It might be difficult to believe, but California’s water-supply reservoirs are not managed according to the weather.
In the midst of the state’s ongoing drought, reservoirs in recent weeks have been releasing huge quantities of water, as seen recently at Folsom Reservoir on the American River near Sacramento. Even though the reservoir is not full, and even though there is no immediate need to empty the reservoir for flood protection, officials still let precious water go.
Why do they do this? It’s because most reservoirs are operated according to rigid rules prescribed by federal law and overseen by the U.S. Army Corps of Engineers that dictate how much water must be released and when. The purpose of these rules, known as a water control manual, is public safety: Dam operators must maintain a fixed amount of empty space in reservoirs during winter months in case that space is needed to store floodwaters.
The rules were mostly written in the 1950s and 1960s, when the dams were originally built. But the rules have not been updated significantly since then, even though there have been profound advances in weather forecasting that could allow that empty space to be managed much differently to hang on to vital water supplies.
Today it is possible to know, for example, that no storms are expected for the entire upcoming week, so there is no reason to release water. But this information has not yet altered the rules.
Rob Hartman is hydrologist-in-charge at the California/Nevada River Forecast Center , a branch of the National Weather Service in Sacramento. He’s an advocate for operating reservoirs using modern weather forecasting tools, an approach now being called Forecast Informed Reservoir Operations (FIRO). Hartman has been nudging the state and federal officials who manage California’s reservoirs to embrace this approach, and he recently spoke with Water Deeply about how it works and why it makes sense.
Water Deeply: What has changed to make Forecast Informed Reservoir Operations possible today?
Rob Hartman: It’s really hard to say there was just one thing that did change. I’ve been here doing this for about 20 years. As I look back 20 years ago, the skill and lead time in the weather forecast has really improved a lot. The computer forecast models are fundamentally better at longer lead times. In the past, we didn’t take anything seriously beyond a couple-three days. We integrated a day of rainfall forecasts and we never forecast anything beyond about a day.
Now, we’re looking seriously at things that are a week out. They may change a lot by the time they reach us. But we’re definitely looking seriously at things that are a week out, and we’re also looking at week two. Now we routinely put five-day forecasts on the web, and we can run out further than that. The lead time has increased dramatically, and it seems like the skill has improved quite a bit. People take the weather forecast a lot more seriously than they used to. That’s really the big shift, and the pressure that’s been put on Corps of Engineers’ water control manuals.
Water Deeply: So why aren’t we using FIRO already?
Hartman: There was a great deal of resistance to this, even a couple of years ago, principally from the Corps of Engineers. In the past, they’ve been very resistant to even consider these sorts of things. They want objective and well-established procedures for being able to manage the flood risk. They have a pretty high bar, and rightfully so. But they’ve really softened their stance on this quite a bit, to their credit.
When those water-control manuals were developed in the 1950s, 1960s and 1970s, the forecasting skill was not good. So as a result, those manuals make the assumption that the big storm is coming tomorrow, and you’ve always got to be ready. We know, because the weather forecasting is better, that the big storm is not coming tomorrow. And it’s not coming the day after, either.
So we don’t need to be in a place with reservoir storage to be able to handle that. We need to be able to get to that place a week from now, should a big storm show up. But we don’t need to be ready for it today. That’s the big change. Nobody is really firmly standing in the way of this now. That wasn’t the case three years ago, even two years ago.
Water Deeply: What eased their concerns?
Hartman: This whole notion has really kind of come of age. The drought in California has, I think, shined a light on it. Times have changed and there is other information that needs to be considered. And we need to manage our water resources in a much more intentional fashion, because it’s limited. It’s really tight here. We don’t have the luxury of excess water all the time. We don’t have to make decisions in the dark, and we shouldn’t be ignoring information that helps us do a better job. I think everybody pretty much accepts that.
The (weather) prediction on the West Coast, I think, is better than it is in many places for inflow forecasts in a three to five or even seven-day period for major flood events. That sort or realization has sort of worked its way into the thought process.
When you have reservoirs that are slightly encroached into their flood-control space during winter season, and you see there’s not much coming in terms of storms for the next week, then the population wonders why you let that water go. And the Corps understands that. So they want to work toward a paradigm where they have rules that allow them to hang on to that water.
Water Deeply: Folsom Reservoir has a new spillway that boosts its operating flexibility. Will the rules be changed to take advantage of that?
Hartman: The new water control manual for Folsom Reservoir will have an element of forecast operations in it. It’s not really full-blown use of the forecast, but it’s something. The Corps has had a very open and transparent process for how that control manual is going to be re-engineered, and has had consistent public forums and opportunities for people to voice concerns and provide feedback. The Sacramento district (of the Army Corps) has really done a nice job of that with Folsom.
It’s been four or five years that they’ve been working on it. There are concerns about the Endangered Species Act and (hydroelectric) power production. It adds substantial expense and time. That all has to be developed and vetted with all of the stakeholders and interests. Otherwise, you just end up in court. So you pretty much have to go through that process.
When the dam was originally constructed, they had essentially 400,000 acre-feet [493 million cubic meters] of flood control space at maximum in midwinter. After the ‘86 flood, when they about lost control of the reservoir, they increased the flood-control space in the reservoir to 600,000 acre-feet. The new water-control manual, it plays in that 400,000 to 600,000 acre-foot range. It doesn’t allow the reservoir to encroach into that 400,000 acre-foot flood-control space in the middle of winter. But it does allow it to modulate in that space between 400,000 and 600,000.
It’s not as aggressive as I would choose. But it’s a foot in the door.
Water Deeply: How would you have preferred to see it done?
Hartman: I have a pretty extreme view, but I’ll tell you anyway. I think rule curves are handy to look at, but I don’t think we should be using them at all. I think what you release from a reservoir today should be a function of everything you know about what the demands are, what the inflow is, what your ability is to release and get rid of the water if something is coming, what the probability of that storm is that may be coming and its magnitude. There’s a way to optimize all of that in a risk-based environment that tells you what you should be releasing in order to optimize the water resources and environmental benefits of the reservoir system.
I would throw the water-control manual out the window and I would replace it with something that was a lot more objective and relied on what the real state of the system is today, rather than having some fixed set of rules that doesn’t necessarily match what’s currently going on.
That is an approach that the Sonoma County Water Agency is experimenting with at Lake Mendocino on the Russian River. They can increase the amount of water that they can store in the reservoir by probably close to 30 percent. So instead of having a wintertime scenario where they store about 65,000 acre-feet, they are able to effectively have that closer to 85,000 acre-feet, with no impact on flood management at all. That’s pretty exciting. That’s just one example they cooked up on their own.
Water Deeply: How important is all this in terms of climate change?
Hartman: It’s probably more important than it was before. We don’t really know what climate change is going to mean in California. What we’ve been seeing, at least for the last couple decades, is increased variability.
People say in a warmer climate we’re going to get more winter rain and less snowpack. And that whole scenario is a disaster for our reservoirs.
If we shift to an environment where all the precipitation came as rain, with no more snow for California, we’d be in a world of hurt. There would be more flooding in the winter and there would be massive water supply shortages in the summer, because the reservoirs would never fill. So it could cause drought.
I think responsible management means taking advantage of all the information and insight available. And I don’t think we’re doing that when we ignore the streamflow forecasts. If we have significant shifts in the timing of the runoff with climate change, this is just one adaptation measure. It won’t in and of itself resolve the problem. But it could help. It definitely won’t hurt.
Water Deeply: What are the downsides of forecast-informed operations?
Hartman: It can potentially increase the risk of flooding. Anything you can predict is going to have an error rate associated with it. Even though we say we’re really good at this, there’s uncertainty. But that may be fully acceptable. It’s a balancing act.
What are you willing to pay to increase your flood risk by 1 percent? If you could pay that 1 percent and instead have 20 percent more reliable water supply, would that be worth it to you? It’s a matter of balancing risks and benefits.
Water Deeply: Does forecast-informed operations put more pressure on forecasters?
Hartman: As far as pressure on the forecasting system, I think it’s welcome. I welcome it. We work hard at producing flow forecasts that are useful and can be put to optimum public benefit. When there are systems put in place that leverage those, we don’t shy away from it. It’s not a problem from my perspective at all.
Top image: Water flows from five of the eight flood gates at Folsom Dam, Folsom, Calif, on Friday, March 18, 2016. The U.S. Bureau of Reclamation, which manages the dam, follows an old rulebook that requires such water releases, even during drought. It might not be necessary if agencies adopted new rules that include modern weather and streamflow forecasting tools. (Rich Pedroncelli, Associated Press)