7. No Specific Plans apply to the site. In addition, the proposed subdivision complies with the minimum lot dimensions and applicable building setbacks of the HR zoning district per SHMC Table 17.16.030 and the lot design requirements of SHMC Section 16.32.050.

2. The proposed subdivision provides, to the extent feasible, for future passive or natural heating or cooling opportunities in the subdivision, as described in the state Subdivision Map Act and any guidelines promulgated by the city council in that the proposed subdivision meets minimum lot dimensional requirements of SHMC Table 17.16.030 and SHMC Section 16.32.050, the majority of units are oriented to provide southeastern solar exposure, and the majority of units would allow for prevailing upvalley breeze to provide for natural cooling if front and rear doors and windows were opened. In addition, the proposed subdivision layout would provide an internal network of private streets that would be shaded by the Project buildings. The Project also provides open space areas that would be landscaped with future construction.

Agenda Item #11.1.

Recommendation

1. Adopt the attached Resolution (Attachment 1)
authorizing the applicant to pay the Water
Neutrality In-Lieu Retrofit Fee pursuant to
SHMC 13.12.030; and

2. Recommend that the City Council adopt the
attached Resolution (Attachment 2)
recommending, pursuant to a statutory
exemption under PRC 21080.66, approval of
the requested Vesting Tentative Map No.

TSM25-001, File #PL24-065 for the Spring Grove
Project.

Agenda Item #11.1.

January 13, 2026 & January 27, 2026
City Council Hearings
• The item was continued to a date certain, January 27, 2026 and February 10, to address council concerns and
public comments
• Public comments and council concerns were on the
following topics (Attachments 12 & 13)
• Occupancy of existing homes
• Water neutrality compliance
• On-street parking concern
• Partial retrofit option
• Fire Access
Agenda Item #11.1.

Existing Occupancy
• State law requires new housing development must produce at least as many units as are being demolished
• Project would demolish 3 units, but produce 41
• State law also requires that any “protected units” that are demolished must be replaced with an equal number of “protected units”
• COA #25 is included to ensure the affordable housing agreement complies with all applicable laws, including those related to demolition of existing housing units • “Protected Unit” status will be confirmed prior to City Council approval of Affordable Housing Agreement. (Not necessary to delay action on Vesting Tentative Map.)
Agenda Item #11.1.

Water Neutrality
• City Council has the authority over granting the Water Neutrality In-lieu Retrofit Fee request (Attachment 1)
• Fee estimate is calculated in Water Neutrality Report, prepared by City Ventures, dated November 10, 2025
(Attachment 11)
• Public Works has reviewed the fee estimate and
accepts the fee estimate calculation
Agenda Item #11.1.

On-Street Parking
• The project proposes 93 off-street parking spaces, which exceeds the 77 parking space minimum requirement per SHMC Table 17.26.040(A) • Condition of approval #19 requires the project to include wayfinding signs along the project’s internal roadway and at the sidewalk connections, and identity a short-term parking area for delivery vehicles • The HOA and CC&R’s will prohibit on-street parking within the development
Agenda Item #11.1.

Partial Retrofit Option-Water Neutrality
• The City Council may consider a condition of approval to require some retrofits and a partial payment of the in-lieu retrofit fee to the extent the applicant is able to find existing demand for retrofits within the community
• However, after discussing this with the applicant, given the challenges of finding demand (e.g. survey) and the uncertainty of how the City would enforce such a condition, the applicant indicated they would not be amenable to a partial retrofit option
Agenda Item #11.1.

Fire Access
• The Vesting Tentative Map proposes a minimum 26-foot-wide (28 ft. in most proposed roadways) Emergency Vehicle Access Easement.
• The project is conditioned to install NFPA13 fire sprinklers in each unit prior to building permit issuance.
• Per California Fire Code, access from two directions is not required for projects with less than 200 units when automatic fire sprinkler system is provided.

Agenda Item #11.1.

Fire Safety Related Concerns
• Dead-End Road Without Turnaround - The 2 intersections are required to meet the width and turning radius requirements of a hammerhead. The entirety of Streets A, B and C are designated as fire lanes and parking is prohibited.
• Fire Hose Reach - Fire Apparatus access roads must extend to within 150 ft. of all portions of the buildings proposed per CFC. Based off the tentative map, this has been achieved on all buildings. • Water Supply and Fire Flow - Adequate water and fire flow must be established at the time of sprinkler plans submittal. If existing water supply is not adequate, fire department requires fire pumps to be installed to make up the deficiencies. This project will not be approved without adequate water flow rates. (Note: Before a building permit can be issued, there must be an approved sprinkler plan). • Staging Constraints - The majority of Roads A, B and C exceed the minimum requirement of 26 ft. wide.

There will be no on street approved parking, allowing emergency vehicles full access to entire width of the proposed roadway. Precise placement of 5 proposed fire hydrants and bollards will be discussed and approved during the permit phase of the project, to address any access constraints. This is standard procedure. • Aerial Apparatus Access - These concerns have been addressed and reviewed. The proposed roadways meets the standard CFC requirements for aerial apparatus.

Agenda Item #11.1.

Questions?
Agenda Item #11.1.

SPRING GROVE
1447-1505-1515 Spring St. - St. Helena - CA - 94574 City Ventures SPRING GROVE: WHAT IS ACTUALLY PROPOSED
A 41-unit, three-story townhouse project on Spring Street—shown as submitted VIEW FROM SPRING ST.

SPRING GROVE
1447-1505-1515 Spring St. - St. Helena - CA - 94574 City Ventures SPRING GROVE: WHAT IS ACTUALLY PROPOSED
VIEW FROM ST. JAMES CT.

41'
18'
ST. HELENA
1 2
3 4
5 6
7 8
9 Outlook
Public Comment and Research Attachments Regarding Zoning and Housing Outcomes From Michelle Covell <[email protected]>
Date Tue 2/3/2026 12:11 PM
To Michelle Covell <[email protected]>
14 attachments (16 MB)
PUBLIC COMMENT - MLC.pdf; ATTACHMENT 1.pdf; ATTACHMENT 2.pdf; ATTACHMENT 3.pdf; ATTACHMENT 4.pdf; ATTACHMENT

5.pdf; ATTACHMENT 6.pdf; ATTACHMENT 7.pdf; ATTACHMENT 8.pdf; ATTACHMENT 9.pdf; ATTACHMENT 10.pdf; ATTACHMENT

11.pdf; ATTACHMENT 12.pdf; ATTACHMENT 13.pdf;
Hi all,
Thank you for your time and for your service to our community.

I am sharing the attached document, “public comment – MLC,” along with thirteen supporting attachments. I am sending this to both state legislators and local leaders because the issues involved sit at the intersection of state mandates and local implementation.

I value your perspectives and your work, and I hope you will not see me as a thorn in your side.

My intent is to contribute constructively.

The research reflected in these attachments, together with my letter, is meant to clearly explain my concerns as a resident and to support more informed discussion.

I recognize that we are dealing with both state mandate requirements and local constraints.

My hope is that we can continue refining our zoning code to create better long term outcomes for our town.

Toward the end of the attachments, I have included my own analysis of local conditions and impacts. With strong teamwork and careful review of the details, I believe improvements are possible. Because timing is critical, I hope you are able to review these materials before the next council meeting.

I also want to share a personal example that informs my concern. I live next door to a two-story apartment building with many residents who work in local restaurants and services. A neighbor who has lived there for twenty two years is now moving out. She works as a housekeeper, her husband is in landscaping, their daughter works for Parks and Recreation, and their son also works locally. After more than two decades, they found a place that costs slightly more but offers a small front and back yard. Their two bedroom apartment rent had risen to more than our mortgage, and their new rent will be close to one and a half times our mortgage, not including the utilities. We have shared lemons, tomatoes, tamales, and tacos over the years, and she has watched my garden grow from her window. She told me she sees projects happening but does not feel comfortable speaking publicly about them. I encouraged her that her voice matters and that I would try to support her. Residents like her are part of the fabric of this town.

One of the studies included in the attachments notes that small towns often lack the resources to conduct this level of analysis, which is why their research was undertaken. I hope these materials are helpful as you evaluate next steps.

Agenda Item #11.1.

Thank you again for your time and consideration, and for reviewing the attached letter and supporting documents before the upcoming meeting.

Respectfully,
As resident, not in any other role.

Michelle Liu Covell
Agenda Item #11.1.

1 Dear Governor Gavin Newsom, Senator Calbadon, Assembly Member Aguiar-Curry, Mayor Dohring, Vice Mayor Deasey, Councilmembers Spadorato, Barak, and Summers, District 3 Board of Supervisor Anne Co%rell, A%orney General Rob Bonta, and State legislators, civic leaders, and whom this concerns, I’m wri)ng to express deep concern about the one size fits all RHNA mandates and the real-world danger they create for fire prone communi)es. Have we already forgo%en the devasta)on of recent catastrophic wildfires and the lives lost when evacua)on planning failed? “An evacua)on disaster was years in the making” is poor city and regional planning, lack of foresight, and negligence of leaders. (See a%achments 6, 7, 8, 9) I have evacuated mul)ple )mes. The memory that stays with me most is the 2017 Atlas Peak Fire driving down Highway 29 as both sides of the valley glowed with fire. I am a licensed architect in the State of California and recently appointed as Planning Commissioner for our town. Today, I am speaking as a resident solely, someone who has seen firsthand how quickly condi)ons turn deadly and how current housing laws are increasing risk rather than reducing it. In 2019, we had to evacuate from the poor air quality with our 3-week-old due to the Kincaide fire, and in 2020, we had to evacuate the Glass Mountain Fire, flames visible from our backyard.

Removing the Growth Management chapter from General Plans is a profound failure of responsibility at the state level. According to a na)onal study by StreetLight, of the U.S. ci)es, they gathered a list of the top 675 communi)es with the most limited evacua)on routes, St. Helena ranks 267 based on popula)on, exit load, and number of exits (see a%achment 4). Increasing popula)on in small towns with constrained evacua)on capacity, high fire severity zones, and limited cumula)ve exits is not a theore)cal planning issue. It is a public health and life safety issue (see a%achment 1, 2, 3, 4, 5) To strip away local jurisdic)ons the Growth Management chapter of the General Plans is to strip away what is protec)ng the town from complete failure. PuEng aside long-term sustainable growth management, with infrastructure and capacity in mind, for a State mandate is like puEng a poten)al death sentence to many towns where these issues could be the )pping point. And what was wri%en to incen)vize more housing could likely end with towns failing and more housing lost. What is the most troubling is that “we” are doing this to ourselves through poor leadership, narrow-sighted one-size-fits all laws that when applied to all ci)es/towns without the lack of a%en)on to the local constraints, we will all suffer.

I am seeing the lesson of The Death and Life of Great American Ci)es by Jane Jacobs play out again. Top-down decisions that ignore local condi)ons ul)mately destroy what they intend to improve. The book is ul)mately a warning that ci)es cannot be engineered like machines. Towns are not like computers where we can simply apply a system update across all. They must be understood as complex ecosystems where safety, vitality, and community emerge from local knowledge and everyday life. We need housing solu)ons where planning is honest about risk, responsive to place, and grounded in life safety.

Agenda Item #11.1.

2 If the state requires increased housing thereby increasing popula)on, it must also require increased evacua)on capacity. Governor Newsom, I know you know Napa Valley. You know the roads are narrow, the lanes are limited, and the valley is long. Requiring addi)onal popula)on without first expanding road capacity and exit routes does not balance needs. It increases the likelihood of people being trapped during the next major fire. That is not acceptable, this is malprac)ce and negligence in a major scale.

I understand the need for affordable housing. That is not the issue. The issue is that the current laws, as wri%en and implemented, are failing to deliver truly affordable housing while simultaneously amplifying risk. (See a%achments 10, 11, 12, 13)
Under our 2023 to 2031 Housing Element, St. Helena is required to plan for 256 housing units, 163 of which are categorized as very low or low income. Yet our zoning code allows “for sale developments” to meet affordability requirements with only 20 percent affordable units without requiring any very low or low income units at all. In prac)ce, this has created a loophole where developments labeled as affordable while mostly are market rate, with the remaining 20 percent priced at 100 percent to 120 percent of AMI which in Napa County is already out of reach for many working residents. (See a%achments 10, 11, 12, 13) The result is predictable. Developers follow the incen)ves while maximizing profit. More market rate housing is built than actual affordable housing. Popula)on increases. Evacua)on capacity does not. For a town of roughly 4,000 housing units, this math produces growth without safety and that is a severe life safety failure.

SB 330 allows projects to be denied only when they pose an adverse risk to public health and safety. Life safety must be included in that defini)on. Yet local jurisdic)ons are forced to evaluate projects in isola)on without any cumula)ve analysis and long-term projec)on of evacua)on capacity, infrastructure strain, or popula)on growth.

What I see instead is fragmented review, narrowed departmental silos, and a system structurally incapable of addressing the scale of risk being created. (See a%achments 1, 2, 3, 4, 5) As the laws and codes are currently wri%en, they priori)ze speed and profit over safety and foresight. They protect those who benefit financially while stripping local jurisdic)ons of the tools needed to protect their residents. That is not fulfilling the intent of the housing laws. It is failing the people of California.

The New York Times headline says it plainly. In the Palisades, an Evacuaon Disaster Was Years in the Making. In St. Helena, we are watching the same condi)ons assemble in real )me. At a Special Joint Mee)ng on August 19, 2025, I asked whether any cumula)ve exit analysis had been done to understand what increased popula)on would mean during evacua)on. Others asked the same ques)on. I raised it again and again at Planning Commission and City Council mee)ngs. I contacted Napa County building officials. There has been no answer. This is all documented.

I am raising this red flag now because one size fits all mandates do not account for towns with limited roads, high fire severity zones, aging water and sewer systems, and constrained evacua)on routes. Increasing popula)on under these condi)ons is not neutral. It compounds risk. If risk is measured on popula)on, exit capacity, and number of exits, (see a%achments 1, 2, 3, 4) by increasing popula)on without addressing exit capacity and number of exits, by simple logical math, is increasing risk.

This is not a NIMBY argument. If housing policies were producing genuinely affordable homes and aligning growth with infrastructure and safety, this conversa)on would be different. Instead, the current framework increases popula)on, increases strain, and increases danger while delivering mostly market rate housing. What I am asking you all, is to think about the intent of these laws, the big picture, and then look at the details and see if the outcome is mee)ng your intent. If it doesn’t, then it is )me to take a deep dive into the reasons why they are not, and fix those things before it is too late. I’m not saying affordable housing cannot be achieved. What I’m saying is affordable housing is not what’s being built, please look at the detail. And what we need to fix is the language in order to achieve the desired outcome, and ensure that the unintended consequences are not poten)al loss of Agenda Item #11.1.

3 lives, and or death of more great ci)es. Massive wildfires are our new normal. Our laws need to take that into account, seriously.

I was very much inspired by a story about a young girl who recognized the warning signs of a tsunami and saved lives because someone listened. I am asking you to listen now before the warning becomes the headline. See story included below.

Sincerely,
Michelle Liu Covell, Resident of St. Helena, Architect. Not represen)ng any other role. 1/30/2026
List of A
achments
Please find the a%ached documents for your reference, some have been highlighted and commented on.

1. KQED: Data Pinpoints 14 California Towns Where an Emergency Escape could be a Problem

2. Ins)tute of Transporta)on Studies: Tes)ng Wildfire Evacua)on Strategies and Coordina)on Plans for Wildland-Urban Interface (WUI) Communi)es in California

3. White Paper: Bay Area Wildland Urban Interface – Review of Risks, Plans, and Strategies

4. U.S. Communi)es with Limited Evacua)on Routes (top 675 towns in the US) Streetlight Data

5. Fire Safety Journal – The analysis of traffic data of wildfire evacua)on: the case study of the 2020 Glass Fire

6. St. Helena 2040 General Plan – Growth Management

7. The Planning Commissioner Handbook: Ins)tute for Local Government – Excerpt on Good Infrastructure

8. The Planning Commissioner Handbook: Ins)tute for Local Government – Excerpt on Public Service Ethics

9. New York Times: Jan 8, 2025 “Some Pacific Palisades residents said the community had long asked for more detailed fire prepara)on plans. When a fire raged through the community this week, gridlock ensued.”

10. The Zoning Code – my observa)ons

11. On Math – My observa)ons

12. This is Not a Nimby-Yimby situa)on – my observa)ons

13. Take a look at these two plans – my observa)ons
Agenda Item #11.1.

4 A story about Tilly who saved lives by being observant, and others listened to her, rather than dismissed her. And she saved lives.

Jaime Tanna – Substack Post
On the morning of 26 December 2004, Tilly was walking along Mai Khao Beach in Phuket with her family, enjoying their first overseas holiday together. The day was calm, bright, ordinary.

Then Tilly no)ced something wasn’t right.

The sea wasn’t moving in its familiar rhythm. It wasn’t gently coming and going.

It was only coming in.

Frothy. Bubbling. Sizzling — “like beer foam,” she later said.

Two weeks earlier, in a geography lesson at Danes Hill School in Surrey, Tilly’s teacher, Andrew Kearney, had shown the class footage of a tsunami and explained the warning signs: unusual sea behavior, receding or surging water, froth and turbulence where there should be calm.

Standing on that beach, Tilly recognized every sign and began shou)ng at her parents: “There’s going to be a tsunami!”
There was no visible wave. The sky was blue.

But Tilly wouldn’t stop. She became more insistent.

Eventually her father, Colin, heard something deeper than fear — he heard certainty. And he chose to trust his daughter.

Nearby, an English-speaking Japanese man overheard the word tsunami. He had just learned of a powerful earthquake off Sumatra. “I think your daughter’s right,” he said.

Hotel staff were alerted. The beach was immediately evacuated.

Tilly’s mother, Penny, was among the last to leave, sprin)ng as the water surged in behind her. They reached the second floor with seconds to spare.

Then the wave hit.

A massive wall of water — around 30 feet high — tore through the beach, ripping up trees, furniture, and debris, hurling everything forward with devasta)ng force.

Across 14 countries, the Indian Ocean tsunami killed more than 230,000 people.

But on Mai Khao Beach, not a single life was lost.

Because a 10-year-old paid a%en)on in class.

Tilly was later called the “Angel of the Beach.” She received interna)onal recogni)on, spoke at the United Na)ons, and her story became part of disaster-educa)on curricula worldwide.

Today, Tilly lives quietly in London. She s)ll credits her teacher. “If it wasn’t for Mr. Kearney,” she once said, “I’d probably be dead — and so would my family.” This isn’t just a story about a tsunami.

It’s a reminder
Some)mes the smallest lesson becomes the loudest lifesaver.

Agenda Item #11.1.





 
 


 !""
"#

""#$% &'()*') +,-,./012301-4.56.7,8093:10,.;3<14.=>?:?.,1.@A?:B?1CD.@4C,2?.73E8F G?.,./:3H8?A I$J$KL MNOPQQRPQSTU VWXYZ[XP\Y]PURPQSQ^ _:-0C8?.`,a?F bcdefghicj k3:EA Agenda Item #11.1.








 







 

 




 

 




!"#$%&!'()*%$+,-./01)223!4567)%8
Agenda Item #11.1.




























































































!
























"



 



#



!














$
%


























&


 



'

$



  
"
()

































 *





&




 +



,
-



 %.,





 












"





"







/

































"





  



























$
 




0

+



 123453678 9

































: 
;















<




9*
 

















"





< #























&











 


=+












"



















:













 




























$





"

Agenda Item #11.1.



 















 










































! 

 "#



















 











$
 "











 





 




%



 

&



$ '

























 










 #












())


&













*
Agenda Item #11.1.


 





   

! "##$ %&

'
()*+, -& 
!.
( -/ 

0 1)2,#%##$ 3+42+

5+62(+ 7# 
''0
 -8+

9 . :;: %& !
9 ;+$ /$ 
90 ! <2=-*+2 -)
' ! (222 -(&+2>+2* 

":+#2 /,+ !

 (:#+?+ 16) !
'
<+##) %& 

! @ABACDEFGHIJKICALMFNLKCDO 16 #+?+,))&++2264+6)2#+)+4,&+&2>2 #$2+2?)2)*62P;-4+,2Q RSTUVWXYTZX[U\]]\^_`aYXbXcdeUf\WXghijjhklmnhopqhqXbXrjhohnjosqhqXbXtijqhjsoulhnopqhquiqvvji Agenda Item #11.1.














































































 









!"#$!%&'



( 
)

*+,-. /



0

 1







1
 





1 

















. 2

3


























 1

3

1






4



5



1




0

6

7


8




*-+





2

1


)


9

2  1 
9


,*
4










2

. 


 




 


















:+1+++1




















 







 




. 


1

7

1




















7











. ;


<%=#$>%?$@"AA=$B"&C



,DD,1




*E


1



4


.
0 &CFG&#




 




H.2.
)
0



























1


 I













 .J
0 
5

 1

















1 


 . 2

3


2










 












 




 












 

. I(
























1














 


























7 1J



. 2












 





















1





1























 . 8


 








1
2



1













. KLMNOP
Agenda Item #11.1.

















 

















 



























!"#$%&'()*+,(-*$.$*/'(-%+.'&0(1+.(2')(03*'+3' +'-0()41)(51))'&0 6 
7
8 


 








 

  

 9:;<=>?@AA<BC<D?BDEF:BCG<H:IJ<IJC<CKL:A<LGGFCDD<A:MNCG<I><=>?F<LEE>?MIO PQRSTUUTVRWXTUUYZRSQ[TR[\]RWP ^_`abcd`e Agenda Item #11.1.

 



 

 

 

 
 








    !

"
#


  



#










 
 

#

# 
# 


$ %&'()&*+,-./,012%,*334 56789:;<7=6;97:>6?@7A6B7B97?C97D6?;7E<F6;=:@E6<G7H89:C97;9:I76?;7H;EJ:KD H68EKDL Agenda Item #11.1.



    



  !" !

Agenda Item #11.1.

RESEARCH REPORT Institute of
Studies
Transportation
Report No.: UC-ITS-2022-34 | DOI: 10.7922/G2XK8CX7
Testing Wildfire
Evacuation Strategies and
Coordination Plans for
Wildland-Urban Interface
(WUI) Communities in
California
Kenichi Soga, Ph.D., Chancellor’s Professor; Donald H. McLaughlin Chair in Mineral Engineering1,2
Louise Comfort, Ph.D., Professor Emerita, Center for Information Technology Research in the Interest of Society (CITRIS)2 Pengshun Li, M.S.1,2
Bingyu Zhao, Ph.D., Postdoctoral Researcher, Soga Research Group2 Paola Lorusso, MSc.1,2 April 2024
1 Department of Civil and Environmental Engineering
2 University of California, Berkeley
Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California ii Technical Report Documentation Page

1. Report No.

UC-ITS-2022-34

2. Government Accession No.

N/A

3. Recipient’s Catalog No.

N/A

4. Title and Subtitle
Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland-Urban Interface (WUI) Communities in California

5. Report Date
April 2024

6. Performing Organization Code
ITS Berkeley

7. Author(s)
Kenichi Soga, Ph.D. http://orcid.org/0000-0001-5418-7892; Louise Comfort, Ph.D. http://orcid.org/0000-0003-4411-1354; Pengshun Li. http://orcid.org/0000-0002-1831-5153; Bingyu Zhao, Ph.D. http://orcid.org/0000-0002-2369-7731; Paola Lorusso

8. Performing Organization Report No.

N/A

9. Performing Organization Name and Address
Institute of Transportation Studies, Berkeley
109 McLaughlin Hall, MC1720
Berkeley, CA 94720-1720

10. Work Unit No.

N/A

11. Contract or Grant No.

UC-ITS-2022-34

12. Sponsoring Agency Name and Address
The University of California Institute of Transportation Studies www.ucits.org

13. Type of Report and Period Covered
Final Report (July 2022 – May 2023)

14. Sponsoring Agency Code
UC ITS

15. Supplementary Notes
DOI: 10.7922/G2XK8CX7

16. Abstract
In the event of a wildfire, government agencies need to make quick, well-informed decisions to safely evacuate people. Small communities, such as in Marin County, with a mix of residences and flammable vegetation in Wildland-Urban Interface zones tend to lack resources to conduct evacuation studies. Consequently, this study uses a framework of wildfire and traffic simulations to test the performance of potential evacuation strategies, including reducing the volume of evacuating vehicles through car-pooling, phasing evacuations by staggering evacuation times by zone, and prohibiting street parking in four representative areas of Marin County. Results show that reducing vehicle numbers lowers the average travel time by 20%- 70% and average exposure time to wildfire by 27%-60% from the baseline. Phased evacuations with suitable time intervals lower the average travel time by 13.5%-70%, but may expose more vehicles to fire in some situations. Prohibiting street parking yields varying results due to different numbers of exits and evacuees. In some cases, prohibiting street parking reduces the average travel time by over 50%, while in other cases it only reduces the average travel time by 9%, contributing little to evacuation efficiency. Altogether, Marin County may want to consider developing a communication and parking plan to reduce the number of evacuating vehicles in wildfire situations. Phased evacuation is also highly recommended, but the suitable phasing interval depends on the speed of fire spread and number of evacuees. Further, whether to establish street parking prohibition policies for a certain area depends on the number of exits and the number of vehicles on the streets.

17. Key Words
Wildfires, evacuation, urban areas, greenways, traffic simulation, advanced traveler information systems, street parking

18. Distribution Statement
No restrictions.

19. Security Classification (of this
report)
Unclassified

20. Security Classification (of this page)
Unclassified

21. No. of Pages
97 22. Price
N/A
Form Dot F 1700.7 (8-72) Reproduction of completed
page authorized
Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California iii About the UC Institute of Transportation Studies
The University of California Institute of Transportation Studies (UC ITS) is a network of faculty, research and administrative staff, and students dedicated to advancing the state of the art in transportation engineering, planning, and policy for the people of California. Established by the Legislature in 1947, ITS has branches at UC Berkeley, UC Davis, UC Irvine, and UCLA.

Acknowledgments
This study was made possible with funding received by the University of California Institute of Transportation Studies from the State of California through the Road Repair and Accountability Act of 2017 (Senate Bill 1).

The authors would like to thank the State of California for its support of university-based research, and especially for the funding received for this project. The authors would also like to thank Mark Brown (Executive Officer of the Marin Wildfire Prevention Authority (MWPA)) and Charlotte Jourdain (Senior Consultant for MWPA) for identifying the case study sites and reviewing the report, and Fehr and Peers for the use of their processed data in validating the simulations.

Disclaimer
The contents of this report reflect the views of the authors, who are responsible for the facts and the accuracy of the information presented herein. This document is disseminated under the sponsorship of the State of California in the interest of information exchange. The State of California assumes no liability for the contents or use thereof. Nor does the content necessarily reflect the official views or policies of the State of California.

This report does not constitute a standard, specification, or regulation.

Agenda Item #11.1.

Report No.: UC-ITS-2022-34 | DOI: 10.7922/G2XK8CX7
Testing Wildfire
Evacuation Strategies and
Coordination Plans for
Wildland-Urban Interface
(WUI) Communities in
California
Kenichi Soga, Ph.D., Chancellor’s Professor; Donald H. McLaughlin Chair in Mineral Engineering1,2
Louise Comfort, Ph.D., Professor Emerita, Center for Information Technology Research in the Interest of Society (CITRIS)2 Pengshun Li, M.S.1,2
Bingyu Zhao, Ph.D., Postdoctoral Researcher, Soga Research Group2 Paola Lorusso, MSc.1,2 April 2024
1 Department of Civil and Environmental Engineering
2 University of California, Berkeley
Institute of Transportation Studies
Agenda Item #11.1.

Table
of
Contents
Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland-Urban Interface (WUI) Communities in California Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California vi Table of Contents
Executive Summary........................................................................................................................... 1 Wildfire-Traffic Simulation Framework............................................................................................................................. 1 Case Studies........................................................................................................................................................................... 2 Introduction ..................................................................................................................................... 5 Evaluating Wildfire Evacuation Strategies........................................................................................... 6 Developing Evacuation Scenarios ..................................................................................................... 10 Wildfire Spread Simulation ...............................................................................................................................................11 Spatial-Queue Based Traffic Simulation.........................................................................................................................12 Evacuation Strategies and Scenario Development.......................................................................................................14 Evaluation of Evacuation Strategies ................................................................................................................................16 Case Studies ................................................................................................................................... 17 Scenario Development.......................................................................................................................................................18 Validation of the Simulation .............................................................................................................................................28 Impact of Background Traffic ...........................................................................................................................................29 Strategy Analysis.................................................................................................................................................................33 Recommended Strategies..................................................................................................................................................78 Policy Recommendations................................................................................................................. 80 Vehicle Reduction...............................................................................................................................................................80 Phased Evacuation..............................................................................................................................................................81 Street Parking Prohibition.................................................................................................................................................81 Summary and Conclusions ............................................................................................................... 83 References ..................................................................................................................................... 84 Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California vii List of Tables
Table 1. Selected Studies That Employ Simulations or Traffic Models for Studying Wildfire Evacuation (adapted from Zhao and Wong, 2021)................................................................................................................................................... 8 Table 2. Description of Scenarios.........................................................................................................................................15 Table 3. Modified Meteorological Parameters for the Ross Valley Scenario...............................................................20 Table 4. Summary of Number of Evacuees for Different Scenarios...............................................................................26 Table 5. Comparison of Model Traffic and Counts from Roadside Detectors during Morning Peak ......................28 Table 6. Comparison of Model Traffic and Counts from Roadside Detectors during Afternoon Peak ...................29 Table 7. Comparison of Average Evacuation Time With and Without Background Traffic.......................................32 Table 8. Average Evacuation Time and Exposure Time for Vehicle Reduction Strategy, All Case Studies ............50 Table 9. Departure Times for Phased Evacuation Strategy, All Case Studies ..............................................................55 Table 10. Average Evacuation Time and Exposure Time for Phased Evacuation Strategy, All Case Studies .........64 Table 11. Average Evacuation and Exposure Times for Street Parking Prohibition Strategy for All Case Studies ....................................................................................................................................................................................................75 Table 12. Recommended Strategies ....................................................................................................................................79 Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California viii List of Figures
Figure 1. Study Flowchart......................................................................................................................................................10 Figure 2. Framework of Fire Simulation..............................................................................................................................11 Figure 3. An Illustrative Node Model ..................................................................................................................................13 Figure 4. An Illustrative Link Model ....................................................................................................................................13 Figure 5. Illustration of Evacuation Time and Exposure Time Calculations .................................................................16 Figure 6. Four Study Areas in Marin County and Fire Hazard Areas.............................................................................17 Figure 7. The Road Network of Marin County...................................................................................................................18 Figure 8. Ignition Points for the Four Cases and the Locations of Weather Stations ................................................20 Figure 9. Fire Propagation for Ross Valley Scenarios.......................................................................................................21 Figure 10. Fire Propagation for Woodacre Bowl Scenarios.............................................................................................22 Figure 11. Fire Propagation for Tamalpais Valley Scenarios...........................................................................................22 Figure 12. Fire Propagation for Novato Neighborhood Scenarios ................................................................................23 Figure 13. Change of O/D which are outside Marin County to the border of Marin.................................................24 Figure 14. The Spatial Distribution of Background OD pairs within Marin County ...................................................24 Figure 15. Temporal Distribution of Background OD Pairs within Marin County......................................................25 Figure 16. Evacuation Destinations.....................................................................................................................................27 Figure 17. Arrival Times for Ross Valley .............................................................................................................................30 Figure 18. The Arrival Times for Woodacre Bowl..............................................................................................................31 Figure 19. The Arrival Times for Tamalpais Valley............................................................................................................31 Figure 20. Arrival Times for Novato Neighborhood .........................................................................................................32 Figure 21. Network Topologies and Exits for the Four Case Study Areas....................................................................33 Figure 22. Baseline Case for Ross Valley ............................................................................................................................35 Figure 23. Baseline Case for Woodacre Bowl ....................................................................................................................37 Figure 24. Baseline Case for Tamalpais Valley ..................................................................................................................39 Figure 25. Baseline Case for Novato Neighborhood ........................................................................................................41 Figure 26. Vehicle Reduction Strategy Performance Measures for Ross Valley.........................................................43 Figure 27. Vehicle Reduction Strategy Performance Measures for Woodacre Bowl .................................................45 Figure 28. Vehicle Reduction Strategy Performance Measures for Tamalpais Valley ...............................................47 Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California ix Figure 29. Vehicle Reduction Strategy Performance Measures for Novato Neighborhood.....................................49 Figure 30. Average Evacuation Times for Different Carpooling Scenarios, All Case Studies ...................................51 Figure 31. Average Exposure Times For Vehicle Reduction Strategy, All Case Studies ............................................52 Figure 32. Evacuation Zones for Ross Valley.....................................................................................................................53 Figure 33. Evacuation Zones for Woodacre Bowl .............................................................................................................54 Figure 34. Evacuation Zones for Tamalpais Valley ...........................................................................................................54 Figure 35. Evacuation Zones of Novato Neighborhood ..................................................................................................55 Figure 36. Phased Evacuation Performance Measures for Ross Valley).......................................................................57 Figure 37. Phased Evacuation Performance Measures for Woodacre Bowl ................................................................59 Figure 38. Phased Evacuation Performance Measures for Tamalpais Valley ..............................................................61 Figure 39. Phased Evacuation Strategy Performance Measures for Novato Neighborhood....................................63 Figure 40. Average Travel Times for Various Departure Intervals for Phased Evacuation Strategy, All Case Studies.......................................................................................................................................................................................65 Figure 41. Average Exposure Times for Different Departure Intervals for Phased Evacuation Strategy, All Case Studies.......................................................................................................................................................................................66 Figure 42. Street Parking Prohibition Strategy Performance Measures for Ross Valley...........................................68 Figure 43. Street Parking Prohibition Strategy Performance Measures for Woodacre Bowl...................................70 Figure 44. Street Parking Prohibition Performance Measures for Tamalpais Valley .................................................72 Figure 45. Street Parking Prohibition Strategy Performance Measures for Novato Neighborhood ()...................74 Figure 46. Average Evacuation Times With and Without Street Parking.....................................................................76 Figure 47. Average Fire Exposure Times With and Without Street Parking ................................................................77 Agenda Item #11.1.

Executive
Summary
Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland-Urban Interface (WUI) Communities in California Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 1 Executive Summary
Due to the combination of warmer temperatures and drought brought on by climate change, wildfires have become an ongoing crisis particularly in transitional Wildland Urban Interface (WUI) areas where residential development has been expanding into forested regions of California. Where lives are at risk, mass evacuation is frequently the sole feasible way to ensure people’s safety. Current evacuation strategies rely on standard, but static, approaches such designating evacuation zones to notify residents in the event of a fire. The most at-risk zones can be evacuated first which helps emergency personnel manage the traffic flow and more easily prevent the traffic jams that could occur when everyone tries to leave the area at the same time. However, these procedures may be difficult to implement in the case of rapidly moving fires and where the community lacks robust communication infrastructure, such as local radio and emergency alert systems.

Many small, resource-strapped communities in WUI zones do not have resources to conduct detailed evacuation studies and many such studies do not consider the impact of background traffic on evacuation processes. Therefore, this study explores the likely performance of a set of generalizable evacuation strategies incorporating background traffic for several areas in Marin County: the Ross Valley, Woodacre Bowl, Tamalpais Valley, and an area near Highway 101 and Ignacio Boulevard in Novato (hereafter referred to as 'Novato Neighborhood'). Using a wildfire-traffic simulation framework several strategies were tested including reducing the number of vehicles involved through carpooling, phased evacuations, and off-street parking prohibitions.

Wildfire-Traffic Simulation Framework
The proposed framework has three parts: (i) wildfire spread simulation, (ii) evacuation behavior, (iii) and traffic evacuation simulation. For the wildfire spread simulation, the model considers terrain topography, fuel, and meteorological conditions along with fire ignition location and time. The results show how the fire is likely to progress over time in one-hour time steps. These data subsequently become inputs for the traffic simulation which provides information to develop a plan of evacuation. For the evacuation behavior portion, three kinds of strategies were considered in this study: (a) vehicle reduction, (b) phased evacuation, and (c) prohibiting vehicles from remaining on the streets (and moving vehicles off the street). These strategies can affect the number of vehicles participating in the evacuation, their respective departure times, and the available road capacity needed for them to reach safety.

For evaluating the vehicle reduction strategy, five different options were tested, including one person/vehicle, two persons/vehicle, three persons/vehicle, four persons/vehicle, and five persons/vehicle. For phased evacuation, four options were tested, including simultaneous departure, 15-minute departure intervals in the daytime scenario or 20-minute departure interval in the nighttime scenario, 30-minute departure intervals in the daytime scenario or 40-minute departure intervals in the nighttime scenario, 45-minute departure intervals in the daytime scenario or 60-minute departure intervals in the nighttime scenario. For parking prohibition Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 2 strategies, we assume that if street parking is allowed, road capacity would be reduced by 15 percent compared with prohibiting street parking. The information from these scenarios is used to inform the traffic simulation model. The traffic simulation model is a mesoscopic traffic model, which examines small groups of vehicles and can track the trajectory of each vehicle during the evacuation. For the traffic simulation, the model produces various metrics of evacuation efficiency, such as the time needed for vehicles to reach two miles away from the fire area, the number of vehicles exposed to fire risk at different times, average evacuation travel time, and average exposure time to the fire hazard, to evaluate the performance of the different evacuation strategies.

Case Studies
The four study sites represent diverse fire and traffic characteristics in high-risk communities. For each site, two different evacuation scenarios were designed: one at 1 a.m., when all the residents are at home, another at 9 a.m., when background traffic is heavy.

The estimated number of residents who need to evacuate is different between the morning and nighttime scenarios because many residents are at work during the day, while most are at home at night. The background Origin-Destination (OD) data from the Metropolitan Transportation Commission identifies the location where each person begins their travel day, and this place is regarded as their residence. Using this information, it is possible to determine how many residents are at home when the fire ignites, and need to evacuate. For Ross Valley, the number of evacuees is 1,965 in the daytime scenario and 4,893 in the nighttime scenario. For Woodacre Bowl, the number of evacuees is 1,356 in the daytime scenario and 2,196 in the nighttime scenario.

For Tamalpais Valley, the number of evacuees in the daytime scenario is 2,010 and 5,492 in the nighttime scenario and for Novato Neighborhood, there are 7,662 evacuees in the daytime scenario and 17,519 in the nighttime scenario.

The analysis showed that different strategies perform better or worse in different areas and times due to differences in road networks, fire ignition points and progression, number of evacuees, and the normal traffic around the areas.

Generally, sharing a vehicle among multiple persons is a good strategy in each case, as this can greatly reduce traffic congestion and speed up the evacuation process. However, it is difficult and time-consuming to coordinate having persons from different households share one vehicle in emergency situations. On the contrary, it is quite easy to encourage persons to share one or two vehicles among their own family members.

We recommend a policy that at least two persons should be in each car during the evacuation.

Phasing the evacuation process improves evacuation efficiency, however, a longer departure interval does not necessarily result in a more efficient evacuation. This is because over time fires may spread quickly and approach zones that were initially far from the initial ignition points. Therefore, it is not advisable to make the persons in these zones wait too long before issuing an evacuation order if the fire is predicted to move quickly.

Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 3 The effect of street parking on evacuation varies with the situation. In some cases, such as the nighttime scenario for Ross Valley and the daytime scenario for Woodacre Bowl, where we found little impact on evacuation. While in the other cases, prohibiting street parking can result in a significant increase in evacuation efficiency. The difference results from having different numbers of evacuees and numbers of exit routes. When the number of evacuees is relatively small compared with the number of background vehicles and the number of exits is high, the effect of street parking is small. Yet, if there are more evacuees and fewer exits, the impact of street parking is high. Overall, prohibiting street parking on fire danger days can effectively reduce evacuees’ travel time. If possible, this strategy should be implemented in all cases. However, where that is not feasible, we recommend at a minimum prohibiting on-street parking during the daytime for Ross Valley, at night for Woodacre Bowl, during both the daytime and nighttime for Tamalpais Valley, and at night for the Novato Neighborhood. Prohibiting street parking on days when fire danger is high could significantly improve evacuation performance.

Agenda Item #11.1.

Contents
Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland-Urban Interface (WUI) Communities in California Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 5 Introduction
Wildfire evacuations have posed persistent challenges in California in recent years particularly in Wildland Urban Interface (WUI) communities where there is a significant mix of residences and flammable vegetation.

Current strategies rely on standard, but static approaches such as issuing evacuation orders by designated zones. However, these approaches are likely to be problematic in actual incidents, given the difficulties implementing them under rapidly moving fire conditions and fragile communications infrastructures.

Most small, resource-strapped WUI communities in Marin County do not have the resources to conduct dedicated evacuation studies. This research evaluates a set of generalizable evacuation strategies for four areas in Marin County: Ross Valley, Woodacre Bowl, Tamalpais Valley, and an area near Highway 101 and Ignacio Boulevard in Novato (hereafter referred to as “Novato Neighborhood”). The strategies include vehicle reduction from carpooling, off-street parking prohibitions, and phased evacuations, which were tested using a wildfiretraffic simulation framework. The strategies were evaluated based on their outcomes in several evacuation scenarios for each of the study communities, as well as their ease of incorporation into standard procedures.

Additionally, the impact of background traffic, which refers to the normal traffic passing through the case study areas when evacuations occur, was taken into account during the evaluation. This is important because the presence of background traffic can cause significant delays in evacuation, and previous research has not adequately considered its impact (Soga et al., 2021). Therefore, this research includes background traffic in the traffic simulation to more accurately evaluate the potential strategies. Based on our results, we provide recommendations for the different responses and strategies.

Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 6 Evaluating Wildfire Evacuation Strategies
Wildfire evacuation strategies have been largely informed by the experiences learnt from other types of disasters (e.g., hurricanes, tsunamis), but also have specific considerations given the dynamic characteristics of wildfire and the spatial areas they impact. Most strategies focus on how to improve evacuation efficiency to reduce evacuation time given the speed and short notice of wildfires, through measures such as temporarily widening lanes (e.g., by using shoulders and medians for vehicle travel). Other evacuation strategies are designed to (i) improve road network capacity through measures such as contraflow (reversing the direction of traffic on individual lanes) (Araujo et al, 2014, Chen et al., 2020) or prohibiting street parking, or (ii) optimize road use by evacuees through strategies such as carpooling, which reduces the number of departing vehicles, and phased evacuations that lower peak traffic volumes on roadways by spreading out evacuees.(Zhao and Wong 2021). On-street parking restrictions can increase the capacity of low-capacity roads (Cova 2005).

Phased evacuations can be staggered to permit those closer to fire danger to leave first and prevent bottlenecks (So and Daganzo 2010). Chen and Zhan (2008) tested all phased sequences in a road network with four evacuation zones and found that if the affected area has a high population density and the road network follows a grid structure, implementing a phased evacuation strategy that involves alternating non-adjacent zones can effectively reduce the overall evacuation time. Encouraging carpooling can also reduce the number of vehicles involved in an evacuation and effectively mitigate resulting traffic congestion (Zhao and Wong 2021).

Although there are many evacuation strategies proposed in existing research, there is no comprehensive study that compares the applicability and relative gains from each type of wildfire evacuation strategies under different network topology, evacuation demand, and fire spread dynamic situations. For many communities affected by wildfires, a preliminary understanding of the potential applicability of previously proposed strategies can still be greatly helpful for developing preliminary plans before carrying out further detailed investigations. One area that has often not received sufficient attention has been the effect of background traffic, which refers to the traffic that was unable to be redirected and passes through the evacuation area, despite its potentially huge impact on evacuation efficiency. For many fire-prone communities along major highways, discussions regarding background traffic are crucial, as during peak hours highways may lack capacity to accommodate the additional evacuation traffic.

Traffic simulations can be useful in testing the performance of various evacuation strategies, as shown in Table

1. Microscopic simulation software, such as Paramics (Cova and Johnson, 2002), can provide fast setup for the simulation model. A dedicated agent-based simulation model can provide greater flexibility and higher simulation efficiency (Soga et al., 2021; Zhao and Wong 2021). Quick, simplified manual calculations (Cova and Johnson, 2002) and adjusted four-step models (de Araujo et al., 2014), can be useful for developing a conceptual understanding of the evacuation situation and potential bottlenecks. The data required for running simulation models are often readily available from sources such as OpenStreetMap and digitized aerial imagery for network inputs, and regional parcel maps can be used for identifying residential locations for estimating the Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 7 number of persons needing to be evacuated. Models can generate aggregated metrics such as travel time, fire exposure time, and link-level congestion status. Altogether, traffic simulation is frequently applied across hazards, with a focus on investigating changes in evacuation performance metrics through parametric studies that analyze detailed evacuation choices, such as departure time, route, and destination (Zhao and Wong, 2021).

Other keys to developing wildfire evacuation strategies concern gathering information on evacuation behavior, which can be obtained through observations and surveys (Gwynne et al., 2023), mathematical models (Lovreglio et al., 2019; McCaffrey et al., 2018), and recently the application of emerging technology such as GPS (Zhao et al., 2022). One important consideration is when evacuees need to be notified of the evacuation warning or order which can be aided by information on how fires spread, and the operation of communication systems, which can be integrated into the evacuation simulations for a more comprehensive depiction of possible scenarios (Soga et al., 2021).

Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 8 Table 1. Selected Studies That Employ Simulations or Traffic Models for Studying Wildfire Evacuation (adapted from Zhao and Wong, 2021)
Reference Model Characteristics Metrics Strategies or
Scenarios
Traffic generation Departure time Destination and
route choices
Simulation
type
Cova and Johnson
(2002)
250 households; vehicles
per household follows
Poisson distribution
(mean: 0.5-3
vehicles/household)
Departure time
follows Poisson
distribution
(mean: 5-25 min
after fire occurs)
The neighborhood
exits or shelters;
dynamically
updated lowestcost route
Microscopic
(Paramics)
Clearance
time, mean
vehicle/
household
travel time
Setting different
number of evacuee
vehicles and
departure time
Cova and Johnson
(2003
30-150 vehicles per zone Uniformly
generated within
15 min
Various static
routing
(minimizing total
travel distance,
minimize merging
or balanced)
Microscopic
(Paramics)
and manual
capacity
analysis
Clearance
time, total
travel
distance,
number of
merges
Reducing
intersection merge
delays with turn
restrictions
Sbayti and
Mahmassani
(2006)
1,794-5,692 vehicles Simultaneous or
optimal
departure time
solved by an
optimization
model
System-optimal
traffic assignment
Microscopic
(DYNASMART
-P)
Clearance
time, mean
travel time
Phased evacuation
Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 9 Reference Model Characteristics Metrics Strategies or
Scenarios
Li et al. (2022) 47,410 vehicles Simultaneous or
optimal
departure time
solved by an
optimization
model
System-optimal
traffic assignment
Nonsimulation,
solved by an
optimization
model
Mean travel
time,
average fire
exposure
time
Phased evacuation
Zhao and Wong
(2021)
7,438 households;
44%/43%/13%
households leave with
1/2/3 vehicles
Departure time
follows normal
distribution (40
min±20 min)
Outside the fire
threat area,
updated lowestcost route
Spatialqueue-based
dynamic
simulation
model
Mean
evacuation
time, total
exposure
time
Slowing fire speed,
number of
evacuating vehicles,
phased evacuation
Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 10 Developing Evacuation Scenarios
The framework of the wildfire and traffic simulation used in this study is shown in Figure 1. The proposed framework covers three parts: (i) wildfire spread simulation, (ii) evacuation behavior, (iii) and traffic evacuation simulation. For the wildfire spread simulation, inputs include terrain topography, fuel (amount and condition of burnable material), and meteorological conditions along with fire ignition location and time. The outputs from this step analyze different aspects of the fire simulation, specifically, fire propagation divided into one-hour time intervals. These data provide fire information for evacuation planning. For the evacuation behavior step, three types of strategies were considered in this study: (a) vehicle reduction, (b) phased evacuation, and (c) prohibition of parking vehicles on the streets (and providing alternative parking to help move vehicles off the street). These strategies can reduce the number of vehicles needed for evacuation, quicken departure times, and increase road capacity, respectively. Such information also become inputs to the traffic simulation step.

For the traffic simulation, the model produces various evacuation efficiency evaluation metrics, such as the number of exposed vehicles at different times, average evacuation travel time, and average exposure time to the fire hazard. These metrics are used to evaluate the performance of the different evacuation strategies.

Figure 1. Study Flowchart
Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 11 Wildfire Spread Simulation
The fire simulation provides information on how quickly the fire may propagate through an area under different meteorological, topographical, and fuel conditions (Finney and Andrews, 1999). For this study, fire simulations were conducted using FARSITE software designed by M.A. Finney for spatial and temporal simulation of fire behavior under various conditions (Finney, 1998). The software requires a variety of input data regarding terrain topography, fuel, and meteorological conditions that are fundamental factors affecting fire behavior. The framework of the fire simulation is shown in Figure 2.

The first step was to define ignition points for each fire both by geography, identifying the areas involved in the fires, and timing, determined by analyzing data for specific days taken from the nearest weather stations.

Following that, data including terrain topography, fuel, and meteorology are collected to run the simulations.

Topographic data refers to the structure and shape of the terrain (such as elevation, slope, and aspect). Tree cover data includes canopy cover, stand height, canopy base height, and canopy bulk density. Fuel data was determined by using the 40 Scott and Burgan Fire Behavior Fuel Model (FBFM40), which divides the area into 40 different fuel types. For meteorological metrics, data on wind speed, wind direction, air temperature, humidity, and solar radiation data were collected and entered into FARSITE. Simulation-related parameters, such as simulation duration, are also set at this step. These steps lead to the fire simulation output, which is the fire propagation status (e.g., areas with active fires, flame length) at each timestamp.

Figure 2. Framework of Fire Simulation
Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 12 Spatial-Queue Based Traffic Simulation
A spatial queue based traffic model, developed by the research team (https://github.com/cbcities/spatial_queue), was used to simulate the evacuation process. We used this model rather than popular microscopic simulators based on car-following and lane-changing behavior since it is less data intensive and easier to program. The simulation model consists of three basic elements: (i) vehicles, (ii) intersections (nodes), and (iii) road links. Figures 3-4 illustrate features of the node model and the link model.

Each vehicle is treated as an independent actor. Each vehicle is assigned an initial evacuation route, and the routes are updated every 10 seconds based on the evolving traffic status (e.g., level of congestion), in a manner that imitates how route recommendations are updated in real time in navigation apps like Google Maps, Apple Maps, and Waze.

Each link in the road network corresponds to a stretch of road between two intersections. The model simulates the time that it takes for a vehicle to move from the upstream node to the downstream node, creating queues and spillbacks, according to the actions of all other vehicles. Vehicles spend at least as long as the free flow travel time on a link before leaving the link or joining a queue at the downstream end. When the end of the queue formed by vehicles reaches back to the upstream end of the link, spillback occurs, and no more vehicles can enter this link.

The link flow capacity is assumed to be 1,900 vehicles/(hour∙lane), and link capacities are imposed in a flip-coin probability manner at each one-second time step, with a probability of a queuing vehicle at the front leaving the current link or entering the next link at 0.53 vehicles/(second∙lane). The model moves vehicles at the front of each link to the next link at each one-second time step, if there is room in that link, and they do not conflict with other vehicles moving through the intersection at that time step, such as those turning left or entering from perpendicular directions.

Vehicles entering an intersection are assumed to have equal priority, except for vehicles already in roundabouts, which have higher priority than vehicles feeding into the roundabouts. All intersections were modeled as non-signalized in anticipation of power failures.

Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 13 Figure 3. An Illustrative Node Model
Figure 4. An Illustrative Link Model
Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 14 Evacuation Strategies and Scenario Development
This study compared the effectiveness of various evacuation strategies/policy options by testing different scenarios under specified conditions. The scenarios consisted of three types:
• Vehicle reduction. As a small increase in the number of vehicles on the roads during the evacuation can significantly increase congestion, this scenario tests whether and how much carpooling can make a difference on evacuation efficiency (Wong and Shaheen, 2019). • Phased evacuation. Many studies have demonstrated the effectiveness of phased evacuation, as it can reduce the traffic congestion during the evacuation (So and Daganzo, 2010, Chen and Zhan, 2008).

While some evacuees further from the fire danger are delayed, allowing those closer to the fire to leave first reduces exposure time and limits the amount of downstream congestion so those leaving later can evacuate more quickly. In this study, we quantify the effects of the phased evacuation strategy on evacuation.
• Prohibiting on-street parking on "Red Flag" days. On-street parking reduces the number of accessible lanes, which in turn reduces road capacity and storage, causing delays in the evacuation progress (Cao et al., 2017, Wijayaratna and Wijayaratna, 2016). Therefore, this scenario analyzes the quantitative impact of prohibiting on-street parking when fire danger is high (i.e., Red Flag Days).

For each set of strategy variables, base case values were chosen for comparison. Details of each strategy are given in Table 2.

Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 15 Table 2. Description of Scenarios
Strategy Options (baseline value
underlined)
Description
Vehicle reduction 1 person/vehicle 1. Reduce the number of vehicles used for evacuation, mainly through carpooling by each household or several nearby households.

2. The baseline case assumes that there is no car-pooling, everyone drives a vehicle.

3. The alternative cases assume increasing levels of carpooling. 2 persons/vehicle
3 persons/vehicle
4 persons/vehicle
5 persons/vehicle
Phased evacuation 0 min 1. Evacuation zone is divided into three zones based on distance to the fire origin.

2. During the day, it is assumed that people can receive the warnings quickly and prepare for the evacuation swiftly, while during the night, evacuee’s response time of becomes longer as they need more time to prepare.

3. The baseline case assumes “no phased evacuation:” all vehicles in the three zones have the same departure time.

4. The alternative cases assume that the evacuation orders are staggered for the three zones at certain time intervals with those zones closest to the fire going first.
15 mins (day) or 20 mins (night)
30 mins (day) or 40 mins (night)
45 mins (day) or 60 mins (night)
Prohibiting onstreet parkingNo street parking 1. On-street parking can affect the number of lanes of roads and further limit road capacity (vehicles/hour) and road storage (vehicles/km).

2. The baseline case assumes there is no on-street parking.

3. The alternative case assumes that all residential roads have on-street parking. According to some studies, street parking can reduce the road capacity by 16-35% (Biswas et al., 2017, Wijayaratna, 2015). As there is no research on street parking in Marin County, we used the average of the reduction values from the existing studies, which is 25%.

With street parking
Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 16 Evaluation of Evacuation Strategies
In this study, two commonly used evaluation performance indicators were employed to evaluate the outcomes under different evacuation strategies: average travel time and average exposure time to the fire hazard. The first metric represents the average time spent in the evacuation process, from leaving the origin to reaching a point at least two miles away from the dynamic fire area. We assumed that reaching two miles away guarantees the evacuees’ safety. The second metric is the average fire exposure time, which is represented by the average duration of time that evacuees are within a half-mile radius of the active fire front. Since this area is considered especially dangerous, it is crucial to know how long an evacuee stays in such a hazardous zone, which reflects the average level of threat that they face. Figure 5 illustrates an example of how to calculate the travel time and exposure time.

Figure 5. Illustration of Evacuation Time and Exposure Time Calculations Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 17 Case Studies
Given the multiplicity of small communities in WUI areas in California and their susceptibility to wildfire threats and losses each year, it is imperative to work with local communities and their planning and fire prevention agencies to find practical and generalizable solutions to address wildfire threats. This study compared the effectiveness of different evacuation response/policy options through scenario testing with control variables.

In previous research, we recognized that the randomness and unpredictability of wildfires, insufficient communications options, as well as the disproportional high evacuation demand in WUI areas are the main factors causing bottlenecks in the evacuation process (Zhao and Wong, 2021). Recognizing the need for designing and upgrading wildfire evacuation strategies for small WUI communities, we partnered with the Marin Wildfire Prevention Authority (MWPA), the single agency in Marin County responsible for preventing wildfire threats, to test the applicability of some commonly adopted evacuation measures over a wide set of case study characteristics and scenarios. MWPA identified four neighborhoods in Marin County as representative case study areas (Figure 6). These four sites represent diverse fire, and traffic characteristics in high-risk communities.

Figure 6. Four Study Areas in Marin County and Fire Hazard Areas Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 18 Scenario Development
This section explains the development of the various scenarios. We first constructed a model of the existing road network. We then chose a fire ignition location for each study area and developed a model to simulate the fire’s propagation and obtained data on the background traffic travelling though Marin County. Finally, we calculated the number of persons and vehicles that would be evacuating each study area during the daytime and nighttime scenarios.

Road Network
The simulated evacuations occur in Marin County, CA. Since background traffic may significantly affect evacuation efficiency the road network used in the simulation models includes not only the four study areas, but all roads in Marin County. The road network for the study area was obtained from OSMnx, a python package (https://osmnx.readthedocs.io/), and then a directed node-and-link-based road network for Marin County was obtained (see Figure 7). Nodes and links are commonly used structures to represent a network, where a node represents one road intersection, and a link represents the stretch of road between two intersections. The final Marin County Road network consists of 40,209 links and 17,857 nodes.

Figure 7. The Road Network of Marin County
Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 19 Fire Propagation
As introduced in the section entitled Wildfire Spread Simulation, several inputs are required to run the fire propagation simulations for the Marin County cases, including ignition locations, day and time of the fire (to retrieve meteorological information), data on terrain topography, and vegetation/fuel status. The ignition points (shown in Figure 8) were chosen to be within a two-mile radius from selected city boundaries and contain fuels such as grass or shrubs. Terrain topography and fuel data were collected from the LandFire database (http://landfire.gov/). The meteorological data were obtained from the FAMWeb Data Warehouse database (http://www.wildfire.gov/application/famweb-data-warehouse) for three stations: Station 042312 - Middle Peak; Station 042309 - Woodacre; Station 042310 - Big Rock. These stations were considered the most suitable for the proposed scenarios due to their proximity and data availability. The locations of these weather stations are also shown in Figure 8.

The typical wildfire season runs from May to November in California, though recently year-long fire threats are becoming more common. For this study, the hypothetical fire event was set to occur on August 8, 2020, the day when the Lightning Complex Fires occurred. The meteorological data for the study areas were extracted from the above identified weather stations for this day.

For the four study areas in Marin County most residents typically work outside the area during the day and return in the evening. Given this, we chose two ignition times: 1 am, when most people are at home and the evacuation travel demand is the highest, and 9 am, when most people have left for work or other places and fewer evacuation vehicles are needed. This produced a total of eight simulations for the four case study areas, with two scenarios per area (daytime and nighttime).

The meteorological conditions for the fire simulations were slightly adjusted for some scenarios to produce ‘dangerous’ or ‘worst case’ scenarios, which would trigger evacuations. In some cases, the fire simulations using the actual recorded data would generate fire propagation scenarios that reach the city's boundaries; in others, this was not possible within the simulation time window of 12 hours. In the case of Ross Valley, fire simulations using the actual recorded meteorological data were a half mile away from the WUI boundary of the town, which likely would not trigger an evacuation. For this reason, the wind direction was slightly adjusted to create fire spread that reaches the Ross Valley boundary. The actual recorded meteorological parameters for Ross Valley on Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 20 August 8, 2020, and the modified inputs for the fire simulation are given in Table 3. Figures 9-12 illustrate the fire progression for Ross Valley, Woodacre Bowl, Tamalpais Valley, and the Novato Neighborhood, respectively.

Figure 8. Ignition Points for the Four Cases and the Locations of Weather Stations Table 3. Modified Meteorological Parameters for the Ross Valley Scenario Date Time Temperature
(ºF)
RH (%) Wind Speed
(GUTS) mph
Real Wind
Direction (º)
Used Wind
Direction (º)
2020/8/18 01:00 a.m. 82 36 18 304 218
2020/8/18 02:00 a.m. 82 36 18 297 215
2020/8/18 03:00 a.m. 81 39 16 311 221
2020/8/18 04:00 a.m. 80 41 18 297 215
2020/8/18 05:00 a.m. 80 39 18 324 234
2020/8/18 06:00 a.m. 80 37 19 324 234
2020/8/18 07:00 a.m. 80 41 19 305 215
Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 21 Date Time Temperature
(ºF)
RH (%) Wind Speed
(GUTS) mph
Real Wind
Direction (º)
Used Wind
Direction (º)
2020/8/18 08:00 a.m. 82 42 17 335 245
2020/8/18 09:00 a.m. 86 31 16 322 232
2020/8/18 10:00 a.m. 88 22 18 313 223
2020/8/18 11:00 a.m. 93 26 17 324 234
2020/8/18 12:00 p.m. 91 25 14 330 240
2020/8/18 01:00 p.m. 93 23 19 332 242
2020/8/18 02:00 p.m. 94 20 21 314 224
2020/8/18 03:00 p.m. 92 20 22 298 215
2020/8/18 04:00 p.m. 88 22 28 309 219
2020/8/18 05:00 p.m. 87 17 36 303 215
2020/8/18 06:00 p.m. 84 16 39 311 221
2020/8/18 07:00 p.m. 82 15 36 327 237
2020/8/18 08:00 p.m. 80 17 39 283 230
(a) Fire Progression from 1 a.m. (b) Fire Progression from 9 a.m.

Figure 9. Fire Propagation for Ross Valley Scenarios
Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 21 (a) Fire Propagation from 1 a.m. (b) Fire Propagation from 9 a.m.

Figure 10. Fire Propagation for Woodacre Bowl Scenarios (a) Fire Propagation from 1 a.m. (b) Fire Propagation from 9 a.m.

Figure 11. Fire Propagation for Tamalpais Valley Scenarios Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 23 (a) Fire Propagation from 1 a.m. (b) Fire Propagation from 9 a.m.

Figure 12. Fire Propagation for Novato Neighborhood Scenarios Background Traffic
The background traffic origin/destination (OD) demand data was obtained from the Metropolitan Transportation Commission (MTC) at https://mtcdrive.app.box.com/v/pba50-2015-TM152-IPA-17. An activity-based travel model was used to simulate the travel-related choices of Bay Area residents. The model used data from the MTC 2015 Bay Area Travel Behavior survey, which included two-day travel diaries from over 15,000 households. The model’s unit of analysis was a “tour” which represents a closed or half-closed chain of trips starting and ending at home or the workplace; it includes at least one destination and at least two successive trips. The model generated tour and trip lists for each resident. The individual and joint trips were later aggregated into origin-destination metrics.

The MTC traffic OD data covers trips in the whole Bay Area. However, in this study, our focus was limited to Marin County. Therefore, trips which do not pass through Marin County were filtered out. The remaining trips that did pass through Marin County were truncated to keep only the trip segments within Marin County to reduce the subsequent computing time. This process is illustrated in Figure 13. After the pre-processing, the number of background OD trips during one modeled day was 823,916. The spatial and temporal distributions of the background traffic OD is shown in Figures 14-15. The ODs are concentrated in the eastern part of the Marin County. The largest number of trips were those originating from/destined to the Golden Gate Bridge and Richmond-San Rafael Bridge. During the peak hours from 7 a.m.-10 a.m. and 3 p.m.-6 p.m., the number of vehicles accounted for around 50 percent of the total background vehicles in a day. If a wildfire occurs during peak hours, then the presence of background vehicle traffic could have a major impact on the evacuating vehicles.

Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 24 Figure 13. Change of O/D which are outside Marin County to the border of Marin Figure 14. The Spatial Distribution of Background OD pairs within Marin County Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 25 Figure 15. Temporal Distribution of Background OD Pairs within Marin County Evacuee OD Demand
The number of residents who need to be evacuated is different between the daytime scenario and the nighttime scenario because many residents would be at work during the day, while most would be at home at night. From the background OD data, the location where each person begins their day is known and this place is regarded as their residence. Using the background OD data, it was possible to determine how many residents were at home when the fire ignited, and these residents were classified as evacuees.

Table 4 provides a summary of the number of evacuees for each scenario.

The simulated evacuation destinations were set on the border of Marin County and other counties, as depicted in Figure 16. Four evacuation destinations were available for evacuees to choose from, and the assumption was that individuals would select the closest option to them. Generally, the evacuees from Tamalpais Valley would choose the southern two destinations (Golden Gate Bridge and Richmond-San-Rafael Bridge) as their destinations; evacuees from Ross Valley and Woodacre Bowl would choose the middle destination; and evacuees from the Novato Neighborhood would choose the northern two destinations (Redwood Highway and Sears Point Bridge) and the middle destination (Richmond-San-Rafael Bridge). The treatment of destination choices is simplistic and does not consider factors such as shelter availability, proximity to resources, and destination safety, which all may affect an evacuee's decision. To minimize these outside influences, this study focused mainly on the most dangerous part of the trip, the time taken to reach a point two miles away from the Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 26 final fire extent. (The time taken to reach the destination is not reported here, as it provides minimal information on the evacuees' risks due to the proximity to the fires.) The evacuees’ departure times were based on a truncated normal distribution (i.e., truncated around the mean ± one standard deviation). The means and the standard deviations of the departure times are different for the nighttime evacuation and daytime evacuation scenarios. We assumed that during the day people are more likely to receive fire alert messages from various media promptly and take action more quickly. Conversely, in the late evening people are likely asleep and less likely to be aware of such messages. Therefore, for the daytime evacuation, the mean departure time was set to 9:10 a.m. ± 10 minutes, while for the nighttime evacuation, the mean departure time was set to 1:20 a.m. ± 20 minutes.

Table 4. Summary of Number of Evacuees for Different Scenarios Study area Fire ignition
time
# residents # residents who have departed to
other places before the fire occurs
# evacuees
Ross Valley 9 a.m. 4,893 2,928 1,965
1 a.m. 4,893 0 4,893
Woodacre
Bowl
9 a.m. 2,196 840 1,356
1 a.m. 2,196 0 2,196
Tamalpais
Valley
9 a.m. 5,492 3,482 2,010
1 a.m. 5,492 0 5,492
Novato
Neighborhood
9 a.m. 17,519 9,857 7,662
1 a.m. 17,519 0 17,519
Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 27 Figure 16. Evacuation Destinations
Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 28 Validation of the Simulation
Before evaluating the evacuation strategies, we conducted a validation analysis on the "no evacuation" scenario, where only background traffic OD data was used to run the traffic simulations, and the simulated traffic volumes at key locations on several main roads at different times were compared with real observations (traffic count data) collected by Fehr&Peers (https://www.fehrandpeers.com/). The comparison results are presented in Table 6 and 6. These two tables show the differences between the traffic volume generated by simulation and count data during morning peak hours and afternoon peak hours. As shown, the percentage difference in volume on most roads is less than 20 percent. According to the 2017 California Regional Transportation Plan Guidelines, a commonly used criterion for volume-count validation is a percentage root mean square error less than 40 percent (Commission, 2017). The value of the percentage root mean square error of our simulation is 18.9 percent, much less than the threshold. Therefore, our simulation performs well.

Table 5. Comparison of Model Traffic and Counts from Roadside Detectors during Morning Peak # Type Direction Location # vehicles during morning
peak period (6 a.m.-10 a.m.)
Percentage
difference
Model Count
1 Mainline North
bound
US 101 at San Francisco
/Marin County Line
12,784 10,000 27.84%
2 Mainline South
bound
21,094 17,984 17.29%
3 Mainline North
bound
US 101 at Corte Madera
Creek
16,591 13,000 27.60%
4 Mainline South
bound
25,718 20,000 28.59%
5 Mainline North
bound
US 101 at San Rafael
Creek
16,042 19,000 -15.57%
6 Mainline South
bound
23,148 28,000 -17.33%
7 Mainline East bound I-580 at Contra
Costa/Marin County Line
8,601 7,900 8.87%
8 Mainline West bound 13,947 12,800 8.96%
9 Mainline North
bound
US 101 at Sonoma
/Marin County Line
6,607 8,100 -18.43%
10 Mainline South
bound
11,031 11,000 0.28%
Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 29 Table 6. Comparison of Model Traffic and Counts from Roadside Detectors during Afternoon Peak # Type Direction Location # vehicles during afternoon
peak period (3 p.m.-7 p.m.)
Percentage of
the difference
Model Count
1 Mainline North
bound
US 101 at San Francisco
/Marin County Line
17,201 19,000 -9.47%
2 Mainline South
bound
13,364 13,000 2.80%
3 Mainline North
bound
US 101 at Corte Madera
Creek
20,304 21,000 -3.31%
4 Mainline South
bound
21,212 20,000 6.06%
5 Mainline North
bound
US 101 at San Rafael Creek 20,522 22,000 -6.72%
6 Mainline South
bound
17,477 20,000 -12.62%
7 Mainline East
bound
I-580 at Contra
Costa/Marin County Line
12,802 12,700 0.80%
8 Mainline West
bound
9,537 8,600 10.90%
9 Mainline North
bound
US 101 at Sonoma /Marin
County Line
8,576 11,900 -27.93%
1 0
Mainline South
bound
5,756 8,000 -28.05%
Impact of Background Traffic
This section analyzes the impact of background traffic on evacuation efficiency. We conducted comparative experiments to show the difference in evacuation time between the scenarios with background traffic and without background traffic at each of the four study sites. The results are shown in Figures 17-20. These figures show the number of evacuees that reach their destinations over time with/without background traffic and during the daytime and nighttime scenarios for the four study sites. Table 7 shows the average travel times for each scenario.

These figures show that background traffic has almost no impact on nighttime evacuations (the nighttime. curve is entirely covered up by daytime curve) but has a significant impact on daytime evacuations. The results Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 30 make sense as there is little background traffic at 1 a.m. and consequently, there would be little effect on evacuation times. In daytime, there are many more commuting vehicles than evacuating vehicles; therefore, the background traffic would have a large influence. Table 7 shows that, when the evacuation/fire ignition starts at 9 a.m., the average evacuation times with background traffic are one and a half hours more for Ross Valley, one- and three-quarter hours more for Woodacre Bowl, one third of an hour more for Tamalpais Valley, and a half hour more for the Novato Neighborhood than without background traffic. These differences highlight the importance of taking background traffic into account when developing evacuation strategies, as it can significantly affect evacuation efficiency.

The impact of background traffic on the evacuation of Ross Valley and Woodacre Bowl is much larger than that of Tamalpais Valley and the Novato Neighborhood. Figure 21 shows that the exits for Ross Valley and Woodacre are concentrated on Sir Francis Drake Blvd, which is a very busy main road. When an evacuation is ordered, it will cause a heavy traffic jam on this road. Compared with these two places, Tamalpais Valley and Novato Neighborhood have many widely dispersed exits. Even if the background traffic volume is high, the evacuees can find many exit routes, hence, the impact of background traffic on the evacuation is comparatively small in these areas.

Overall, background traffic needs to be taken into account when planning for daytime evacuations, but its impact differs with the topology of the road networks.

Figure 17. Arrival Times for Ross Valley
Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 31 Figure 18. The Arrival Times for Woodacre Bowl
Figure 19. The Arrival Times for Tamalpais Valley
Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 32 Figure 20. Arrival Times for Novato Neighborhood
Table 7. Comparison of Average Evacuation Time With and Without Background Traffic Study site Fire ignition time Average evacuation time [h] Difference With background Without
background
Absolute Percent
Ross Valley 1 a.m. 1.30 1.30 0 0%
9 a.m. 1.79 0.28 1.51 532%
Woodacre Bowl 1 a.m. 0.71 0.71 0 0%
9 a.m. 2.04 0.33 1.71 518%
Tamalpais Valley 1 a.m. 0.58 0.58 0 0%
9 a.m. 0.49 0.16 0.33 206%
Novato
Neighborhood
1 a.m. 1.21 1.21 0 0%
9 a.m. 0.82 0.34 0.48 141%
Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 33 (a) topology and exits for Ross Valley. (b) topology and exits for Woodacre Bowl (c) topology and exits for Tamalpais Valley (d) topology and exits for Novato Nhbd Figure 21. Network Topologies and Exits for the Four Case Study Areas Strategy Analysis
The purpose of this study was to understand the applicability of commonly used evacuation strategies under different settings. These strategies, including vehicle reduction, phased evacuation, and banning on-street parking were shown in Table 2 above. Each alternative strategy option was simulated, and the results compared against the base scenario (the underlined values in Table 2). The following sections present evacuation performance metrics and other results for the baseline scenario and the various alternative scenarios.

Recommendations are given at the end.

Baseline
The base cases for the wildfire simulation for the four case studies are: no car-pooling, no phased evacuation, and no on-street parking. Figures 22-26 show the number of evacuees that reach a point at least two miles Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 34 outside the fire area over time, number of evacuees still within a half mile of the fire area over time, and the cumulative person-hours of exposure to the fire over time for the four case studies.

For the daytime evacuation, the distribution of departure times follows a normal distribution with a mean value of 9 hours 20 minutes and a standard deviation of 10 minutes. For the nighttime evacuation, the distribution follows a normal distribution with a mean of 1 hour 40 minutes and a standard deviation of 20 minutes. This is consistent with the research of Zhao et al. (2021) which assumed that departure time would follow a normal distribution since once residents receive an evacuation notification, most will need time to prepare to leave and this will delay the peak departure time.

Figures 22(a)-(f) illustrate how the evacuation process for the Ross Valley proceeds. For the daytime evacuation scenario (fire ignites at 9 a.m.), all vehicles would reach two miles away from the fire area before 11:11 a.m. (solid blue line in Figures 22(a)-(b), and the average evacuation time would be one and a quarter hour. For the nighttime evacuation scenario (fire ignites at 1 a.m.), all vehicles would reach two miles away from the fire area before 4:15 a.m. (solid blue line in Figure 22(b)), and the average evacuation time would also be about one and a quarter hour. The average evacuation time for the nighttime scenario is similar to that during the day despite the much-reduced background traffic, mainly because of the higher number of evacuees in the nighttime scenario. The impact of the background traffic is nevertheless observed in another of the daytime scenarios. Figure 22(c) shows that, during the period of 9:30 a.m.-10 a.m., very few additional vehicles leaving the fire danger area make it beyond one half mile of the fire because background vehicles block the evacuation routes during these morning peak hours. However, this does not occur in the nighttime case (Figure 22(d)) when background traffic is far less. Figures 22(e)-(f) present the cumulative exposure time of all the evacuees for the daytime and nighttime cases. The cumulative exposure time when all the evacuees reach the destination is about 230 hours for the daytime scenario and 1050 hours in nighttime scenario. On average, though, evacuees in the nighttime scenario are less exposed to the fire, about seven minutes per person compared to 13 minutes for those in the daytime scenario.

Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 35 (a) #vehicles that reach 2 miles away from the fire area (daytime scenario)
(b) #vehicles that reach 2 miles away from the fire area (nighttime scenario)
(c) #persons within 0.5 mile of fire (daytime
scenario)
(d) #persons within 0.5 mile of fire (nighttime
scenario)
(e) Cumulative fire exposure (hours) (daytime
scenario)
(f) Cumulative fire exposure (hours) (nighttime
scenario)
Figure 22. Baseline Case for Ross Valley
Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 36 For the Woodacre Bowl case, if the fire ignites at 9 a.m., all vehicles reach two miles away from the fire area before 10:30 a.m., while if the fire ignites at 1 a.m., all vehicles reach two miles away before 2:30 a.m. Figures 23(c)-(d) show that the number of evacuees within a half mile of the fire first increases and then plummets. The reason for the initial increase is that the fire ignites one mile away from the residential area but covers the sole evacuation road. Therefore, evacuees need, first, to enter the fire area and then evacuate to the safe locations.

Figures (e)-(f) present the cumulative exposure time of all the evacuees for the daytime and nighttime scenarios. Cumulative exposure when all the evacuees reach the safe destination is about 100 hours in the daytime scenario and 120 hours in the nighttime scenario. On average, each evacuee is exposed to the fire for around four and a half minutes in the daytime scenario and just over three minutes in the nighttime scenario.

Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 37 (a) #vehicles that reach 2 miles away from the fire area (daytime scenario)
(b) #vehicles that reach 2 miles away from the fire area (nighttime scenario)
(c) #persons within 0.5 mile of fire (daytime
scenario)
(d) #persons within 0.5 mile of fire (nighttime
scenario)
(e) Cumulative fire exposure (hours) (daytime
scenario)
(f) Cumulative fire exposure (hours) (nighttime
scenario)
Figure 23. Baseline Case for Woodacre Bowl
Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 38 For the Tamalpais Valley case, if the fire ignites at 9 a.m., all vehicles will reach two miles from the fire before 10:20 a.m., while if the fire ignites at 1 a.m., all vehicles reach two miles away before 3 a.m. The evacuation time in the nighttime scenario is longer than in the daytime scenario because the number of evacuees is larger in the nighttime scenario. Even though there is little background traffic in the nighttime, the combined effect of having so many evacuees severely affect evacuation efficiency. Figures 24(c)-(d) illustrate that the evacuees move out of the danger area quickly within 45 minutes in the daytime, but it takes 100 minutes after the fire ignites in the nighttime. Figures 24(e)-(f) present the cumulative exposure time of all evacuees for the daytime and nighttime cases. The cumulative exposure when all evacuees reach the safe destination is about 90 hours in the daytime scenario and 650 hours in the nighttime scenario. On average, the evacuees in the nighttime scenario are exposed to the fire longer, for around three minutes, compared to seven minutes for those in the daytime scenario.

Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 39 (a) #vehicles that reach 2 miles away from the fire area (daytime scenario)
(b) #vehicles that reach 2 miles away from the fire area (nighttime scenario)
(c) #persons within 0.5 mile of fire (daytime
scenario)
(d) #persons within 0.5 mile of fire (nighttime
scenario)
(e) Cumulative exposure (hours) (daytime scenario) (f) Cumulative exposure (hours) (nighttime scenario) Figure 24. Baseline Case for Tamalpais Valley
Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 40 For the Novato Neighborhood case, if the fire ignites at 9 a.m., all vehicles reach two miles away before 11:40 a.m., while if the fire ignites at 1 a.m., all vehicles reach two miles away before 5:00 a.m. Figures 25(c)-(d) show that some evacuees become blocked from 2:00-3:00 a.m. Since there is little background traffic at this time, the backup is presumably due to downstream evacuating vehicles blocking the way of the upstream vehicles.

Figures 25(e)-(f) present the cumulative exposure time for all evacuees to reach the designated safe destination in the daytime is about 50 hours and 500 hours in the nighttime. On average, each evacuee is exposed to the fire for around one minute in the daytime and two minutes in the nighttime.

Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 41 (a) #vehicles that reach 2 miles away from the fire area (daytime scenario)
(b) #vehicles that reach 2 miles away from the fire area (nighttime scenario)
(c) #persons within 0.5 mile of fire (daytime
scenario)
(d) #persons within 0.5 mile of fire (nighttime
scenario)
(e) Cumulative exposure (hours) (daytime scenario) (f) Cumulative exposure (hours) (nighttime scenario) Figure 25. Baseline Case for Novato Neighborhood
Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 42 Vehicle Reduction
In this section we analyze the impact of reducing the number of vehicles involved in an evacuation through carpooling based on our model. Figures 26-29 show the number of evacuees that would reach two miles away from the fire area over time from the fire’s ignition, number of evacuees still within a half mile of the fire area over time, and cumulative exposure time to the fire over time depending on whether one, two, three, four or five persons share a single vehicle for each of the four case studies. Table 8 and Figures 30-31 present the average travel time and average exposure time for each of the five different carpooling assumptions.

For the Ross Valley case, Figure 26 shows that if each vehicle has an average of one, two, or five persons, all evacuees will reach two miles away from the fire area before 11:10 a.m., 10:40 a.m., and 10:10 a.m., respectively in the daytime scenario and all evacuees will reach two miles away from the fire area before 4:20 a.m., 2:50 a.m., and 2:20 a.m., respectively in the nighttime scenario. The number of persons who remain within a half mile of the fire area is lowest with four or five persons sharing one vehicle and nearly all leave the danger area much more quickly, while having only one person in each vehicle strands most evacuees in the congested danger area for around one and a half hours in the daytime scenario and two and a half hours in the nighttime scenario.

Table 8 shows that the average evacuation time decreases with more carpoolers. For each additional person sharing a vehicle, the average travel time will be reduced by about 10 percent. Further, average fire exposure time is reduced from about seven minutes without carpooling to about four minutes with two persons sharing one vehicle. When three or more persons share one vehicle, the average exposure time drops to about three minutes. With three or more persons per vehicle the fire exposure threat for all evacuees is a bit less than three minutes.

For the nighttime scenario, the average travel time significantly decreases to about 32 minutes if an average of two persons shares one vehicle, compared with around 74 minutes when there is no carpooling. Further, the average exposure time is reduced to about 5 minutes. When every vehicle is shared by three or more persons, the average travel time is reduced to a little over 15 minutes and the average exposure time stays the same at about 5 minutes.

Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 43 (a) #vehicles that reach 2 miles away from the fire area (daytime scenario)
(b) #vehicles that reach 2 miles away from the fire area (nighttime scenario)
(c) #persons within 0.5 mile of fire (daytime
scenario)
(d) #persons within 0.5 mile of fire (nighttime
scenario)
(e) Cumulative exposure (hours) (daytime
scenario)
(f) Cumulative exposure (hours) (nighttime scenario)
Figure 26. Vehicle Reduction Strategy Performance Measures for Ross Valley For the Woodacre Bowl case, Figure 27 shows that if each vehicle carries an average of one, two, three, four or five persons, all evacuees reach two miles away from the fire area before 10:30 a.m., 9:50 a.m., 9:40 a.m., 9:40 a.m., and 9:40 a.m., respectively in the daytime scenario and two miles away from the fire area before 2:30 a.m., 2:25 a.m., 2:25 a.m., 2:25 a.m. and 2:25 a.m., respectively in the nighttime scenario. The number of persons who remain within a half mile of the fire area is lower if there are at least two persons per vehicle and nearly all evacuees leave the fire danger area much more quickly, while the having only one person in each vehicle strands most evacuees in the congested danger area for around 0.75 hour and 1.5 hours, respectively in the daytime and nighttime scenarios.

Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 44 Table 8 shows that in the daytime scenario, the average evacuation time is reduced nearly half when there are at least two persons per vehicle. When there are three persons per vehicle the time for everyone to reach a safe location is reduced by over half, though there is not much additional improvement in evacuation efficiency when there are four or more persons per vehicle. The same is true for average exposure time which remains about three minutes when there are more than three persons per vehicle. In the nighttime scenario, attaining at least two persons per vehicle can already significantly reduce traffic jams and speed up the evacuation progress; reducing the average evacuation time by about 47 percent and limiting the average exposure time by around 30 percent. Again, any additional increase in carpooling provides only marginal improvement.

Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 45 (a) #vehicles that reach 2 miles away from the fire area (daytime scenario)
(b) #vehicles that reach 2 miles away from the fire area (nighttime scenario)
€ #persons within 0.5 mile of fire (daytime scenario) (d) #persons within 0.5 mile of fire (nighttime scenario)
(e) Cumulative exposure (hours) (daytime scenario) (f) Cumulative exposure (hours) (nighttime scenario) Figure 27. Vehicle Reduction Strategy Performance Measures for Woodacre Bowl Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 46 For the Tamalpais Valley case, Figure 28 shows that if evacuating vehicles carry an average of one, two, three, four or five persons, all evacuees will reach two miles away from the fire area before 10:20 a.m., 10:00 a.m., 10:00 a.m., 9:50 a.m., and 9:50 a.m., respectively in the daytime scenario and will reach two miles away from the fire area before 3:00 a.m., 2:40 a.m., 2:30 a.m., 2:20 a.m. and 2:20 a.m., respectively in the nighttime scenario. With at least two persons per vehicle nearly all evacuees leave the fire danger area much more quickly, while having only one person in each vehicle strands most evacuees in the fire danger area for around three quarters of an hour in the daytime scenario and one and three quarters hours in the nighttime scenario.

Table 8 shows that, in the daytime scenario, carpooling improves evacuation efficiency. With an average of two persons sharing one vehicle the average evacuation time is 21 minutes, about 16 minutes less then without carpooling. Little improvement results from having three persons per vehicle but having five riders per vehicle can shave an additional seven minutes off the evacuation time. Average exposure time is less than three minutes for all alternatives. Simulation results indicate that even without carpooling there is almost no heavy congestion within a half mile of the fire center. In the nighttime scenario, having an average of two persons sharing one vehicle reduces the average evacuation time by half from 44 minutes to 22 minutes and average exposure time from seven minutes to six minutes. With more persons per vehicle, the average evacuation time continues to decrease to about 13 minutes when all vehicles carry five persons. However, the average exposure time remains about six minutes which suggests that evacuation efficiency is maximized with an average of just two persons per vehicle.

Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 47 (a) #vehicles that reach 2 miles away from the fire area (daytime scenario)
(b) #vehicles that reach 2 miles away from the fire area (nighttime scenario)
(c) #persons within 0.5 mile of fire (daytime
scenario)
(d) #persons within 0.5 mile of fire (nighttime
scenario)
(e) Cumulative exposure (hours) (daytime scenario) (f) Cumulative exposure (hours) (nighttime scenario) Figure 28. Vehicle Reduction Strategy Performance Measures for Tamalpais Valley Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 48 For the Novato Neighborhood case, Figure 29 shows that if the evacuating vehicles carry an average of one, two, three, four or five persons, all evacuees will reach two miles away from the fire area before 11:40 a.m., 10:40 a.m., 10:20 a.m., 10:00 a.m., and 10:00 a.m., respectively in the daytime scenario and will reach two miles away from the fire area before 5:00 a.m., 3:20 a.m., 3:00 a.m., 2:50 a.m. and 2:40 a.m., respectively in the nighttime scenario. For the daytime strategy, carpooling has little effect on the number of evacuees reaching a half mile from the fire area over time or the cumulative exposure time. However, for the nighttime scenario, having only one person per vehicle traps over 150 persons within the fire danger area for almost 1.5 hours.

Table 8 show that for both the daytime and nighttime scenarios average evacuation times decrease with more than two persons per vehicle, but average fire exposure times do not. Having an average of at least three persons per vehicle in the daytime scenario and a least two persons per vehicle in the nighttime scenario can reduce the average evacuation time by over 50 percent.

Overall, the modeling shows that when the number of evacuating vehicles is reduced by 1/2, 1/3, 1/4, and 1/5, evacuation times can be reduced by 50, 66.7, 75, and 80 percent, respectively. The only exception is the scenario for the Ross Valley case study. Beyond a certain level, average evacuation times will not continue to decrease in some circumstances (e.g., from two persons/vehicle to five persons/vehicle in Woodacre Bowl nighttime scenario) because, other factors, such as the amount of background traffic, may be difficult to overcome, For example, there are only three exits in Woodacre Bowl (see Figure 21), and most of the evacuees will use the middle exit, which leads to congestion.

In summary, the value of reducing the number of evacuating vehicles depends on various factors. For instance, if there is a high level of background traffic (e.g., Ross Valley daytime scenario) or if there are only a few escape routes (e.g., Woodacre Bowl) reducing the number of evacuating vehicles may have little effect.

Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 49 (a) #vehicles that reach 2 miles away from the fire area (daytime scenario)
(b) #vehicles that reach 2 miles away from the fire area (nighttime scenario)
(c) #persons within 0.5 mile of fire (daytime
scenario)
(d) #persons within 0.5 mile of fire (nighttime
scenario)
(e) Cumulative exposure (hours) (daytime scenario) (f) Cumulative exposure (hours) (nighttime scenario) Figure 29. Vehicle Reduction Strategy Performance Measures for Novato Neighborhood Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 50 Table 8. Average Evacuation Time and Exposure Time for Vehicle Reduction Strategy, All Case Studies Indicators
Average
travel
time [h]
%change
from
baseline
Average
exposure
time [h]
%change
from
baseline
Average
travel time
[h]
%change
from
baseline
Average
exposure
time [h]
%change
from
baseline
Ross Valley Daytime case Nighttime case
Baseline 1.250 - 0.120 - 1.230 0 0.223 -
2 persons/veh 1.005 19.6% 0.071 40.8% 0.532 56.7% 0.087 61.0% 3 persons/veh 0.906 27.5% 0.047 60.8% 0.391 68.2% 0.086 61.4% 4 persons/veh 0.790 36.8% 0.048 60.0% 0.301 75.5% 0.085 61.9% 5 persons/veh 0.651 47.9% 0.047 60.8% 0.254 79.3% 0.085 61.9% Woodacre Bowl Daytime case Nighttime case
Baseline 0.577 - 0.074 - 0.389 - 0.054 -
2 persons/veh 0.300 48.0% 0.054 27.0% 0.206 47.0% 0.038 29.6% 3 persons/veh 0.260 54.9% 0.049 33.8% 0.196 49.6% 0.036 33.3% 4 persons/veh 0.246 57.4% 0.049 33.8% 0.197 49.4% 0.036 33.3% 5 persons/veh 0.238 58.8% 0.047 36.5% 0.195 49.9% 0.036 33.3% Tamalpais
Valley Daytime case Nighttime case
Baseline 0.618 - 0.045 - 0.731 - 0.117 -
2 persons/veh 0.356 42.4% 0.045 0.0% 0.365 50.1% 0.099 15.4% 3 persons/veh 0.356 42.4% 0.045 0.0% 0.263 64.0% 0.100 14.5% 4 persons/veh 0.308 50.2% 0.045 0.0% 0.230 68.5% 0.100 14.5% 5 persons/veh 0.252 59.2% 0.045 0.0% 0.222 69.6% 0.100 14.5% Novato
Neighborhood Daytime case Nighttime case
Baseline 0.830 - 0.007 - 1.061 - 0.027 -
2 persons/veh 0.480 42.2% 0.006 14.3% 0.519 51.1% 0.015 44.4% 3 persons/veh 0.359 56.7% 0.007 0.0% 0.464 56.3% 0.015 44.4% 4 persons/veh 0.282 66.0% 0.007 0.0% 0.407 61.6% 0.015 44.4% 5 persons/veh 0.271 67.3% 0.007 0.0% 0.361 66.0% 0.015 44.4% Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 51 Figure 30. Average Evacuation Times for Different Carpooling Scenarios, All Case Studies Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 52 Figure 31. Average Exposure Times For Vehicle Reduction Strategy, All Case Studies Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 53 Phased Evacuation
This section analyzes the effects of a phased evacuation strategy. First, the four study areas were divided into smaller traffic analysis zones as designated by the Metropolitan Transportation Commission. These are seen in Error! Reference source not found.-Error! Reference source not found.. The zones in red (high-risk) are the closest zones to the modeled fire ignition point; therefore, residents in these zones would be the first to leave (Phase 1). Those in orange (middle-risk) and pink (low-risk) zones would be the second and third groups to leave (Phase 2 and Phase 3). Three scenarios with different phased intervals, described above in the “Scenario Development” section, were tested and the designated departure times for each Phase are shown in Table 9. In the daytime scenario evacuees from the designated zones depart at either 15, 30, or 45-minute intervals, while in the nighttime scenario, the intervals are 20, 40, or 60 minutes apart.

The results for each of the four locations are presented in Figures 36-39 which show the number of evacuees that reach a point at least two miles away from the fire area over time, number of evacuees within a half mile of the fire area over time, and cumulative fire exposure over time under each phased evacuation strategies for the four case studies. Table 10 and Figures 40-41 present the average travel time and average exposure time or each of the different alternatives.

Figure 32. Evacuation Zones for Ross Valley
Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 54 Figure 33. Evacuation Zones for Woodacre Bowl
Figure 34. Evacuation Zones for Tamalpais Valley
Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 55 Figure 35. Evacuation Zones of Novato Neighborhood
Table 9. Departure Times for Phased Evacuation Strategy, All Case Studies Interval Phase 1
(zones)
Phase 1
(departure
time)
Phase 2
(zones)
Phase 2
(departure
time)
Phase 3
(zones)
Phase 3
(departure
time)
Day
time
15 mins
Fairfax
(1,3)
Woodacre
(1,2)
Tamalpais
(1)
Novato
(6,10)
9:20±10
mins
Fairfax
(2,4,5,6)
Woodacre
(4,5)
Tamalpais
(2,3,4)
Novato
(1,5,9,12)
9:35±10 mins
Fairfax
(7,8,9,10)
Woodacre
(3,6)
Tamalpais
(5,6,7)
Novato
(2,3,4,7,8,1
1)
9:50±10
mins
30 mins 9:20±10
mins
9:50±10 mins 10:20±10
mins
45 mins 9:20±10
mins
10:05±10
mins
10:50±10
mins
Night
time
20 mins 1:40±20
mins
2:00±20 mins 2:20±20
mins
40 mins 1:40±20
mins
2:20±20 mins 3:00±20
mins
60 mins 1:40±20
mins
2:40±20 mins 3:40±20
mins
Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 56 For the Ross Valley case, Figure 36 shows that in the daytime scenario when the interval between phases is 15 minutes it takes two hours to complete the evacuation. When the interval is increased to 30 minutes the evacuation takes three and a half hours and when it is increased to 45 minutes the evacuation is completed in four hours. In the nighttime scenario, if the phased intervals are 20, 40 and 60 minutes, the evacuation times last 3, 3, and 4 hours, respectively. Notably, longer lasting evacuation times do not necessarily represent high evacuation efficiency. From the perspective of people within 0.5 mile of the fire area and cumulative exposure time, the baseline leads to the lowest exposure (around 85 hours) and quick evacuation, and longer interval time leads to worse evacuation performance at the daytime scenario. In the nighttime scenario, the best option is the 20-minute interval (corresponding cumulative exposure time is 600 hours) and the worst is the interval of 60 minutes, of which the exposure time reaches over 700 hours.

Table 10 and Figures 40-41 show that the average evacuation time decreases as the phasing interval increases from one and a quarter hour when there is no phased interval to one hour when the interval is 45 minutes, a reduction of 23 percent. However, the longer interval time forces some people to stay longer in the fire danger area. When the interval time is 45 minutes, the average exposure time increases by 17 percent, compared to the baseline. However, when the interval time is 15 minutes, both average travel time and exposure time are reduced. Hence, the optimal interval for Ross Valley is 15 minutes. For the nighttime evacuation, a longer interval time leads to lower evacuation time, but higher exposure time. When the interval time is 20 minutes, the average evacuation time is 40 percent of the baseline and 60 percent of the time with a 40-minute interval.

In addition, the average exposure time with a 40-minute interval is 27 percent less, compared with a reduction of 38 percent with a 20-minute interval and 6 percent with a 60-minute interval. Therefore, a 20-minute interval is the best option for the Ross Valley nighttime scenario.

Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 57 (a) #vehicles that reach 2 miles away from the fire area (daytime scenario)
(b) #vehicles that reach 2 miles away from the fire area (nighttime scenario)
(c) #persons within 0.5 mile of fire (daytime
scenario)
(d) #persons within 0.5 mile of fire (nighttime
scenario)
(e) Cumulative exposure (hours) (daytime scenario) (f) Cumulative exposure (hours) (nighttime scenario) Figure 36. Phased Evacuation Performance Measures for Ross Valley) Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 58 For the Woodacre Bowl case, Figure 37 shows that in the daytime scenario if the phased intervals are 15, 30, and 45 minutes, respectively, the evacuation takes one, one and a half, and two hours to complete. In the nighttime scenario if the phased intervals are 20, 40, and 60 minutes, respectively the evacuations last one and a half, two and a half, and three hours. Notably, longer evacuation time does not necessarily represent low evacuation efficiency.

An interval of 30 minutes leads to the lowest exposure time (around 58 hours) and quick evacuation, while the baseline performs the worst in the daytime scenario (up to 100 hours of cumulative exposure time). In the nighttime scenario, all evacuation intervals perform similarly, with exposure time of around 80 hours, while for the baseline, the exposure time reaches 120 hours.

Table 10 and Figures 40-41 show that, similar to the Ross Valley case, the average daytime and nighttime evacuation times decrease as the phased evacuation interval time increases. However, the average exposure times of both scenarios first decrease and reach their lowest value with a 30/40-minute interval. However, as the interval time continues to increase, the average exposure times increase, and the speed of the evacuation becomes slower than the speed of the fire spread. Hence, the ideal interval time for the daytime evacuation is 30 minutes and 40 minutes for the nighttime evacuation.

Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 59 (a) #vehicles that reach 2 miles away from the fire area (daytime scenario)
(b) #vehicles that reach 2 miles away from the fire area (nighttime scenario)
(c) #persons within 0.5 mile of fire (daytime
scenario)
(d) #persons within 0.5 mile of fire (nighttime
scenario)
(e) Cumulative exposure (hours) (daytime scenario) (f) Cumulative exposure (hours) (nighttime scenario) Figure 37. Phased Evacuation Performance Measures for Woodacre Bowl Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 60 For the Tamalpais Valley case, Figure 38 shows that in the daytime scenario when the interval between phases is 15 minutes it takes one and a half hours to complete the evacuation. When the interval is increased to 30 minutes the evacuation takes two hours and when it is increased to 45 minutes the evacuation is completed in two and a half hours. In the nighttime scenario, if the phased intervals are 20, 40 and 60 minutes, respectively, the evacuations last two and three-quarter hours two and three-quarter hours and three and a half. Notably, longer evacuation times does not necessarily represent low evacuation efficiency. For the daytime scenario, the baseline has the lowest fire exposure time (around 85 hours) and quick evacuation, and longer interval times reduce evacuation performance. For the nighttime scenario, the all-phasing intervals perform similarly, with exposure times around 600 hours, though the 60-minute interval exposure time reaches over 700 hours.

Table 10 and Figures 40-41 show that a larger interval time results in higher evacuation efficiency in terms of the average evacuation time for both the daytime evacuation and nighttime scenarios. However, average fire exposure time, even for the 15/20-minute interval is no better than the no-phase scenario. In the daytime evacuation, a 15-minute interval reduces the average evacuation time by 34 percent, while increasing the average exposure time by 16 percent. If the focus is on reducing fire exposure time as much as possible, then for the daytime evacuation, the best strategy is to let all evacuees leave at the same time. For the nighttime evacuation, applying a 20-minute interval reduces the average evacuation time markedly and reduces the exposure time somewhat. Therefore, this is the optimal strategy for the nighttime case.

Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 61 (a) #vehicles that reach 2 miles away from the fire area (daytime scenario)
(b) #vehicles that reach 2 miles away from the fire area (nighttime scenario)
(c) #persons within 0.5 mile of fire (daytime
scenario)
(d) #persons within 0.5 mile of fire (nighttime
scenario)
(e) Cumulative exposure (hours) (daytime scenario) (f) Cumulative exposure (hours) (nighttime scenario) Figure 38. Phased Evacuation Performance Measures for Tamalpais Valley Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 62 For the Novato Neighborhood case, Figure 39 shows that in the daytime scenario when the interval between phases is 15, 30 and 45 minutes, respectively the evacuation is completed in 1.75, two, and 2.5 hours. For the nighttime scenario if the phased intervals are 20, 40, and 60 minutes, respectively, the evacuations last two and a half, two and a half and four hours.

Longer evacuation times do not necessarily represent low evacuation efficiency. An interval of 30 minutes for the daytime scenario leads to the lowest cumulative fire exposure time (around 45 hours) and quick evacuation, while the baseline leads to the worst evacuation. For the nighttime scenario, intervals of 20, 40, and 60 minutes perform similarly, with exposure times around 250 hours while the baseline exposure time reaches around 500 hours. Table 10 and Figures 40-41 show that, as for the other three cases, average evacuation times decrease as the phasing interval is increased. Average exposure times are mostly unchanged across these three different intervals. Additionally, the performance of the 30/40-minute interval is close to that of a 45/60-minute interval. Therefore, 30/45-minute intervals for daytime evacuation and 40/60-minute intervals for nighttime evacuation are the best options according to the model.

In summary, as the phasing interval is increased, the evacuation takes longer, at least up to a point, because the traffic congestion almost disappears. When implementing a phased evacuation strategy, it is important not only to focus on travel time but also the duration of fire exposure; longer waiting times in high-risk areas can pose a significant threat to residents.

Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 63 (a) #vehicles that reach 2 miles away from the fire area (daytime scenario)
(b) #vehicles that reach 2 miles away from the fire area (nighttime scenario)
(c) #persons within 0.5 mile of fire (daytime scenario) (d) #persons within 0.5 mile of fire (nighttime scenario) (e) Cumulative exposure (hours) (daytime
scenario)
(f) Cumulative exposure (hours) (nighttime scenario)
Figure 39. Phased Evacuation Strategy Performance Measures for Novato Neighborhood Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 64 Table 10. Average Evacuation Time and Exposure Time for Phased Evacuation Strategy, All Case Studies Indicators
Average
travel
time [h]
%change
from
baseline
Average
exposure
time [h]
%change
from
baseline
Average
travel time
[h]
%change
from
baseline
Average
exposure
time [h]
%change
from
baseline
Ross Valley Daytime case Nighttime case
Baseline 1.250 - 0.120 - 1.230 0 0.223 -
15/20 mins 1.081 -13.5% 0.088 -26.7% 0.883 -28.2% 0.138 -38.1% 30/40 mins 1.079 -13.7% 0.110 -8.3% 0.505 -58.9% 0.166 -25.6% 45/60 mins 0.963 -23.0% 0.140 +16.7% 0.412 -66.5% 0.210 -5.8% Woodacre Bowl Daytime case Nighttime case
Baseline 0.577 - 0.074 - 0.389 - 0.054 -
15/20 mins 0.328 -43.2% 0.054 -27.0% 0.203 -47.8% 0.037 -31.5% 30/40 mins 0.236 -59.1% 0.046 -37.8% 0.197 -49.4% 0.035 -35.2% 45/60 mins 0.236 -59.1% 0.050 -32.4% 0.194 -50.1% 0.036 -33.3% Tamalpais
Valley Daytime case Nighttime case
Baseline 0.618 - 0.045 - 0.731 - 0.117 -
15/20 mins 0.408 -34.0% 0.052 +15.6% 0.463 -36.7% 0.116 -0.9% 30/40 mins 0.308 -50.2% 0.059 +31.1% 0.301 -58.9% 0.122 +4.3% 45/60 mins 0.296 -52.1% 0.064 +42.2% 0.282 -61.4% 0.133 +13.7% Novato
Neighborhood Daytime case Nighttime case
Baseline 0.830 - 0.007 - 1.061 - 0.027 -
15/20 mins 0.346 -58.3% 0.007 0.0% 0.502 -52.7% 0.015 -44.4% 30/40 mins 0.220 -73.5% 0.006 -14.3% 0.322 -69.7% 0.015 -44.4% 45/60 mins 0.220 -73.5% 0.007 0.0% 0.311 -70.7% 0.015 -44.4% Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 65 Figure 40. Average Travel Times for Various Departure Intervals for Phased Evacuation Strategy, All Case Studies
Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 66 Figure 41. Average Exposure Times for Different Departure Intervals for Phased Evacuation Strategy, All Case Studies
Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 67 On-street Parking Prohibition
This section analyzes the impact of street-parking on evacuation efficiency. Figures 42-45 show the number of evacuees that reach a point two miles away from the fire area over time, number of evacuees within a half mile of the fire area over time, and cumulative exposure over time with and without street parking for all four cases.

Table 11 and Figures 46-47 present the average evacuation time and average exposure time for each alternative parking prohibition strategy.

For the Ross Valley case, Figure 42 shows that if street parking is allowed, the evacuation lasts over four hours for the daytime scenario, compared with two hours if street parking is prohibited. However, for the nighttime scenario, there is no significant difference with evacuation time lasting around 3.5 hours.

Street parking leads to slower evacuations and increases the total cumulative exposure time in the daytime scenario to over 400 hours, which is double compared to the situation without street parking. For the nighttime scenario, the total exposure time when street parking is permitted is only 100 hours more than when it is prohibited.

Table 11 and Figures 46-47 show that, if street parking is permitted during the daytime, the average evacuation time increase significantly by up to 88 percent and fire exposure time to over 78 percent. However, street parking does not affect nighttime evacuations as much, as there is little of background traffic in the way so even though permitting street parking reduces road capacity, vehicles can still evacuate smoothly.

Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 68 (a) #vehicles that reach 2 miles away from the fire area (daytime scenario)
(b) #vehicles that reach 2 miles away from the fire area (nighttime scenario)
(c) #persons within 0.5 mile of fire (daytime scenario) (d) #persons within 0.5 mile of fire (nighttime scenario) (e) Cumulative exposure (hours) (daytime
scenario)
(f) Cumulative exposure (hours) (nighttime scenario)
Figure 42. Street Parking Prohibition Strategy Performance Measures for Ross Valley Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 69 For the Woodacre Bowl case, Figure 43 shows that if street parking is allowed, the evacuation lasts for over one and a half hours for the daytime scenario, which is very close to the evacuation time when street parking is prohibited. For the nighttime scenario, street parking extends the evacuation time from one and a quarter hour to one and a half hours. In the daytime scenario, the cumulative exposure times are similar at around 100 hours; however, for the nighttime scenario, street parking increases the total exposure time by 20 hours to 140 hours and strands persons in the vicinity of the fire area for a longer duration. Table 11 and Figures 46-47 show the opposite result compared to the Ross Valley case. Street parking has a very small effect on the daytime evacuation, but a huge impact on the nighttime evacuation. During the daytime, there is little background traffic around that area and the number of evacuees is relatively small, so the evacuees can evacuate quickly.

However, during the nighttime, the number of evacuees is nearly double the number of evacuees during the daytime. There are only two exits to leave this area and if evacuees want to leave, they need first to enter the fire danger area and then drive through these two exists. With the reduced road capacity caused by street parking, the fire danger area becomes congested which leads to longer exposure and evacuation times.

Therefore, officials should consider notifying residents to move their vehicles off the streets in these situations.

Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 70 (a) #vehicles that leave 2 miles away from the fire area (daytime scenario)
(b) #vehicles that leave 2 miles away from the fire area (nighttime scenario)
(c) #persons within 0.5 mile of fire (daytime scenario) (d) #persons within 0.5 mile of fire (nighttime scenario) (e) Cumulative exposure (hour) (daytime scenario) (f) Cumulative exposure (hour) (nighttime scenario) Figure 43. Street Parking Prohibition Strategy Performance Measures for Woodacre Bowl Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 71 For the Tamalpais Valley case, Figure 44 shows that if street parking is allowed, the evacuation lasts over five hours in the daytime scenario, four hours longer than without street parking. For the nighttime scenario, street parking only extends the evacuation time by one hour to around two and a half hours. Street parking extends fire exposure times in both scenarios by 40 hours and 400 hours, respectively. Table 11 and Figures 46-47 show that street parking has a significant influence on evacuation efficiency in both the daytime scenario and nighttime scenario, where the average evacuation times increase by 19 and 12 percent, respectively, and the average exposure times increase by 52 and 58 percent respectively. Based on the simulation results, officials should consider prohibiting on-street parking whenever fire danger in Tamalpais Valley is high.

Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 72 (a) #vehicles that reach 2 miles away from the fire area (daytime scenario)
(b) #vehicles that reach 2 miles away from the fire area (nighttime scenario)
(c) #persons within 0.5 mile of fire (daytime scenario) (d) #persons within 0.5 mile of fire (nighttime scenario) (e) Cumulative exposure (hours) (daytime
scenario)
(f) Cumulative exposure (hours) (nighttime scenario)
Figure 44. Street Parking Prohibition Performance Measures for Tamalpais Valley Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 73 For the Novato Neighborhood case, Figure 45 shows that if street parking is allowed, the evacuation lasts over two and a half hours for the daytime scenario, which is very close to the evacuation time when street parking is prohibited. For the nighttime scenario, street parking extends the evacuation time from four hours to four and a half hours. Street parking almost has no impact on the cumulative fire exposure time and number of persons within a half mile of the fire area in the daytime scenario. In the nighttime scenario, street parking hinders the evacuation process, causing the exposure time to increase to 700 hours, which is 250 hours more compared to the situation without street parking. Table 11 and Figures 46-47 show that street parking has a tiny effect on daytime evacuation, but a significant influence on nighttime evacuation. During the day, the number of evacuees is small, and there are many exits. Therefore, the reduced road capacity for residential roads does not influence the evacuation much. However, there are 17,519 evacuees during the night. Even though there is almost no background traffic, the dual effect of having so many evacuees and the fewer number of accessible lanes creates delays. These findings show how banning on-street parking at night when fires occur in the Novato Neighborhood could greatly increase evacuation efficiency.

In evaluating banning on-street parking, we assumed that, if the road capacity is reduced by 25 percent, the average evacuation time would increase by around 25 percent. Yet, the results of many scenarios did not reflect this pattern. Instead, the evacuation times only increased about 10 percent in some scenarios (e.g., Ross Valley nighttime scenario and Woodacre Bowl daytime scenario), but rose by 50 percent in others (e.g., Ross Valley daytime scenario, Novato Neighborhood nighttime scenario). The impact depended on the different number of background vehicles and evacuee vehicles. Thus, the merits of this strategy depend on the characteristics of the local areas, including the ratio of number of evacuees to volume of background traffic, the spatial distribution of evacuees, and the number of exits.

Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 74 (a) #vehicles that reach 2 miles away from the fire area (daytime scenario)
(b) #vehicles that reach 2 miles away from the fire area (nighttime scenario)
(c) #persons within 0.5 mile of fire (daytime scenario) (d) #persons within 0.5 mile of fire (nighttime scenario) (e) Cumulative exposure (hours) (daytime
scenario)
(f) Cumulative exposure (hours) (nighttime scenario)
Figure 45. Street Parking Prohibition Strategy Performance Measures for Novato Neighborhood () Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 75 Table 11. Average Evacuation and Exposure Times for Street Parking Prohibition Strategy for All Case Studies
Indicators Average
travel time
[h]
%change
from
baseline
Average
exposure
time [h]
%change
from
baseline
Average
travel
time [h]
%change
from
baseline
Average
exposure
time [h]
%change
from
baseline
Ross Valley Daytime case Nighttime case
Without street
parking (baseline)

1.250 - 0.120 - 1.230 - 0.223 -
With street
parking

2.353 +88.2% 0.214 +78.3% 1.341 +9.0% 0.243 +9.0%
Woodacre Bowl Daytime case Nighttime case
Without street
parking (baseline)

0.577 - 0.074 - 0.389 - 0.054 -
With street
parking

0.632 +9.5% 0.076 +2.7% 0.480 +23.4% 0.063 +16.7%
Tamalpais Valley Daytime case Nighttime case
Without street
parking (baseline)

0.618 - 0.045 - 0.731 - 0.117 -
With street
parking

0.738 +19.4% 0.070 +52.2% 0.817 +11.8% 0.184 +57.3%
Novato
Neighborhood
Daytime case Nighttime case
Without street
parking (baseline)

0.830 - 0.007 - 1.061 - 0.027 -
With street
parking

0.896 +8.0% 0.007 0.0% 1.327 +25.1% 0.039 44.4%
Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 76 Figure 46. Average Evacuation Times With and Without Street Parking Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 77 Figure 47. Average Fire Exposure Times With and Without Street Parking Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 78 Recommended Strategies
The previous section analyzed different evacuation strategies in terms of two summary statistics: (i) average time for evacuees to reach two miles away from the fire area; and (ii) average exposure time within a half miles of fire area. From the results shown in Tables 8, 10-11 and Figures 30-31, 40-41, and 46-47, different strategies exhibit varying performance in different areas and times. This is due to differences in road networks, fire ignition points and progression, the number of evacuees, and the normal traffic around the areas. Table 12 lists recommended strategies for each case.

Sharing a vehicle among multiple persons is a good strategy among these different cases, as sharing vehicles can greatly reduce traffic congestion and speed up the evacuation. However, it is difficult and time-consuming to coordinate having persons from different households share one vehicle in emergency situations. On the contrary, it is quite easy to encourage persons not to use all their cars, but to share one or two vehicles among family members. We recommend that at least two persons should share a car during fire evacuations.

We recommend cities consider adopting a phased evacuation strategy. However, longer departure intervals do not necessarily result in a more efficient evacuation. This is because over time fires can spread and approach zones that were initially far from the point of origin. Therefore, it is not advisable to make persons in these zones wait too long time before issuing an evacuation order.

The effect of street parking on evacuation varies with the circumstances. In some cases, such as the nighttime scenario for Ross Valley and the daytime scenario for Woodacre Bowl, street parking had little impact on the evacuation, while in the other cases, prohibiting street parking can result in a significant increase in evacuation efficiency. The difference results from varying numbers of evacuees and exit routes. When the number of evacuees is relatively small compared with the number of background vehicles and the number of exits is large, the effect of street parking is small. Yet, with are more evacuees and fewer exits, the impact of street parking is high. Overall, prohibiting street parking on fire days can effectively reduce evacuation times. This strategy would be advantageous in all cases, particularly Tamalpais Valley, and during the daytime for Ross Valley, and at night for Woodacre Bowl, and the Novato Neighborhood. Prohibiting street parking in these cases could significantly improve evacuation performance.

Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 79 Table 12. Recommended Strategies
Case Daytime Scenario Nighttime Scenario
Ross Valley i. ≥3 persons sharing one vehicle
ii. 15-min departure interval
iii. No street parking
i. ≥2 persons sharing one vehicle
ii. 40-min departure interval
Woodacre Bowl i. ≥3 persons sharing one vehicle
ii. 30-min departure interval
i. ≥2 persons sharing one vehicle
ii. 40-min departure interval
iii. No-street parking
Tamalpais Valley i. ≥2 persons sharing one vehicle
ii. No phased evacuation
iii. No-street parking
i. ≥2 persons sharing one vehicle
ii. 20-min departure interval
iii. No-street parking
Novato Neighborhood i. ≥3 persons sharing one vehicle
ii. 30/45-min departure interval
i. ≥3 persons sharing one vehicle
ii. 40/60-min departure interval
iii. No-street parking
Agenda Item #11.1.

Testing Wildfire Evacuation Strategies and Coordination Plans for Wildland Urban Interface (WUI) Communities in California 80 Policy Recommendations
The following policy recommendations are based on our analysis of the three evacuation strategies.1 Vehicle Reduction
Based on the analysis presented in the previous sections, reducing the number of evacuating vehicles is a highly recommended strategy for Marin County. In Marin County, the average household size in 2019 was 2.45 and the number of vehicles per household was 1.9 (US Census Bureau, 2021). Due to the high level of vehicle ownership, there may be less incentive to carpool in emergency situations. The relevant benefits of carpooling have already been advocated on official websites (Firesafe Marin, 2019, 2020).

This strategy was also analyzed in Zhao and Wong (2021), and the relevant policy recommendations discussed there remain valid based on results from this study:
• Recommend residents take as few vehicles as possible (just enough to transport people and key belongings) through an educational and informational campaign. • Suggest residents pre-pack their vehicle(s) during high fire danger periods so that vehicle space is used efficiently.
• Develop an equitable insurance framework for protecting residents’ vehicles in high-risk fire areas. • Develop plans for parking areas outside of potential evacuation zones for residents to take additional vehicles during high fire danger weather conditions as part of their emergency preparedness planning).

Zhao and Wong (2021)
In addition, the following policies are also recommended:
• Establish a communication system to provide real-time updates on car-pooling opportunities, match drivers with potential passengers, and disseminate information about carpooling resources and schedules.
• Continuously monitor and evaluate the effectiveness of vehicle reduction/car-pooling policies after a wildfire. Collect feedback from users, measure the impact on transportation patterns, and make necessary adjustments to improve the effectiveness and operational feasibility of this strategy. 1 The analysis framework and some content of this section can be referred to Table 4 in Zhao and Wong (2021). Direct quotes are used for readability. References to the original content are given whenever applicable.

Agenda Item #11.1.