TRT Responds to SFPUC Misinformation

The State Water Resources Control Board is in the process of updating the Bay Delta Water Quality Control Plan to help restore the estuary and rivers that feed it, including the Tuolumne.

Comments submitted by the City and County of San Francisco, the Bay Area Water Supply and Conservation Agency (BAWSCA), many of BAWSCA’s 26 member agencies, and several other entities exhibited a number of misconceptions about potential impacts of the Plan. Many of the figures cited were based on misleading information provided by the San Francisco Public Utilities Commission (SFPUC).

For example, eight members of the Bay Area Legislative Caucus wrote, “If approved, the SWRCB’s current plan would substantially cut water supplies to 2.6 million Bay Area residents in San Francisco, Silicon Valley and the East Bay by up to 50% at the first sign of any future drought.”

Comment letters from 16 of BAWSCA’s member agencies were virtually the same. The agencies were provided with a form letter and encouraged to fill in a few blanks. Among other things, these letters stated, “As a wholesale customer of SFPUC … water supply available to (city or agency) under the SED proposal could be reduced more than 50% under drought conditions for multiple consecutive years.”

Objective modeling demonstrates the SFPUC could manage a reoccurrence of the drought of record, even with the 40% unimpaired flow requirement in place.

TRT has modeled the impact of a reoccurrence of the 1987-92 drought under the SFPUC’s current baseline demand – 223 million gallons per day (MGD) or 250,000 acre feet/year (AF/Y) – and 40% unimpaired flow between February and June. We found that the SFPUC would not run out of water under such a scenario. We based our model on a modest average of 10% rationing during the six-year period. We assumed there would be no rationing in the first two years of the drought, 10% rationing in years three and four, and 20% rationing in years five and six. For comparison, water demand in FY 2015/16 was 175 MGD – 21.5% below the 223 MGD baseline. Demand in FY 2016/17 was 180 MGD – 20% below the baseline.

The SFPUC has not challenged our model, and in fact, their own modeling demonstrates they would not run out of water if the drought of record were to reoccur and the SED were in place. However, they base their planning on a “Design Drought” (see below).

During the height of the recent 4-year drought (2012-2015), the SFPUC had enough water in storage to last more than three years. Had the Bay Delta Plan’s 40% unimpaired flow between February and June been in place during the drought, the SFPUC still would have had at least 2-years-worth of water in storage at any given point.

Following the normal water year in 2016, the SFPUC’s Tuolumne River storage rebounded to 85% of capacity by December 11. By early January of 2017, it was clear that all the reservoirs on the Tuolumne would easily fill, so water had to be released from Don Pedro Reservoir at the maximum amount allowed by flood control rules for months. Had the SED been in place during the drought, all of the Tuolumne’s reservoirs would still have filled in January, and any previous storage deficit would have been erased.

SFPUC figures are based on an arbitrary 8.5-year “Design Drought.”

While the SFPUC has not challenged our modeling, they assert that their much more extreme rationing scenario is based on an 8.5-year “Design Drought,” which includes the 1987-92 drought of record, followed immediately by the driest 2-year period on record – 1976/77. They assume every year is either the beginning of, or middle of, this Design Drought, and thus factor in much higher levels of rationing than would have been required at any time in the past.

The SFPUC does not explain its Design Drought when sharing information with others, leaving the impression that impacts would be much greater than should be reasonably expected. Furthermore, the scenario they share is based on Plan Bay Area 2040 demand projections (265 MGD) vs. the current baseline (223 MGD) or internal SFPUC 2040 demand projections (250 MGD).

The SFPUC model assumes rationing would begin in the first year of drought, and would be 39% for the first three years (under 223 MGD demand) and 49% for the remaining years. This policy would leave 576,000 acre-feet remaining in storage at the end of the six-year drought of record if it were to reoccur. This is enough water to last more than two years at pre-drought demand, and almost three years at FY 2015/16 demand. All of the projected job losses and socioeconomic impacts associated with excessive rationing would have been unnecessary, making the Design Drought a very expensive insurance policy.

The SFPUC has projected jobs and economic losses from such a conservative rationing scenario to be 445,905 annual full-time equivalents and more than $116 billion. Again, these impacts would have been unnecessary. These socioeconomic impacts are based on a 2017 Brattle Group report commissioned by the SFPUC.

The 2017 Brattle Group report is flawed.

Along with its SED comments, the SFPUC submitted a document entitled, Bay Area Socioeconomic Impacts Resulting from Instream Flow Requirements for the Tuolumne River, prepared by The Brattle Group. This report states, “The method used to estimate these impacts is described in the report Socioeconomic Impacts of Water Shortages within the Hetch Hetchy Regional Water System Service Area, prepared by The Brattle Group in 2014.”

Shortly after the 2014 report was issued, a coalition of NGOs submitted extensive comments that identified a number of flaws in the Brattle Group analysis.

The recent drought proved that prior Brattle Group studies were seriously flawed (see TRT presentation). For example, their 2014 report projected that 30% rationing would result in the loss of 25,000 jobs and $7 billion. However, between 2006 and 2016, water demand in the SFPUC service territory decreased by 30%, with no job losses and no negative economic impacts. In fact, 125,000 jobs were added in San Francisco alone between 2010 and 2015, and the regional economy improved to its strongest position ever.

SFPUC drought planning harms the Tuolumne River ecosystem unnecessarily.

During the recent drought, releases into the lower Tuolumne were critically low, and fish and wildlife suffered. Had the SED been in place, river conditions would have been improved, and SFPUC storage would currently be the same as it is today. This graph shows that under the current flow regime, the SFPUC needed to capture 373,000 acre-feet coming into Water Year 2017 to fill its reservoirs on the Tuolumne. Actual water available to San Francisco was 3.1 million acre-feet – enough water to fill all of the SFPUC’s reservoirs, including those in the Bay Area, more than twice. In other words, during Water Year 2017, the SFPUC had the right to capture enough water from the Tuolumne (if storage were available) to last more than 12 years.

Click here for a series of graphs that compare unimpaired flow, actual flow, and 40% unimpaired flow between February and June between 2012 and 2017. The difference between 40% unimpaired and actual flow is additional water that would have benefited the Tuolumne River ecosystem had the SED been in place. Rather than releasing this water during the drought, it all (and much more) got dumped in 2017, providing much less benefit to the river ecosystem than had it been apportioned over the dry years of the drought.

TRT has attempted to correct the record.

The Tuolumne River Trust and other NGOs have attempted to convince the SFPUC to correct the record, with limited success. The SFPUC issued a brief, to which we responded with our critique, and finally the SFPUC responded to our critique. We believe these documents make it clear that information distributed by the SFPUC has been extremely misleading.

Conclusion

The Delta Reform Act of 2009 established a State policy aimed at achieving the co-equal goals of ensuring water supply reliability and restoring the Bay-Delta ecosystem. Unfortunately, many water agencies have chosen to ignore the second of the goals. This is unfortunate, because under a reasonable scenario, both objectives could be achieved.