Farms as floodplains: A summary after five years of experimental flooding of agricultural land
Studies show the rice-field fish are larger, healthier and more robust than those in the river at the same age
One of the recent and notable success stories for California has been the ongoing studies in the Yolo Bypass to use flooded rice fields as habitat for young migrating salmon. The results from the five years of study show the rice-field fish are larger, healthier and more robust than those in the river at the same age.
Lynn Takata began with the historical context of the Delta. In the mid-1800s, settlers came to California for the Gold Rush and settled in the Delta area. “The settlers came to the quick realization that the Delta is an area of pretty frequent and sometimes catastrophic flooding,” she said, noting the picture on the slide showing an artist’s rendition of the 1861-2 megaflood, a rain event that lasted for almost 43 consecutive days. “With that realization, settlers started an era of dam construction and levee construction, and really a whole infrastructure of flood control in the Delta to protect developments and also to protect farms from the tides and the floods that came with the landscape.”
Because of a lot of that levee construction and dam construction, the floodplains and wetlands that were really characteristic of the Delta really disappeared from the landscape, Ms. Takata said, presenting a map from the San Francisco Estuary Institute’s historic ecology study.
“On the left hand side, those green and blue colors are the historic Delta’s wetlands and floodplains that used to cover a large portion of the area, and the map on the right, the pink color is agricultural land and weedy vegetation that has now replaced all those floodplains and wetlands,” she said. “So in comparison to the Delta before human habitation, the landscape has really transformed dramatically to really one of agriculture from one that used to be mostly floodplain and wetlands.”
“So while there are a lot of stressors in the modern Delta for native fishes, it’s believed that that loss of floodplain habitat and wetland habitat is amongst the major reasons that native species, particularly species like Delta smelt, longfin smelt and salmon are shriveling in the Delta, but on the flip side of the coin, it’s also been that the rehabilitation and the restoration of wetlands and the reconnection of floodplains is looked to as maybe one of the ways that we might be able to support these species into the future.”
The Yolo Bypass is one of the areas of particular interest because it’s one of the largest floodplains on the west coast; when inundated, it doubles the wetted surface area of the Delta, providing a lot of habitat for native fish species and waterbirds, she said.
The Yolo Bypass is part of the flood control system that was implemented in the early 20th century to control floods, but it functions quite differently than levees and dams, she said. “It acts as something of a pressure relief valve for the Sacramento River, allowing water to flow down into it for flood control,” she said. “The way that it works is the northern end of the bypass is the Fremont Weir, a low concrete two-mile long sill, and when the Sacramento River reaches 33 1/3 feet in stage, water begins spilling over that sill into the Yolo Bypass. When overtopping occurs in the Yolo Bypass, anything that’s transiting through the Sacramento River, like juvenile salmon, has the opportunity to enter the floodplain and take advantage of the habitats that are out there.”
Just as it is engineered to flood, the Yolo Bypass is also engineered to drain relatively quickly and efficiently, Ms. Takata noted. “The whole bypass is graded to the southeast, so as water comes over Fremont Weir, it then drains to the southeast to a perennial canal on the eastern side called the Toe Drain,” she said. “The Toe Drain drains the Yolo Bypass, channels the water down south, joins up with the Deep Water Ship Channel, and then eventually joins back up with the Sacramento River around Rio Vista.”
The Yolo Bypass also has four tributaries that enter from the western side, which irrigate agricultural lands within the Yolo Bypass during the growing season. These tributaries can cause some localized flooding, but this does not bring an influx of salmon in as the there is only one small population of salmon in one of the tributaries, Putah Creek.
“Based on studies that began in the late 1990s by Dr. Ted Sommer and others, we know that juvenile salmon that can gain access to the Yolo Bypass really benefit as far as growth goes and survival goes,” she said. “We also know from studies since then that the Yolo Bypass during overtopping is a really productive system. There is a lot of lower trophic productivity, a lot of phytoplankton production, and invertebrate production that can then get transported down to the lower estuary subsequently as it drains.”
A large portion of the Yolo Bypass is used for agriculture during the growing season, and so one of the big impetuses for the agricultural floodplain project is how to utilize or leverage the existing infrastructure and land uses, which is mostly agriculture in the Yolo Bypass, to benefit juvenile salmon, she said.
Ms. Takata said she would be discussing where the project has gone over the last 5 years. “It has been an adaptive research project where we modify our goals and questions for subsequent years based on what we learned during the current year, based on opportunities that come up with different landowners who want to participate, but also what we anticipate will happen with Mother Nature and how much water she might provide in the following year, and so we have been very dynamic in that way in trying to plan our research.”
The past five years in brief: The first year, 2012, was a pilot year where they tried maintaining juvenile salmon on an artificially irrigated rice field. In 2013 and 2014, they worked on questions about agricultural substrate and depth heterogeneity and did those things make a difference for salmon. In 2015, they were able to expand to some agricultural floodplains outside of the Yolo Bypass. In 2016, they were able to try the ‘floodplain extension potential management strategy.’
Ms. Takata then went through each year, giving more details.
Year 1: 2012
In the first year, the goal of the study was to see if it was possible to put juvenile salmon out on a field that is irrigated artificially and will they survive. The first landowner participant was the Knagg’s Ranch property, which is irrigated from Knight’s Landing Ridge Cut located in the northern part of the property.
“We placed hatchery salmon on a small 5-acre field in the southeastern corner of the property with a variety of depths and a variety of substrate or vegetation types, and we found that we could maintain juvenile salmon out on that field, and they actually grew really well,” she said. “Growth rates were about .7 mm per day, which is amongst the highest growth rates in the region and also on par with growth rates that Ted Sommer had observed in the late 90s in the Yolo Bypass floodplain. Despite a berm breach where a number of our fish escaped, we recovered almost 60% that we actually stocked in that field.”
Year 2: 2013
In 2013, the studies focused on questions of agricultural substrate and if that made a difference to the salmon. For the study, they examined three fairly common substrate types commonly found in rice fields: “One is rice stubble, which is when the rice crop for the year is chopped to a uniform length and left; one was stomped, which is when rice is cut really close to the ground and pressed into the mud, giving a muddy substrate; and one was fallow, which is a field that hasn’t been in cultivation during the growing season and weedy vegetation is allowed to grow,” she said.
So they used nine 2-acre fields on the Knagg’s Ranch properties, with each of the fields having one of the three agricultural substrates, and then they put hatchery salmon on the fields to see how they did. “We found that growth rates were really, really high this time, .93 mm per day, that’s some of the highest growth rates reported in the Central Valley for salmon, but we found that survival was pretty low that year,” she said. “It ranged from 0 and 10% depending on the field, and that’s fairly low, it’s not really what we had expected. What we did find though was neither growth or survival was related to the substrate treatment.”
Ms. Takata offered a possible explanation for the low survival rate. “We conducted the project in March and April, which turned out to be fairly warm and it was fairly late; it was after most of the surrounding landowners had drained their fields in preparation for the growing season, and so our little experimental set up there was the only patch of water in the Bypass at the end of our experiment,” she acknowledged. “It attracted a lot of fish eating birds. Lesson learned. And that’s why we think the survival was very low and not related to our substrate types.”
They also wanted to know if juvenile salmon were given the opportunity, would they choose certain vegetation types over others, so they implanted fish using very small PIT tags (or passive integrator technology). They then created two circular enclosures that had a variety of habitat types to test would they prefer to be in certain parts of the field over others; one of those circular enclosures had all three of those habitat types, the stomped, the fallow, and the stubble, and then the other enclosure which was kind of a control, had nothing but the stomped, muddy vegetation. They then put PIT tag implanted salmon on those fields, and put PIT tag antennas in those different habitats to see where the salmon would chose to spend their time.
“We found that they really didn’t seem to care what the vegetation type was, but rather, they wanted to be closer to where the water was coming in,” said Ms. Takata, noting that there is a paper released earlier this year, written by Louise Conrad, that has more details on the work with the agricultural substrate.
Year 3: 2014
Given what they had learned about low survival and bird predation, and other studies that have suggested that fish tend to go to deeper areas to try and get away from predation, as well as some anecdotal observations of juvenile salmon from their fields, in 2014 they wanted to know whether depth and proximity to inflow made a difference for juvenile salmon.
“Again, we were faced with a dry year, so these were hatchery salmon, because wild salmon couldn’t access our fields and we had to irrigate the fields artificially,” she said. “Again, we utilized those 2-acre fields at the Knagg’s Ranch property and we treated each field to a different depth treatment; some of the fields had no trench, some fields had an 18 inch trench, and some of the fields had a 36 inch trench, and we put hatchery salmon out on those fields.”
“For the preference side, seeing where salmon want to go if given the choice, we created a field that had various depths with various proximity with relation to inflow,” she said. “We used what’s called a factorial design, so that every depth is represented at every proximity with relation to inflow, and when you use that design, you can basically look at two variables at the same time. So in this case, we looked at both the depth variable and the proximity to inflow variable at the same time using this.”
“Again we used our PIT tag technology and implanted juvenile salmon with the PIT tags and put antennas out in the various depths, and let them do what they wanted, and we collected the data at the end to see where they preferred to spend their time,” she said. “In our 2-acre replicated fields, we found that survival is a little bit better this year. We got smarter, we did the experiment a little bit earlier when surrounding fields were also filled with water, and we found that survival was anywhere from 22% to 64%, depending on the field, but again, it wasn’t related to the depth treatment.”
“In our habitat preference experiment, we found that during the daytime, salmon preferred to be in the deeper panels, but at night they seemed willing to go out to the shallower areas and preferred the shallower areas,” she said. “No matter what time of day, we found that salmon preferred to be closer to inflow again, confirming what we had anecdotally observed the year before.”
Year 4: 2015
2015 was again predicted to be a dry year, and this time they had interest from other landowners in participating, so they were able to expand beyond the Yolo Bypass to study other agricultural floodplains outside of the Yolo Bypass.
“We were able to get a rice field out in the Sutter Bypass that was interested in participating, and we had a new landowner in the Yolo Bypass that wanted to participate, and interesting thing about this landowner, Conaway Ranch, was that they had two different field types that were right next to each other. One was rice, and one was a conservation easement, which had some really heavy vegetation and trees growing on it. The third property they used for the experiment on was the Dos Rios Ranch on the San Joaquin River.”
All of these sites were fed with different water sources, so for that year, they looked at how the food web develops in the different geographic locations in Delta. “Is it similar, or is it different? What do the fish eat in these places? Do they eat different things depending on where they are and what the water source is? What about growth, does it differ between these places?”
They put enclosures in each of the areas, and put salmon in the enclosures, and monitored them for a four-week period. “We found that no matter where the field was in the floodplains, inside or outside the Delta, invertebrate densities were pretty high,” she said, noting that invertebrates are the fish food that develops on the field. “What was also really interesting is that those two fields on the same property that were right next to each other had very different invertebrate densities at the fourth week of the study; that’s the Conaway rice field and the Conaway easement field. The Dos Rios Ranch down on the San Joaquin had comparatively high densities similar to the Conaway easement field.”
Even though the densities were high, the composition of the invertebrates was really different, Ms. Takata said, presenting color-coded pie charts for each of the fields to illustrate the differences in infield composition. “What’s interesting is what the salmon chose to eat,” she said. “While the proportions are different, what they preferred seemed to be similar across the different geographies. Cladocerans and dipterans is what they sought out, and again as we had observed in other years, growth rates were really high and fairly similar across all these different geographies and field types.”
Year 5: 2016
With 2016 being a potentially wet year, they were exited to test what they called flood extension potential management strategy. “All along, while we have had to use hatchery fish, our target was to benefit wild salmon, and so if we could get an overtopping event at Fremont Weir that would bring in wild salmon, we were hoping to get the opportunity to test extending the duration of a flood event using infrastructure that exists on an agricultural field,” Ms. Takata said. “The way the Yolo Bypass floodplain is engineered, it’s engineered to flood, but it’s also engineered to drain relatively quickly, so once overtopping stops, water doesn’t stay in the floodplain very long. We still wanted to get a handle on what survival was, and since we wouldn’t know how many wild fish came on to our fields during an overtopping event, we also wanted to stock hatchery salmon on our flood extension fields just to get a sense of what their survival was like, both when they left the field, and also perhaps when they entered the ocean fishery.”
“We also wanted to know how the food web would develop in this very different scenario where presumably it’s a wet year, the water is going to be cooler, and the Sacramento River is going to be running a little faster, and so we were curious to know would this be different than all our prior years when we had to irrigate artificially.”
Knagg’s Ranch and Conaway Ranch would be participating, as well as Swanston Ranch, a new landowner in the middle portion of the Yolo Bypass.
There was overtopping at Fremont Weir, but it occurred later than they had hoped. “It was good for the ecosystem, but not good for our experiment,” Ms. Takata said. “The hatchery wasn’t able to hold the hatchery fish for us as long as we needed, and so we weren’t able to plant the hatchery salmon in our flood extension fields, and perhaps it’s just as well. As you get later in the season, the weather gets warmer, the fields can warm up pretty quickly; there was at least one field that got warmer than we had liked, and the dissolved oxygen dropped below what we would have liked, so it possibly was just as well that we couldn’t plant hatchery salmon out there.”
“We did find that wild salmon did make it on to our fields, and in a large range of sizes,” she said. “Although they were all genetically fall and late-fall, they came in at a range of sizes that would falsely suggest they were of all the runs of salmon that exist in the Central Valley. We also found that a lot of natives made their way onto our fields, particularly splittail and blackfish.”
In summary …
After 5 years of research, what have they learned? “Growth is really high on the floodplain, and that seems to be universal, no matter what the year is and no matter what the geography is,” Ms. Takata said. “Also, zooplankton production and invertebrate production (fish food) is really high, but the composition can differ, depending on where it’s located in the bypass or in the floodplain. It can also depend on what field type you have, what was grown the prior season or what wasn’t grown the prior season.”
“As far as fish are concerned, for survival and for where they prefer to spend their time, substrate doesn’t really seem to make a difference to them, but they do seem to prefer deeper areas, particularly during the daytime, and the shallower areas at night. In this sort of floodplain environment, you have to imagine that the agricultural fields, particularly when they are irrigated artificially, are more akin to ponds then a river or a lake, and so juvenile seem to really be attracted by that inflow. Timing is really important. Don’t go too late in the year when everybody else is drained all their fields, and that you can really run into some water quality issues if the timing is a little bit late.”
“Most importantly, we’ve learned that how to work an adaptive research project that can be really dynamic depending upon what comes, what the landowner participation is like, and what the water year is going to look like,” she said. “We’ve also learned the adaptive research that we conducted can be really complicated, especially when Mother Nature doesn’t’ really go the way you want; she can be really dynamic, she can be really sarcastic, and what we’ve learned about how landowners manage and grow on their land is likewise very dynamic and very complex, so just implementing our little project here required a lot of communication and frequent communication. In the future, should any of these management scenarios be implemented at a large scale in the Yolo Bypass and in agricultural floodplains, it’s going to require some really tight, open, and frequent communication with landowners to get it to work right.”
Looking forward …
Looking forward, they still are interested in trying out the flood extension management scenario so we can see how extending the duration of a flood event might benefit salmon.
Ms. Takata noted that their section at DWR is doing other work in the Yolo Bypass on salmon and on other fish; they are also using telemetry studies for juvenile salmon survival, and also to see how adult salmon stray. “We do get salmon coming up, looking for the Sacramento River because that’s basically the water coming through the Toe Drain when it’s dry. Where do they go, how far do they go before the turn around and how far do they need to go when they won’t turn around.”
They are also looking the lower trophic production in the Yolo Bypass floodplain gets transported downstream and how it develops through the flood cycle. “We’re hoping to use otolith microchemistry, which is basically using the earbone in a fish to figure out where that fish might have reared to see if we can get a Yolo Bypass signal in adults that return to spawn to see did they rear in the Yolo Bypass when they were young or did they rear elsewhere.”
“And that’s it … “
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!