Update on DWR’s Reservoir Reoperation Study

Shasta sliderboxDWR’s Ajay Goyal tells the Delta Stewardship Council that there are benefits to reoperation of the state’s system, but they look to be limited

Senate Bill X2 1, passed in 2008,  allocated resources for the planning and feasibility studies to identify potential options for the reoperation of the state’s flood protection and water supply systems that would optimize the use of existing facilities and groundwater storage capacity.

At the June meeting of the Delta Stewardship Council, Mr. Ajay Goyal, Ajay Goyalprincipal engineer and branch chief of the Statewide Infrastructure Investigations Branch of the Department of Water Resources, gave a summary of the system reoperation study, its findings, and the next phase of the program.

Ajay Goyal began by saying that the state legislature directed the Department of Water Resources to conduct a planning study to identify potential options for reoperation of the state’s flood protection and water supply systems with an interest of optimizing the existing facilities in conjunction with groundwater, and to simultaneously improve water supply reliability, flood protection, and enhance the ecosystem.  The legislation specified that the study should look at integrating flood protection and water supply systems,  reoperating reservoirs in conjunction with groundwater storage, and improving conveyance.

Goyal DWR-System Reop_Page_03The first step was to develop a list of reservoirs, so they identified the large reservoirs in the state that they felt had potential of providing additional benefits through reoperation: Shasta, Oroville, Folsom, Don Pedro, Comanche, Pardee, and New Exchequer Dam (Lake McClure), and Friant.  They also decided to look at improving the integration of CVP and the SWP systems.  They then met with the reservoir owners to see if they were interested in participating in the project; they also met with more than a dozen groundwater districts in the San Joaquin Valley, in the Kern area, and also in Southern California.

We found that most of the reservoir owners were not comfortable with DWR trying to reoperate the reservoir, and some were going through FERC relicensing,” Mr. Goyal said.  “South of the Delta, most of the groundwater districts have a lot of empty space, and are looking for water.  They were all interested.”

Goyal DWR-System Reop_Page_05With that information, they narrowed it down to four strategies that were evaluated in the study:  reoperation of Oroville, Shasta, and Lake McClure which is on the Merced River operated by MID, and looking at trying to improve integration between the Central Valley Project and the State Water Project.

Chair Fiorini asks why New Melones was not on the list.  “We have not considered that.  Our team felt that it did not have much flexibility at the time we were working on it, but we can always look at more reservoirs in future phases,” Mr. Goyal answered.

The analysis was divided into four reoperation components that matched the objectives:  Improving supplemental ecosystem flows, groundwater conjunctive management, forecast-based operations for reservoirs, and water resource system integration. The last component mainly applies to the fourth strategy which is improving integration between the CVP and SWP.

For ecosystem flows, our focus was to improve spring flows and cold water pool,” he said.  “For groundwater conjunctive management, we did groundwater substitution in Feather River basin and Sacramento River basin.”

Mr. Goyal then presented a series of graphic of reservoir space to explain how reservoir reoperation was modeled.  He noted that within reservoirs, typically there is dead pool, conservation storage space, and flood control space.

The dam is operated based on the Corps of Engineers flood control diagram,” he explained.  “Reservoirs normally go into flood control from November through March.  Levels are dropped down to accept flood flows. In the analysis here for the forecast-based analysis, we allowed encroachment in to flood control up to 25% of the total flood control capacity in the reservoir.  In the same volume, we allowed to go into the conservation pool, depending on weather conditions.  If we expected less inflow coming in, if the weather was sunny in the forecast and less inflow was to come in, we allowed encroachment into flood storage space and captured more water.  But if we expected a lot of water coming in, then we allowed spilling from the reservoir and encroachment into the conservation space.  That’s how we have analyzed it here.”

The first challenge was to determine how aggressively the reservoirs could be operated, so they did a  trade-off analysis where, through modeling, they reoperated Oroville and released supplemental ecosystem flows ranging from 25 – 1000 acre-feet and monitored what happens in the system.  They next modeled groundwater substitution in Feather River basin ranging 25- 100,000 acre-feet; they then analyzed both combined.  They then did the same experiment for Shasta.

Goyal DWR-System Reop_Page_11We found that while we were able to provide a good amount of ecosystem flows, we started impacting carryover storage and had low cold water pool left,” Mr. Goyal said.  “We started reducing water supply and then also had an impact on hydropower.  But one of the main objectives that we had in the study was to meet all the three objectives simultaneously.  We cannot afford to impact one objective when we tried to improve another objective.”

So they fine tuned the flows and came down to ecosystem flows of a maximum of 150,000 acre-feet from Shasta, 150,000 acre-feet from Oroville, and then connected the groundwater management in Feather River and Sacramento River but limited to 100,000 acre-feet in each basin, he said.

Five strategies were analyzed: Oroville operations separately, Shasta operations separately, Oroville and Shasta operations combined, Lake McClure, and then improved integration between the Central Valley Project and the State Water Project.

Goyal DWR-System Reop_Page_13These are the assumptions we have made in the analyses,” Mr. Goyal said.  “We did supplemental ecosystem flows in wet, above-normal, and normal years from March through May.   We did conjunctive management from May through August during dry and critical years only, meaning we supplied surface water during wet and normal years to current groundwater users because we have separate flows in the system; during dry and critical years when water is less, we held back the water in the reservoir to improve cold water pool and had increase in groundwater pumping, meaning we put some of the surface water users on groundwater during dry and critical years.”

Mr. Goyal noted that this analysis was done prior to the approval of the Sustainable Groundwater Management Act.  “Now all the groundwater basins will have to show sustainable levels, and groundwater substitution could be difficult to do everywhere, although in some areas it may still be possible,” he said.

Forecast-based operations were typically done from November through March when the reservoir is going in flood control.

Mr. Goyal then presented the results of the joint reoperation of Shasta and Oroville, noting that in this study, they also included Folsom.  The modeling assumptions were forecast-based operations for all three reservoirs together; ecosystem flows were released from Shasta and Oroville with a target of 150,000 acre-feet in each of the rivers, and conjunctive management was kept at 100,000 acre-feet per basin.

He presented three different scenarios: one under existing Delta conveyance, one with climate change, and the third with Cal Water Fix.  “We wanted to see how much improvement we can do to ecosystem flows, so when we had surplus flows, meaning we had met all the requirements in the biological opinions and in D-1641, we did not violate any of the requirements, and we met storage requirements in Shasta and Oroville; then we did supplemental flows with the 150,000 acre-foot target from each reservoir in mind; if we didn’t have enough water then we did not do it that year.

Goyal DWR-System Reop_Page_14So on average you will see that we improve total ecosystem flows; we had a target of 300,000 acre-feet but on annual average it came to about 80,000 acre-feet because in dry and critical years we did not do, plus we could not do in every normal or wet year.  Some years it happened and some did not, so the average was 80,000 acre-feet,” said Mr. Goyal.  “Water supply – this is export from the system, mostly south of Delta exports – some are north Sacramento Valley exports also, but most of them are south of Delta, and that export went from 37 from existing current baseline conditions.

Groundwater pumping, which under SGMA may or may not be possible, we had a limit of 200,000 acre-feet during dry and critical years only.  On average we came to about 55,000 acre-feet.  We had to do some dry and critical years, not all the years.  Then Delta outflow went up, but this number doesn’t mean much because annual Delta outflow is about 16 million acre-feet.  What I am showing you is the incremental change resulting from the reoperation.  Change in carryover storage –  Shasta went up some, but looking at the size of the reservoir, it’s a small volume; Folsom had an improvement and Oroville had a reduction.  So what you can see that it’s not that we’re getting a lot of benefit but there are some benefits, so that’s one observation to be made.”

Goyal DWR-System Reop_Page_16They next considered reoperation under climate change.  “We would see increase in same kind of benefit we would see even then,” he said.  “We would have inflow and more water supply coming out, we’ll have improvement in water supply, and similar ecosystem flows.  Groundwater pumping would be a little more; in dry years, we would have to pump a little more.  Outflow went down.  Reservoir storage, with climate change, and because of the hydrology and operation it came up that Oroville would have increase in carryover storage a little bit.

They lastly looked at reservoir reoperations with the California Water Fix.  “The water supply went up even more, mostly south of Delta,” he said.  “We had several conditions in Water Fix: one was we used the bypass flow criteria for the Water Fix; second is we let all the water for ecosystem flows go out to the ocean, and third was we picked up all the water.  The one we are showing here is using the bypass flow criteria for Water Fix.  You see improvement in reduction in groundwater pumping and reduction in Delta outflow; it’s because most of that water is already covered in the baseline in Water Fix.  There are higher outflows; that’s why it’s not showing that.  Similar change we see in carryover storage.”

So in a way, yes, we can provide benefits, but it’s kind of limited,” Mr. Goyal said.  “We also showed that the existing system is highly optimized and is operated quite well to meet current regulations in the biological opinions.”

Goyal DWR-System Reop_Page_17Mr. Goyal then presented the results for reoperation of Lake McClure on the Merced River, noting that the Merced Irrigation District had given them a limit on how much to flex the reservoir of 15,000 acre-feet, so their analysis was based on that.

For forecast-based operations, we allowed encroachment up to 50,000 acre-feet, so most of benefit which we see is coming from forecast-based operations.  Change in storage, there was an improvement and we were able to do ecosystem flows plus get water supply.  So it’s pretty reasonable. We were able to get some benefits.

Goyal DWR-System Reop_Page_18The other strategy was looking at improving integration between the Central Valley Project and the State Water Project.  “We found that both systems are very well integrated,” Mr. Goyal said.  “We prepared a spreadsheet model and gamed it in all different ways.  We combined both CVP and SWP and operated them as one single system; storage was shared, conveyance and diversion capacities were shared.  Where we saw there was potential was where we had gaps in banks to convey additional water, we did, and then also used combined reservoir storage.  The additional benefit we could get is anywhere from 150,000 acre-feet of additional water supply, and this depends on how aggressively we repoperate the system.”

Another thing we learned is that some of these techniques we tried here are already being applied by the operators, but we don’t know which ones and to what extent they’ve been applied, so the benefits could be less than this because some of it is already captured,” Mr. Goyal added.  “For a system that yields 4.5 million acre-feet, 100,000 acre-feet is very small number.  It’s like marginal.”

“So in conclusion of this study, we say, yes you can get benefits but they are going to be limited.  With Water Fix and climate change, under those conditions we would have improved benefits, benefits would go up, and then FBO and conjunctive management are already being applied by the operators informally.


In the next phase, they plan to make sustainable groundwater management an objective of the study, and so they will try to meet that in future reoperation, and also include some of the potential reservoirs that would be funded by Prop 1 when we look at reoperating the existing system.  “When we have more storage, we might be able to get greater benefit from reoperation of the whole system with the integration concept,” he said.  “We also want to look at reoperating reservoirs with the Water Fix attached to it and try and see how much more we can get.  We didn’t analyze it to the level we could, but maybe in the future we might look at that.”

The other thing is over the last 4 years, we had some special concessions made to the system because of drought,” he said. “Operators were allowed to modify the operations as compared to what is in the biological opinions, and D 1641 because of drought, so we want to capture some of those in our modeling going forward.”

Mr. Goyal said that in the SGMA scenario, there are five basins in the Sacramento and Feather River systems which are high priority basins, so the study will look at reoperating Shasta, Oroville, and Folsom, and conveying all the water to these five areas, how much would it produce.

We supplied surface water to these areas during wet and normal years and had them reduce groundwater pumping, but during drier years, we allowed them to increase pumping and had surface users put on groundwater to protect cold water pool in the reservoir and storage in the reservoir,” he said.

He presented a chart of groundwater operations, noting that all the excess water goes into recharge.  Orange shows benefits through reoperation and blue is unappropriated flows that could have captured and provided for recharge, and below the line is groundwater pumping.  “So on average we were able to get some benefits for reoperation and from unappropriated flows; groundwater pumping was about an average about 35,000 acre-feet.

If you were to look at the 82 year period, we would have put in storage about 58,000 acre-feet per year, adding up to about 5 million acre-feet, and 57,000 AF per year is coming from reoperation, 36,000 is coming from unappropriated flows, and pumping is on average is about 35,000 AF.  So this slide shows change in outflow on average a reduction in outflow of about 57,000 acre-feet and no change in exports.  We left exports as it is, we took all the benefits and provided it to these five areas for recharge.”

That completes my presentation.


Mary PiephoCouncilmember Mary Piepho asks, “You indicated on slide 16 the compilation of the three different scenarios, with California Water Fix and the Delta outflows at a -68, and overall you are referring to an improved benefit.  I’m trying to correlate Delta outflow reduction with a system classified as an improved benefit, because from my perspective, that doesn’t equate.”

That outflow reduction is at certain times of the year,” Mr. Goyal responded.  “But we are supplying water that in spring periods especially when it is beneficial to the tributaries in the Sacramento River, so there is a tradeoff always.  Either we are improving cold water pool or we are improving flow conditions in the river, so we have to look at all of them together.  And see which one is more important to us.  It is a tradeoff all the time.  The reduction in outflow that you see, there is total outflow of 16 million acre-feet per year, so we are reducing it by 60,000 acre-feet.  Percentage it’s like .0 something percent.  Very small percent, but what we also have to look at is what period it is.  And we can always adjust it.  We did not do that level of a detailed analysis here.

Councilmember Piepho asks what time-step is the frequency of the modeling; is it monthly?  Mr. Goyal says it is monthly; they are using CalSIM.  Ms. Peipho notes that monthly timestep isn’t always the best timestep.  “We’re talking about capturing the most when we have it and the changes in the system do happen more frequently than a monthly data point, so I would encourage consideration of a more current, even if it’s weekly,” she said.

Randy FioriniChair Randy Fiorini echoed the sentiment.  “We’ve been told even hourly time steps are an important for water and ecosystem management, but daily, weekly would provide a more accurate depiction of what actually is available,” he said.

It seems like the one constraint that had some results was the redefinition of the flood control pool and the carryover pool,” said Councilmember Ken Weinberg.  “Did you feel like maybe the current flood control rule curves might be too conservative, and maybe that warrants some adjustment to preserve greater carryover storage, or is that something you can take away from this?

If it officially gives us an allowance to be able to flex the curve up and down depending on weather conditions and coordination with other reservoirs, that will lead to greater benefits,” said Mr. Goyal.

Ken WeinberegI was interested on the SGMA scenario – I think you held the deliveries to contractors, you said you held that constant, so did you reduce outflow to recharge groundwater basins?” asked Councilmember Weinberg.

During wet years, we reduced outflow by 111,000 acre-feet, but on average it comes down to 50,000 acre-feet in reduction because it has to come from somewhere – that water for recharge,” said Mr. Goyal.  “This was without Water Fix and under current conditions and current demands and this was above and beyond what was allowed by the biological opinions and D 1641. … It would be very difficult to implement this kind of scenario because of course we are taking some risk.  All of this is based on risk, and some of the contractors might be concerned at being put at risk.

Jessica Pearson asked what can decision makers do, based on this information?

What we have is that we have created a process as to how Jessica Pearsonto evaluate a reservoir when looking at reoperation, how you can decipher benefits and different types of objectives you can see, so this framework that we have developed can be used by other projects, other reservoir owners, in looking at reoperating their facility,” said Mr. Goyal.  “We are happy to assist them with it, guide them as to how we’ve done it, and use this work, apply it.  That’s the benefit.

But there’s not really a formal mechanism for Mark Cowin and the Bureau to say do this,” said Ms. Pearson.

The benefits I showed are dependent upon groundwater substitution in Sacramento and Feather River basin,” Mr. Goyal said.  “Many of those, some of them have expressed concerns to us, saying they are uncomfortable tying their groundwater to the system, so certainly it will have to be worked out with them.  As far as Merced Irrigation District’s reoperation, that’s a very easy project to implement because that’s one district, and it has good benefits; though we looked at a smaller amount, if you looked at larger amount, it might provide more, and they might be interested in proceeding with that also.”

For more information …

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