BAY-DELTA SCIENCE CONFERENCE: Managed Floodplain Fish Food: Bringing Dry Side Food Webs to Wet Side Fish

Two centuries ago, the floor of the Central Valley was primarily a marshy wetland. In the springtime, the snowpack would melt, swelling the rivers beyond their banks and casting the young fish out onto the floodplains. There they would stay for months, fattening up on the abundant zooplankton and invertebrates until the floodplains drained, signaling the time to migrate to the ocean.

As modern day California was developed, the Central Valley’s waterways were re-engineered and channelized to control the floods and divert water for human uses. Today, 95% of the Central Valley’s historical floodplains are cut off from the river by levees, preventing chinook salmon and other native species from accessing the abundant food resources of floodplains.

Floodplain habitat restoration for salmon in California’s Sacramento Valley falls into two strategies: providing access for fish to the foraging and rearing habitat on inundated floodplains (wet-side) and exporting the productivity of inundated floodplains to the river (dry-side). Fish in the rivers are cut off from accessing these habitats, regardless of environmental conditions. So while restoration to increase the frequency and duration of wet-side floodplains is critical for the Sacramento Valley’s salmonid populations, including dry-side managed floodplains in the conservation strategy may also contribute to achieving population goals.

At the 2021 Bay-Delta Science Conference, Jacob Montgomery, Project Manager for Cal Trout’s Central Valley program, presented a potential new management strategy for managing Central Valley chinook salmon.

Central Valley chinook salmon are specially adapted to rear on floodplains. As shown on the slide on the lower left, after just a few weeks, the fish on floodplains are twice as large as the same fish that spent that time in the river.

“One of the main reasons that this happens is because of the aquatic food web that develops on a floodplain,” said Mr. Montgomery. “You can see the invertebrate and zooplankton biomass that is produced on a floodplain is orders of magnitude more dense than that in a river-type ecosystem.”

The map on the slide at the lower left is of the Central Valley in 1873; the shaded region along the Sacramento River corridor were perennial wetlands. The river was unmanaged and flooded every winter, possibly multiple times. In the wet years, floodwaters and floodplains could have extended from the foothills in the Coast Range across the entire valley to the foothills in the Sierra Nevada.

However, now most of the Central Valley’s perennial wetlands have been converted into agriculture, and the rivers managed to stay within the levees.

“We have cities and agriculture that we need to protect from flooding, but it makes it hard for a fish that’s adapted to rear on floodplains to find any,” said Mr. Montgomery.

Cal Trout has developed several programs to reconnect salmon populations with floodplains. For the last decade, Cal Trout’s wet-side programs have been working in the Yolo and Sutter Bypasses, which still flood occasionally and are accessible to fish. The new programs are called ‘dry side’ programs, which are conservation strategies to include all the flooded acreage existing outside the levee boundaries that still active floodplains but that fish don’t have access to.

Wet side programs

One of the well-known studies on floodplains in the Central Valley is the Nigiri Project, which began with the basic idea of getting fish onto flooded bypass habitats. When activated by floodwaters, the flooded bypass habitats act very similarly to historical floodplains, and fish achieve the growth benefits from the aquatic food web that develops there.

“We’re working on increasing both the frequency of flooding and the duration of flooding in all but the wettest years in the flood bypass systems,” said Mr. Montgomery. “When we’re in an extreme flood, we don’t need to change this system at all; we need this to function as a flood control mechanism. But in any other scenario, a small flood will drain out of this bypass very quickly, and the length of time that the floodplain habitat exists is very short. So we want to, in less extreme scenarios, increase the number of floods and the duration that those floods last for, so that fish that come in with those floodwaters can have access to floodplain habitat for a longer time.”

Dry side programs

Outside of the Sacramento River levee and the flood bypass system, there are about 500,000 acres of functional floodplain habitat that exists in wildlife refuges, waterfowl hunting clubs, winter flooded rice fields for rice decomposition, and wading bird and shorebird conservation programs, All of these flooded habitats create a dense zooplankton food web just as the natural floodplains would have. So the dry side programs are about managing those floodwaters and exporting them back to the river where fish can get some of that benefit from the food web.

“So on the wet side, we’re bringing fish onto floodplains, and we’re doing that with the Nigiri project,” said Mr. Montgomery. “On the dry side, we’re bringing floodplains back to the fish. We call that fish food.”

Implementing the dry side program

Cal Trout got the first chance to test the dry side program at scale a couple of years ago with Reclamation District 108 in northern Yolo County, just north of Knight’s Landing in Northern California. The Reclamation District, a local agency that manages water infrastructure, has all the infrastructure needed to irrigate agriculture, provide water supply, and drainage for flood control. All this infrastructure is necessary to manage floodplain water and export food webs back to the Sacramento River.

In 2019, just under 5500 acres participated in the pilot program. Rather than draining the fields immediately when done with the flooding, the farmers were asked to drain their fields such that a prolonged pulse of floodplain water was released over four weeks through the canal system and out to the river. The yellow star on the map shows RD 108’s main flood control pumping station.

Cal Trout monitored the river ecosystem to see how it responded with four monitoring stations:

a monitoring station upstream of the pumping station to assess the baseline conditions in the Sacramento River;
a monitoring station inside the floodplain canal before water is exported to the river so they would know what was being exported and the potential for maximum food web density and fish growth;
a monitoring station at the outlet where the floodplain water is mixing with the Sacramento River; and
a monitoring station a mile downstream to see any signal from this action could be detected further downstream.

Every week, the zooplankton community assembly, the abundance and fish growth with individually pit-tagged juvenile Chinook salmon from the Feather River Hatchery in enclosures. Each individual fish was identified weekly, their fork length and body mass measured, and their condition and growth rate assessed throughout the management action.

The results

The chart shows the data from the floodplain canal during the period of active management of the floodplain drainage. The colored area in the background is zooplankton density, and the white box and whisker plots are the fish growth.

“There’s a big spike in zooplankton density in the canal when we’re actively moving that water through it,” said Mr. Montgomery. “The guilds of zooplankton change a little bit from maybe a large portion of copepods to a higher proportion of large and small cladocerans during the drainage. Then the zooplankton abundance drops off pretty sharply after we stop managing for this floodplain export.”

“There was positive fish growth and increasingly positive fish growth every week that the fish were exposed to these conditions,” he added. “So basically, the more we feed them out here, the faster they grow.”

The charts show the data from the upstream baseline station and the floodplain outlet station where the floodplain food web subsidy initially mixes with the river. With zooplankton, the upstream baseline conditions are meager, and few food resources are available for fish.

At the floodplain outlet, there are occasional pulses as the drainage water comes off the floodplain.

“During the time when we’re actively managing for this floodplain export, a much higher concentration of zooplankton food is getting added to the river,” said Mr. Montgomery. “It’s those floodplain-specific bugs, those large and small cladocerans that are dominating the assemblage when we are exporting the floodplain water to the river. And likewise, the density of food added to the river drops off pretty quickly after we stop draining floodplain water into it.”

Fish growth pretty much follows the zooplankton pattern. The baseline conditions upstream for fish are pretty low across the board; there are even negative growth rates, which means they’re not consuming enough calories to compensate for the calories; they are just swimming in a holding station at that location in the river. He noted that those fish aren’t caged out in the middle of the river but rather in edge habitats, backwater eddys, and other good quality fish habitats.

“These should be habitats where a fish can make a living, but they’re not able to do that at those baseline food web conditions upstream of our floodplain outlet,” said Mr. Montgomery.

The pattern at the outlet is similar to the canal; there are positive growth rates that increase as they are fed more of the floodplain zooplankton.

“This is a really positive result,” he said. “It seems to indicate that the simple concept adding food from off-channel, managed floodplains back to the river can help subsidize food webs and growth rates for fish in the Sacramento River.”

He noted that at the beginning of the study in mid-February 2019, all these fish were about the same size and looked pretty lean. “By the end, those fish at that baseline upstream river station, they’ve grown a little bit, but they still have that kind of lean look to them. Whereas fish at the outlet and even fish a mile downstream, they’re bigger, they’re thicker, and their body condition is more robust. So these fish just look much healthier. And that’s after only four weeks of food web subsidies from this managed floodplain.”

“At the floodplain outlet, we have fish growing on average five times faster than the fish that were growing upstream. And even a mile downstream of the outlet, we’re still seeing fish growing three to four times faster than the baseline river fish upstream of these food webs subsidies.”

Conclusions and implications

So what does this mean for the Sacramento Valley? “We definitely have another tool that we can add to our conservation and management portfolio, and that is exporting fish food from managed floodplains on the dry side of the river and bypass levees back to the Sacramento River,” said Mr. Montgomery. “Fish actually do get benefits from that.”

He also noted that 2019 was a wet year, and typically a wet year is a pretty good year to be a fish in the Sacramento Valley.

“Those bypasses typically are activated for long periods, so there’s a lot of floodplain habitat available,” he said. “Survival and growth are pretty good if those fish can find the floodplain. But that’s not always the case. For example, this year is a very dry year, and there hasn’t been a single flood event in any of the bypasses. And so there’s virtually been zero floodplain habitat available for fish in the river system.”

The study continues in 2021

Cal Trout is running the project this year with 7000 acres under management, hoping to put a bigger signal in the river and have it last a little bit longer. They are also monitoring even further downstream to see if they can determine how far the reach is, so there is a monitoring station six and a half miles downstream from the floodplain outlet, in addition to periodically from the outlet down to that six and a half-mile mark.

They are also looking to see if, rather than one pulse of floodplain water, if it’s possible to get multiple flood, drain, and reflood cycles out of a single piece of acreage. So 2000 acres were drained in mid-January, reflooded, and drained again with the rest of the 7000 acres in mid-February to mid-March.

UPDATE: This presentation was prepared in March of 2021 before the study concluded for the season. Mr. Montgomery sent me the following update:

“We tracked the foodweb subsidy much further downstream in 2021 – 6 miles, as opposed to 1 mile in 2019. The simple story we learned in 2021, relative to 2019, was that the baseline river growth is much lower in dry years (30% lower by our data), which makes a much greater relative impact of a river reach with a subsidized foodweb for the salmon population. Also, the subsidized food web shows positive effects on the fish growth rate at least up to 6 miles downstream. So, this is not just a localized effect but a reach-scale effect. We do intend to continue running this Fish Food program in 2022 as well.”

“We think that this program has a much greater potential to have a positive effect on the salmon population in dry years, particularly because those floodplain habitats in the bypasses are not available in dry years.”