BAY DELTA SCIENCE CONFERENCE: Checking assertions with data: Untangling factors that constrain water exports from the San Francisco Bay estuary

The Banks Pumping Plant in the South Delta (Photo by DWR)

In recent years, media reports and editorials have suggested that environmental regulations have made Delta outflows increasingly large, project water exports are frequently limited or even halted by regulations to protect endangered species, and that regulations designed to protect endangered species, the Delta Smelt in particular, are the principal restrictions on exports and are thus responsible for most of the water that flows from the Central Valley to San Francisco Bay.  But are these assertions true?  Are Delta smelt really responsible for as much as they get blamed for?

The Bay Institute studied water management practices and constraints that affect the flow of water into and through the San Francisco Bay and the Delta estuary, which is home to six imperiled fish species and water export facilities owned by the state and federal governments that serve millions of people and large expanses of agricultural land.

Using several public data sources, The Bay Institute analyzed the long‐term trend in the net effect on Delta outflow of water diversions in the estuary’s Central Valley watershed, and the frequency and magnitude of specific regulatory and infrastructural constraints on the two water export facilities.

Greg Reis, a scientist with the Bay Institute, as well as an Information & Restoration Specialist at Mono Lake Committee, gave a presentation at the 2018 Bay Delta Science Conference to discuss the results.

The last time people talked about water ‘wasting’ to Mono Lake was decades ago, so its disappointing to hear that on this side of the mountains still in the mainstream media and in Congress and various venues,” he said.  “That clearly is false, because we know rivers need water, estuaries need water, terminal lakes need water, and preferably in a natural pattern and volume and a somewhat natural range of variability.”

Assertion: Environmental regulations have made Delta outflows increasingly large

The first assertion is that the outflow from the Delta to the San Francisco Bay (and ultimately to the ocean) has become increasingly large.  So in looking at unimpaired flow versus actual flow, is Delta outflow increasing or decreasing?

Mr. Reis presented a chart, with the blue line showing unimpaired flow and the red line depicting actual Delta outflow; the cubic feet per second of outflow to the estuary is on the Y axis.

Unimpaired flow is defined as ‘the flow that would occur if all runoff from the watershed remained in the river, without storage in reservoirs or diversions, such as irrigation, power generation, or water supply.  “Unimpaired flow is an index of runoff conditions in the watershed,” Mr. Reis said.  “It is different than natural flow, which some people disagree about prehistoric landscape conditions.  You can have debates about natural flow, but unimpaired flow is pretty much agreed upon.

For the analysis, they looked at the daily estimates of unimpaired flow, applied a daily pattern to a monthly volume, and then compared it to actual flows.   “In the 2015 daily example, we’re capturing a big part of the peak flows – 39% of the December peak and 73% of that February peak,” he said.

When you compare unimpaired flow versus actual flow on an annual basis, the trend over time can be determined.  “Here we have Delta outflow in the months of February through June in million acre-feet on the y axis,” Mr. Reis said, noting that February through June is one of the most ecologically important as it’s when most of the native fishes are in the estuary, the most protections are overlapping, and it tends to be when the most water is flowing through the system.  Focusing on that critical period, they considered unimpaired flow in quintiles, and subsorting it into bins of 20% for each year type.  The wettest 20% of years are in blue, then green, yellow; the dry years are brown, and the supercritical years – the driest 2% of years, are shown in black.

Anything drier than 2015, which in the 1922-2016 record occurred twice – in 1925 and 1977, so those 2% or super-critical years are pretty rare,” he said.  “What did the estuary actually experience?

He presented a slide comparing the actual outflows (lower chart) to the unimpaired flows (upper chart), pointing out there are a lot more black bars in the actual outflow in recent years.  “Since the 1995 water quality control plan, it’s over 40%,” he said.  “Since 1967, over 40% of the years are in that driest 2% category of unimpaired flow.  If you look at the red and black parts together, what would on a unimpaired basis would be about 20% of the years, but over 60% of years have actual outflows that low going to the estuary.”

If we look at a percent of unimpaired flow trend, back in the 30s and 40s compared to present day, now you get 75% only in the wettest years,” he said.  “We can’t hold the wet years, they happen, but we’re mining out the middle years and turning the middle years into dry years, and we are actually capturing more of the wet year flows as well.”

So as to the first assertion: are flows increasing or decreasing, apparently they are decreasing still, he said, adding “although the caveat is that the Water Quality Control Plan has set some very basic standards to make a floor for the worst years, so we are actually getting some improvement at the bottom end, but mostly they are decreasing.”

Assertion:  Project water exports are frequently limited, or even halted, by regulations to protect endangered species

Mr. Reis next presented a slide showing the primary data sources for the frequency and volume analysis; he presented another slide showing the different types of constraints:

He noted that the various sources give various answers to what limited exports on any given day, so he had to interpolate for the days that were missing and figure it out from flows and regulations.  The result was a daily analysis from 2011 to 2016, with the major categories of constraints being the Water Quality Control Plan, the Endangered Species Act, and system capacity.

The Water Quality Control Plan has two different elements: the hydraulic salinity barrier necessarily for human uses and fish and wildlife protections.  “We used a layered approach,” Mr. Reis said.  “You need the hydraulic salinity barrier to keep the Delta fresh for human uses to even have exports, so that came first.  Then we layered on the fish and wildlife protections, and then under the Endangered Species Act, anadromous fish and the smelt Reasonable and Prudent Alternatives (RPAs) under the biological opinions.  And we separated out those categories.”

Then system capacity is physical capacity such as full storage, canal capacity, pump capacity, or maintenance shutdowns, he said.

He presented a slide showing the temporal analysis which examined what days had limiting factors from the three main categories: the Water Quality Control Plan in blue; the RPAs from the biological opinions in orange for anadromous fish and in pink for Delta smelt.  He pointed out that not all of these factors are in effect all year long; for example, the anadromous fish RPA could be controlling November through June potentially, and Delta smelt September through June potentially.  He also noted this is not always in every year, but those are the times of the year they could be controlling things.

He presented a slide of the daily analysis of what was controlling for the years 2011 – 2013.  “Clearly fish and wildlife objectives (shown in blue) was not controlling all year, but you can see the times it was,” he said.  “The hydraulic salinity barrier is at the top in red; in 2011 was a pretty wet year, 2012 reservoirs were still pretty full, so that wasn’t controlling that much.  There was a lot of water to move across the Delta and that means the anadromous fish and the Delta smelt RPA did a lot of controlling and down at the bottom.  There were a lot of capacity constraints in 2011 because of the wet year.”

I think of these RPAs as kind of like speed limits on how much water you can take across the Delta,” he added.  “When there’s a lot of water there to move, it’s setting a limit and saying this is the limit at which you can do it, and in the drier years, you never even hit those limits some times.”

Mr. Reis then presented the next three years of the analysis from 2014-2016, noting that the drought was pretty deep during that time and so there were no infrastructure limits that were hit, and that the hydraulic salinity barrier (shown in reddish brown) was controlling most of the time.

He pointed out that in some years like 2014, the smelt controls didn’t happen at all.  “So contrary to what news articles you might have been reading in 2014, Delta smelt didn’t control anything that year,” he said.

Mr. Reis then presented a tabular display of the last chart, highlighting the main categories in yellow.  “If you look at that left column, the hydraulic salinity barrier controlled about 35% of the time, the water quality control protections for fish and wildlife about 29% of the time, anadromous fish RPAs 17% of the time, smelt about 6% of the time, and system capacity 11% of the time,” he said.  “Clearly there is variation between years, but this is a pretty good representative set of years.  We had a wet year in there and then some dry years.  So Delta smelt weren’t ending the world there with only 6% of time controlling.”

Assertion:  Regulations designed to protect endangered species, and Delta Smelt in particular, are the principal restrictions on exports and are thus responsible for most of the water that flows from the Central Valley to San Francisco Bay.

The third assertion is a question of volumes of water.  Mr. Reis explained that even if something is controlling, on a certain day, and that protection is removed, it doesn’t necessarily mean that water is freed up; there is something else that will be controlling.

So we parsed out the volumes on each day,” he said, presenting a slide showing results for five different individual days.  “There’s a center bar with export volumes on the left, and outflow of the Delta to the Bay on the right.  The graphic shows five different days from our analysis as examples with what was controlling on that day named above the date.  On the right side of the graph are the acronyms, and whatever is closest to the bar was controlling that day.”

He focused on the information for February 23, 2016 (the third from the top), noting that the Endangered Species Act and Fish and Wildlife were the controlling factors; the stipled area represents the export capacity.

So if you were to eliminate the ESA protections and move that water over to the left side or the export side, the next thing to control would be fish and wildlife, and if you were to eliminate the water quality control plan protections and move that volume over to the left side, you would max out the export pumps and not be able to take any of the additional uncaptured outflow (AUO),” said Mr. Reis.  “If you expanded the size of the export pumps or built new tunnels, then you can start digging into the AUO but we did not look into that analysis.  The HSB, the hydraulic salinity barrier, is assumed to be necessary for exports to exist in the first place, so we didn’t parse that one out as far as a limit on exports, but we did quantify it.

He then presented the results from the volumetric analysis in a pie chart, with the large chart showing the results for all the years in aggregate, and the bottom showing the pie chart for each individual year.  He noted that the Delta smelt is parsed out for the larger pie chart, but in the pie charts for the individual years, it is lumped in with the ESA total amount.

As you can see, the amount attributed to Delta smelt is small – just 1%; it’s actually about .6% if you don’t do the rounding.  Anadromous fish is 3% and Fish and Wildlife is about 4%,” he said.  “The hydraulic salinity barrier was responsible for 17% of the watershed runoff being used to repel salt and Delta exports; Delta exports were 18%; Delta net use 5%; upstream net use was 25%; and additional uncaptured outflow was 27%.  I should mention, some of the additional uncaptured outflow is required by the water quality control plan, but if it exceeded the export capacity, we didn’t quantify that in our analysis.  We’re looking at the existing capacities and what actually occurred.  We didn’t reoperate reservoirs or do any modeling, it was just what actually happened.”

Mr. Reis acknowledged there is some uncertainty.  For their analysis, they overestimated the key things such as the anadromous fish, the smelt, and the water quality control plan, so those are maximum numbers; underestimated volumes are the HSB, hydraulic salinity barrier and infrastructure.

In conclusion …

He then gave his conclusions.  “Are Delta outflows increasingly large? We found they are not.  Project water exports are frequently eliminated or halted is not true, and Delta smelt really being responsible for most of the water going out to the Bay is also not true.

Question:  For the hydraulic salinity barrier, there are two reasons for that.  There’s a PPIC report on a similar subject broken out.  There are the flows you need going out in order to have the exports, and the flows you need going out in order to keep the salinity ok for some of the urban and agricultural uses of the Delta.  Do you have any comment on that?

We were doing our study at the same time that Greg Gartrell and PPIC were doing theirs, and once that study came out, I pasted in Greg’s numbers to compare,” said Mr. Reis.  “I was starting to get negative numbers at times where his requirement for outflow was higher than the actual outflow on that day, so I didn’t take the time to try and figure that out.  I think his numbers were more modeled than ours, and outflow is a modeled number too.  There’s about a million more acre-feet that Greg found than what we have in the hydraulic salinity barrier, so we were conservative on that, and we used a number that the water board was using, 4800 cfs as kind of a long-term average, and or whatever actually observed on that day if that was what was occurring.”

(Jeanette Howard from the Nature Conservancy and Jonathan Rosenfield at The Bay Institute also contributed to this work.)

Daily emailsSign up for daily email service and you’ll never miss a post!

Sign up for daily emails and get all the Notebook’s aggregated and original water news content delivered to your email box by 9AM. Breaking news alerts, too. Sign me up!

Print Friendly, PDF & Email