CA WATER COMMISSION: The Yuba-Feather Forecast-Informed Reservoir Operations Program

Reservoirs are often constructed and operated for multiple purposes, such as water supply, flood control, hydropower, recreation, and environmental enhancement. However, managing the reservoir for multiple purposes is challenging and often involves tradeoffs between sometimes competing needs.

Forecast-informed reservoir operations (FIRO) is a reservoir-operations strategy that uses enhanced monitoring and improved weather and water forecasts to inform decision making to selectively retain or release water from reservoirs to optimize water supply reliability and environmental co-benefits and to enhance flood-risk reduction.  FIRO provides an effective means of increasing the efficiency and resiliency of existing water resources infrastructure – all without costly construction projects.

John Leahigh is the Water Operations Executive Manager for the Department of Water Resources with 25 years of experience either in or overseeing the State Water Project Water Operations Office for the Department.  At the May meeting of the California Water Commission, he spoke about the Yuba-Feather FIRO program, the promise of Forecast-Informed Reservoir Operations, and how it can be an adaptive response to the effects of climate change.

Mr. Leahigh began by noting that California’s hydrology is unique in that it has the highest variation in annual precipitation of any state in the entire country.  It’s just a handful of storms every winter that will basically either make it or break it for the state’s water supply.

The chart shows the runoff in the Sacramento Valley for the past 120+ years. 

“What strikes you as obvious is the varied pattern that we see in terms of the annual precipitation that varies up to tenfold from one year to the next,” Mr. Leahigh said.  “This year’s most recent drought emergency is example number one. We’re more likely to end up as the third driest in the 100-year record and the driest since 1977. And this is coming just four years after the wettest year on record 2017 for the north part of the state.”

“Some would look at this graph and conclude that there really is no such thing as a typical year; it’s either boom or bust when it comes to floods and droughts in the state.”

It’s only going to get more challenging with the warming climate, Mr. Leahigh said.  The chart on the right shows the clear trend of increasing temperatures in the state over the last several decades.  The warmer temperatures will change the proportion of precipitation, with more precipitation falling as rain versus snow in the Sierra Nevada, leading to larger floods in the winter and less snowpack to store water to make it through the long, dry summers.

Current water management in the state and flood risk analyses depend on the historical estimates of hydrology, but those assumptions will be tested by observations not experienced in the last 100 years, both in terms of floods and droughts.  This requires us to better utilize the advances in forecasting that have been made in recent decades, he said.

FIRO: Leveraging improvements in weather forecasting

“What we want to do is quantify and leverage those improvements, continuing to invest in those improvements as we move forward,” said Mr. Leahigh.  “If we can forecast a large event approaching the watershed, reservoir releases can be made in advance of the peak inflow to create more vacant space in the reservoirs and thereby attenuate the peak outflow downstream to achieve flood control benefits. And on the flip side of that, if no storms are forecasted, especially late in the spring, water can be held in storage or conserved, yielding an increase in water supply benefits.”

He described these rule curve changes as software changes to potentially improve the performance of existing infrastructure, much like a well-known electric car company offers software upgrades to download into the vehicle to enhance the car’s performance.

One of the first reservoirs where FIRO has been implemented is Lake Mendocino in Sonoma County. Rather than using the fixed guide curve or rule curve, a FIRO space is used to provide increased flexibility on the amount of vacant space required at any given time, leveraging the inflow forecast by providing additional empty reservoir space when major storms are forecasted and requiring less flood space at the end of the season to maximize water in storage when storms are no longer a threat.

Lake Mendocino is one of at least three official ongoing FIRO projects in the state. Prado Dam in Southern California was the second, and the third project is the combination of New Bullards Bar Dam operated by Yuba Water Agency and Lake Oroville operated by DWR. 

“This constitutes the official third FIRO project in the state and the first one that is a multi-dam project attempting to integrate the FIRO project,” said Mr. Leahigh.  “This would be the first project where the rain-snow interface is a key component in achieving accurate forecasts for inflows.  And Lake Oroville is a key component of the overall state water project system. The water supply from the project touches a great portion of the state’s cities and farmland.  It is a huge contributor to the economy of the state.”

The slide shows the current status and timelines of the various FIRO projects.  The Lake Mendocino FIRO project is now finalizing its viability assessment for that project; Prado Dam is just finishing its preliminary viability assessment; and the Yuba-Feather RIver FIRO project is finishing up the work plan that will set the roadmap for both the preliminary viability assessment and the final viability assessment.

He noted that the fourth project on the slide is the Howard Hanson Dam in Washington State, which is forming a steering committee to look at the potential for that reservoir.

There are many partners on the Yuba Feather FIRO project.  Besides the Department of Water Resources and Yuba Water Agency,  Dr. Marty Ralph from Scripps Institute at UC San Diego, the National Weather Service, and the National Marine Fisheries Service are partners. In addition, two branches of the Army Corps are involved: the research and development division of the Army Corps out of Vicksburg, Mississippi, and, locally, the Sacramento district water operations arm of the Army Corps. Finally, the Sonoma Water Agency is also involved to leverage the lessons learned from the work at Lake Mendocino.

FIRO builds on previous programs

The Yuba-Feather FIRO initiative builds upon past programs and, in particular, the Forecast Coordinated Operations Program, which was the initial partnership but did not have the research element to it that the FIRO program does.  That program kicked off approximately 15 to 20 years ago with the goal to better coordinate operations and build common decision support tools for operating during significant storm events that hit both the Feather in Yuba watersheds. This partnership is vital because Yuba Water Agency and the Department of Water Resources have a common downstream control point for flood management purposes in both the water control manuals, and so that coordination is critical.

The FIRO program goes beyond that and initiates research investigations to improve the forecasts to build upon the forecast coordinated operations program. In addition, the program will be developing and conducting viability assessments for possible formal operational rule changes at both Lake Oroville and New Bullards Bar.

“The benefits we expect to receive from this is both of flood risk reduction and possible water supply savings as well,” said Mr. Leahigh.  “We hope this effort will help inform the development of a water control manual update; that process is also currently underway for both Lake Oroville and New Bullards Bar. Those existing manuals are dated back to the early 1970s, so it definitely would be a good thing to take a fresh look at those and update those Corps manuals.”

“Ultimately, at the top of the pyramid, we are looking to operationalize and leverage the forecasting improvements in the decisions that are made for flood management purposes as part of an update to the Army Corps water control manual,” said Mr. Leahigh.  “We’re looking for something that’s very resilient into the future that can continue to leverage any improvements that we continue to see on the forecasting capabilities.”

Investing in atmospheric river research

Understanding atmospheric rivers is critical for both floods and droughts, as the state’s water supply is linked directly to these atmospheric river events.

“It’s important to understand when and where these events will occur,” said Mr. Leahigh.  “But it’s also important when we expect not to see these events in any given year, especially towards the end of the season, when the flood control manuals allow for the refill of the various reservoirs.   Therefore, we’re putting a substantial amount of investment in improving the monitoring capabilities, both ocean reconnaissance and land-based monitoring.”

The slide lists some of the research investments.  In terms of ocean reconnaissance, the C 130 Hurricane Hunters that help provide surveillance for hurricanes during hurricane season can be redeployed on the West Coast for flights over the Pacific to observe atmospheric rivers as the two seasons are counter to one another.

In terms of land-based monitoring, the snow level vertically pointing radar is essential.  “We’ve invested in new sites on the coast and inland to the foothills near Lake Oroville and other places along the foothills as it’s extremely important to get that that snow-rain elevation dialed in as much as possible,” said Mr. Leahigh.  “That is the key to determining how big or how hazardous a flood event would potentially be. So those investments are going to be critical in feeding into the weather models.”

For forecasting runoff, understanding soil moisture is critical, so the program is investing in soil moisture instrumentation. “You can do very well in forecasting the meteorological events, but translating that into actual runoff into the reservoir is just as important a component of the forecasting.”

A scale for atmospheric rivers

Researchers at Scripps have developed a scale for atmospheric rivers, similar to the way hurricanes are rated in intensity.  The two primary attributes of an AR are the intensity and the duration. Categories one and two are considered mostly beneficial in terms of providing water supply.  Categories four and five are considered hazardous as the volume of runoff can overwhelm the dams and levee systems.  Category three lies somewhere in between.

“That’s where it’s critical to get the freezing level,” said Mr. Leahigh.  “The wetness of the watershed plays an extremely important role on whether those kind of middle type atmospheric rivers are going to be on the mostly beneficial or hazardous side.”

Benefits of FIRO

Lake Oroville’s existing flood control diagram may be a natural fit for FIRO space.  “The amount of flood space required varies from 11% to 22% of the capacity of the lake that needs to remain empty, depending on the wetness of the watershed,” he said.  “That’s a function of how recent storms have hit the region and how many.  That same variable space could be utilized in a more sophisticated way, we think, in which the wetness of the watershed is just one of many inputs that would be part of the forecasted inflow due to an AR or any storm approaching the Feather watershed.”

As for flood risk reduction benefits, the other consideration for the Yuba Feather FIRO program are the common downstream control points within the levee system on the lower Feather River, downstream of the confluence with the Yuba. Only one of the major branches of Yuba River is controlled by New Bullards Bar, which the Yuba Water Agency operates.  It is thought that by making moderate releases in advance of a forecasted storm, higher releases can be avoided later in the storm, thereby decreasing the high flow events on that downstream levee system. 

Regarding potential water supply reliability improvements, the graphic on the slide shows a hypothetical application of a FIRO type approach to the operations in 2004. 

“This particular hypothetical shows the potential for water supply reliability improvements of over about 100,000 acre-feet above what the existing 1970 water control manual rules would have required for filling the reservoir later in the spring,” said Mr. Leahigh.  “Additional conservation in these types of years would go a long way to better prepare the system for the inevitable droughts that we are faced with, the current year being a prime example of that.”

Schedule and study outcomes

The current project schedule for the FIRO program is being interwoven with the parallel effort of the Corps’ update of the water control manuals for both Lake Oroville and New Bullards Bar.

“There’s a lot of interplay between the two processes,” he said.  “I think the Corps is very encouraged about this approach, and the same agencies are involved in both of these processes moving forward. So we’re expecting to have some output from both these processes by the mid-2020s.”

There are two expected outcomes for the Yuba Feather FIRO program:  The first is the operationalizing of the forecast informed reservoir operations process in an updated set of flood rules in the water control manual that is more resilient. The second is continuous investments in forecasting capabilities.

“By resilient, I mean a water control manual that is flexible enough that it could leverage any continuous improvements in the forecasting capabilities that we might see out of our ongoing investments in the AR research and forecasting capability improvements,” said Mr. Leahigh.  “Those investments are being made through a partnership that includes all levels of government – federal, state, and at the local levels. So those would be my two take-home points in terms of what we’re hoping to achieve from the FIRO program, and specifically the Yuba Feather.”

John James, Yuba Water Agency

Next, John James, the Water Operations Project Manager for Yuba Water Agency and one of the co-chairs of the Yuba-Feather FIRO program, gave a brief background on the agency, the program, and its relationship with the State Water Project.

The Yuba Water Agency is a standalone public agency established in 1959 to reduce flood risk and improve water supply to benefit the people of Yuba County.  Mr. James noted that Yuba Water has had a long history of proactive project and program development within the county, beginning with the Agency’s Yuba River development project and the construction of New Bullards Bar Dam and Reservoir.

“With that, we’ve also had a long history of partnership and cooperation with DWR and the State Water Project, particularly with the coordinated flood operations between New Bullards Bar and Lake Oroville,” said Mr. James.  “As part of the Yuba Feather FIRO program, Yuba Water is taking the same progressive forward-looking approach to building climate resilience in our watershed by further enhancing operational partnerships with DWR and other state and federal partners.”

In parallel with the Yuba-Feather FIRO Program, Yuba Water Agency is currently planning a new secondary spillway at New Bullards Bar that is designed to be operated under FIRO.  This additional outlet will have gates lower in elevation in the reservoir, which will significantly increase release capacity and allow for increased flexibility in managing the reservoir and the downstream watershed by creating additional flood storage space ahead of flood events which can then be used to reduce regional downstream flows during the peak.

The updates to New Bullards Bar and Lake Oroville water control manuals and the new spillway project at New Bullards Bar will significantly improve public safety and climate resiliency, especially given the increased flood events and droughts projected under climate change.  

“These climate change impacts are already being observed and will challenge our infrastructure going forward,” said Mr. James.  “The current and projected conditions indicate how important FIRO implementation and complementary projects are in mitigating the impacts now and into the future. FIRO is truly using the latest and best science and technology. Our collective partnership with Scripps, the Army Corps, DWR, and others is really moving the needle with forecast improvement and confidence in atmospheric river forecasts and other weather and hydrologic forecasting that water managers rely on for their operations. FIRO emphasizes the flexibility in operations and allows us to get the most value out of existing and future infrastructure, improving public safety and water management capability.”

“FIRO also helps inform related forecasts improvement opportunities, specifically with seasonal water supply forecasting that can support adaptive water management strategies in critical years such as this one,” said Mr. James.  “The program supports continued strengthening and institutional trust between partnerships and organizations and enhancing these partnerships, improving technology, and implementing flexibility into our operations will be essential to adapting and managing our water resources in the era climate change.”

QUESTION & ANSWER HIGHLIGHTS

QUESTION: Chair Teresa Alvarado asked, why would you not use FIRO? What would be the hesitation?

“For one, unless we actually formally change the rules that we’re required to operate under the Corps manual, that change is not available to us,” said Mr. Leahigh.  “Also, before we want to make a change, we want to be absolutely certain – this is a significant departure from what the flood operating rules are today, in terms of relying more on those forecasts.  I think we need to be very deliberate in capturing the level of risk that we would potentially be looking at in terms of relying on forecasts more than we do today. I think it is essential, though, to start leveraging the technological improvements that we have seen over the last few decades. It’s very clear to me that we can do better through the use of those forecasts, but we need to be deliberate about it.”

QUESTION: Chair Teresa Alvarado noted that oftentimes, there are trade-offs and different stakeholder interests that came into play around environmental protection and habitat and recreation, and other things.  What about stakeholder involvement?

“As part of the process, there will be a fully vetting in terms of potential environmental effects,” said Mr. Leahigh.  “We do think that there’ll be substantial improvements to the reduction in flow frequencies at certain levels to the downstream levee system. But we have to also look at them and ensure that this will not put us sideways somehow with the environmental flows.”

“I think there is potential for improvements on the on the environmental side in terms of being able to capture more of the peak winter inflow, hold it as storage, should there be no threat for additional storms, this could actually stabilize flows a little bit more in the spring, later in the spring, which would be a plus I think, for the fishery,” said Mr. Leahigh.  “So rather than cutting back releases to a minimum when we’re trying to fill during that spring fill period on the existing flood control diagram, if we’re able to capture some of those flows earlier in the year, we wouldn’t need to reduce the flows as much in the spring period. So I think there actually could be potential for side benefits for the fishery.  At the very least, we want to ensure that we’re not impacting the fishery from the status quo. So we have to be very deliberate about every step of the way, along here. Certainly, we’re still a little bit early in the game compared to some of the other FIRO projects. But as part of both the water control manual update process and the FIRO, we will be amplifying the public outreach parts of this project and this program moving forward.”

“NOAA Fisheries is part of the Yuba-Feather FIRO program, and additionally, that as part of the updates to the water control manuals for New Bullards Bar and Lake Oroville, they are scoping out that public stakeholder process,” added John James.

QUESTION:  Commissioner Solorio asked about releasing water earlier.  “Any early release is good, so we can put it to better use, whether it be in Orange County, with our groundwater basin, or other parts of the state. My question is, how much of an early release? Are we talking about, like a day or two, a few hours or a week or two weeks?”

“It would be measured in days for the earlier release, and it’s going to be dependent on the reliability of those forecasts and how we quantify that reliability,” said Mr. Leahigh.  “We’re talking days in advance, not hours.  We used to be able to forecast only a day out; now, we have the same level of skill closer to five days out. So, that gives you an idea of potentially somewhere in that range. And, one of the things we’re looking at is, as those forecasts improve over time into the future, we want to be able to utilize those improvements as part of a more flexible water control manual that would come out of this whole process.”

COMMISSIONER MAKLER:  If you were to think about the deployment of this technology, what is the potential for changing the priorities or the level of capital investment?  Do you think that this level of the forecasting and the level of flexibility that may provide you in the future would materially impact kind of capital planning decisions?

“I think there’s a lot of promise in that area,” said Mr. Leahigh.  “At least for Lake Oroville, the current rules are kind of blind to the forecast in terms of the requirements.  By making it more dynamic, in a sense, we’re almost building out additional reservoir space for both on the flood side … One of the objectives, for example, is the Marysville dam, which was never built on the Yuba river system was going to provide a certain amount 100-200,000 acre-feet of additional flood space for the system.   One of our objectives here is to essentially make up for the fact that the dam was never built and be able to provide the same sort of performance through just the software changes that would have existed with the existence of that physical structure. So on the flood side, that’s the kind of benefit that we’re hoping to achieve.”

“I gave an example for water supply reliability of 100,000 acre-feet saved,” Mr. Leahigh continued.  “It would be the equivalent of an additional expanding reservoir by 100,000 acre-feet. And that’s strictly done through rule changes or software changes. So I think to the extent that we get that bang for the buck, in terms of those kinds of changes, that just frees up resources for other infrastructural projects elsewhere.”

QUESTION:  Commissioner Curtin asked if releasing water and moving it into groundwater basins was being considered as part of the plan.  You could use that as your additional flood space and also have environmental benefits.  Is that part of what you are working on, or is that another silo that we have to connect?

“Flood MAR is an example of that, and while that’s not explicitly part of this particular project, I think that represents the same concept – how can we improve management of the system without building new dams and without necessarily putting in major infrastructure,” said Mr. Leahigh.  “Certainly, there’d have to be some infrastructural improvements to maximize recharge and recovery of these groundwater basins, of course, but probably minuscule compared to a major dam project, but you could achieve the same sorts of improvements as you’re suggesting. So I think there is a lot of great promise with those as well. And so I don’t think it’s an either-or; I think it’s ‘and.’ … “