OUR MISSION: To provide our customers with high-quality, efficient and reliable water, power and sewer services in a manner that values environmental and community interests and sustains the resources entrusted to our care.
525 Golden Gate Avenue, 13th Floor
San Francisco, CA 94102
T 415.554.3155
F 415.554.3161
TTY 415.554.3488
San Francisco Public Utilities Commission
Correspondence Log
September 25, 2023 through September 25, 2023
Date Received From Subject
September 25, 2023 Dave Warner Remarkable Similarities to San Diego
San Francisco Public Utilities Commission
Correspondence Log
October 2, 2023 through October 2, 2023
Date Received From Subject
October 2, 2023 Dave Warner Comparing Design Drought Risk to Sewage Overflow Risk
September 25, 2023
Re : Remarkable Similari(g4)5es to San Diego
Dear
Commissioners, General Manager Herrera and Assistant General Manager Ritchie, This chart likely looks familiar to you: Demand declining since 2000 while projec)gU6-ons show demand increasing:
But it’s not ours. The one above is the San Diego County Water Authority’s (SDCWA). This is ours below:
216
195
100
150
200
250
300
350
2000 2005 2010 2015 2020 2025 2030 2035 2040 2045
MGD
SFPUC Combined Retail and Wholesale Projections
Comparison to Actual Demand*
Actual 2000 UWMP 2005 UWMP
2010 UWMP 2015 UWMP 2020 UWMP
*Comparison is between the UWMP projections(dashed lines) and the finance department sales projections (dotted lines) where 6 mgd of projected water loss (non-revenue) has been added to the finance department figures. 2015 finance department projections were visually estimated from the charts in General Manager Herrera's letter to commissioners dated July 5, 2022.
~190
What’s more remarkable is the impact on rates. From 2007 to 2024, both agencies experienced their water rates more than tripling:
San Diego’s rates went up in part due to investments in alterna)gU6-ve water supplies including their Carlsbad desalina)gU6-on plant combined with an unexpected 40% drop in demand. SDCWA has been cri)gU6-cized for inves)gU6-ng in AWS given the large drop in demand. One analysis stated that without the Carlsbad desal plant, wholesale rates would have been $200 less even at the lower level of demand.
The SFPUC’s rates went up in part due to WSIP investments along with a 27% drop in demand. Of course, the WSIP investments were needed to maintain exis)gU6-ng supplies rather than inves)gU6-ng in new supplies. The point is that when demand goes down and investments increase, rates necessarily go up and now both us and San Diego have remarkably high wholesale rates (although we s)gU6-ll top theirs).
A learning from San Diego’s experience is understanding SDCWA’s fixed and variable revenue and cost analysis:
The table shows that 85% of SDCWA’s opera)gU6-ng costs are fixed while only 28% of opera)gU6-ng revenues are fixed. When demand declines, costs don’t decline by as much, hence driving the need to increase rates. It would be interes)gU6-ng to see the SFPUC’s split between fixed and variable costs and revenues to help understand the rate risk of declining revenues in a period of increasing costs.
The point in comparing ourselves to San Diego is that we’ve already seen our rates jump because of WSIP and declining demand/revenues. To do it yet again by inves)gU6-ng in AWS would put our rates through the roof, so to say (if they are not already through the roof), par)gU6-cularly if each drop of AWS is not fully u)gU6-lized. Higher rates don’t help economic ac)gU6-vity.
Sincerely,
Dave Warner
cc: BAWSCA Board of Directors, Nicole Sandkulla, BAWSCA CEO, Adrain Covert, Bay Area Council $-
$500
$1,000
$1,500
$2,000
$2,500
2007 2024
BAWSCA & San Diego Wholesale Rate
Comparison
(per acre foot)
BAWSCA San Diego
October 2, 2023
Re: Comparing Design Drought Risk to Sewage Overflow Risk
Dear Commissioners, General Manager Herrera and Assistant General Manager Ritchie,
My understanding is that San Francisco’s sewer system is currently designed with a recurrence interval for a raw sewage overflow event once every five years and that the goal is to increase the recurrence interval to once every 100 years. In an undated SFPUC memo to file regarding the SFPUC’s Long Term Vulnerability Assessment 1, the design drought was found to have a recurrence interval of 25,293 years. While the memo pointed out that there was a large uncertainty related to extreme drought events, the lower bounds of the 95% confidence interval was still over 12,000 years.
If I were a San Francisco resident, I would gladly trade off reducing the design drought recurrence interval by an order of magnitude in exchange for lengthening the recurrence interval to 100 years or more for a raw sewage overflow. This trade off could easily be achieved by reducing the design drought by a year, thereby reducing the need for alternative water supplies by 20-25 mgd, which frees up funds for sewer system reliability improvements.
In other words, taking a wholistic view of risks across the SFPUC’s services, the design drought stands out as an excessively over-engineered system in the context of competing needs for SFPUC resources.
In the light of competing needs, please ask staff what an appropriate recurrence interval should be for a drought planning model and what such a model would look like.
Best regards,
Dave Warner
1 The title of the undated memo to file is “Issues that arose during the review of working draft final report and how we agreed to resolve them.” The working draft final report referred to is the Long Term Vulnerability Assessment. The recurrence interval of 25,293 years for the design drought can be found in the table under Issue 2.