METROPOLITAN BAY-DELTA COMMITTEE: Salmon predation in the Delta

Study results suggest removing predators won’t address the problem, but removing submerged aquatic vegetation might

At the May meeting of Metropolitan’s Special Committee on the Bay-Delta, Alison Collins, Senior Resource Specialist for Metropolitan, presented the results of a recent study looking at the effects of predation on juvenile salmon.  The study was funded by Metropolitan Water District and the Department of Fish and Wildlife; other partners included Cramer Fish Sciences and ICS.

The predation study is part of a larger effort to understand the stressors that are negatively impacting salmon.  “Conceptually, we think there are a lot of stressors that are negatively impacting fish, and our job is to identify what these stressors are, understand the magnitude of the impacts on the fisheries, and then identify management actions that can be implemented to try and modify these stressors,” said Ms. Collins.

The real challenge is to identify management actions that we can have control over that will result in changes or modifications to the stressors.  “All these stressors might be negatively impacting salmon in some sort of way, but it doesn’t mean we can have control over all those different management actions.  For example when we think about predation, it’s actually really difficult to control predator populations.  But maybe we can modify the habitat through which salmon and predators interact in a way that gives salmon some sort of an advantage.”

Ms. Collins noted that there was a request from a recent Metropolitan meeting for information that identifies, categorizes, and ranks all the stressors as manageable or unmanageable, but unfortunately, no such list exists as of today.  She noted it would require a lot of studies and work to inform such a list, but the studies she will be describing today are examples of how they start to gain information on what management actions would be feasible.  As management actions are implemented and monitored, they will be able to evaluate the success of those management actions and create a list of actions that are feasible and could be successful.

She noted that they are actively working on studying all the stressors, not just predation, through collaborative partnerships.  “Through these collaborative partnerships, we’re understanding these stressors and we’re able to implement more effective management to help restore and recover salmon populations,” she said.  “The more that we can recover and build resilient salmon populations, the more flexibility we’re going to have with water project operations, so that’s the ultimate goal here is to understand this water delivery system in combination with the ecosystem needs that we have out there.”


The California Delta has been highly modified.  There has been extensive channel modification with more structures near and in the water, multiple invasions of non-native species including different types of predators, as well as invasive aquatic vegetation that fills in habitats that fish used to use with thick, dense strands of vegetation.  These landscape scale changes have really changed the way that salmon and the predators use the habitat and interact within the habitat.

One of the questions they sought to answer was, how do landscape modifications like artificial structures and vegetation influence predation on juvenile salmon?  So they had two study questions.  The first was, do the number of predators influence juvenile chinook salmon survival?  Ms. Collins noted that in the study, they observed that predator density did not change the survivorship of juvenile chinook salmon.  The second question was does habitat type modify juvenile chinook salmon survival?  She said they did observe that submerged aquatic vegetation, or SAV, did decrease salmon survival.


Ms. Collins then described how they arrived at those two conclusions.  For the study, they  built three large enclosures with reinforced PVC pipe that were fully enclosed by mesh.   These enclosures allowed researchers to control how many predators were in each enclosure.  The mesh around the enclosures was small enough so predators stayed in but large enough that juvenile salmon could swim through these enclosures.

The salmon were tagged with PIT or passive integrated tag, which is essentially a microchip; when the fish swam out of the enclosure, the PIT antennas would read the tag.  The tag data provided information on how long fish spent in the enclosure, which direction fish swam out of the enclosure, and whether they survived or were eaten by a predator.

They conducted the study in 2017 in Little Holland Tract, and in 2018 on Bouldin Island and an old duck club.

To answer the question, does predator density affect salmon survival, the enclosures allowed for different numbers of predators to be put into each enclosure; they had a low, medium, and high predator density treatment.  They used largemouth bass as the predator for this experiment which would be put into these enclosures along with an equal number of tagged juvenile salmon; then they would measure how many survived and how many were eaten.

To answer the question does habitat affect salmon survival, for this experiment the number of predators was the same in each enclosure, so they were only manipulating the habitat.  The three habitats they used for the study were submerged aquatic vegetation or SAV, a simulated dock habitat, and a no-habitat treatment.

For the simulated dock habitat, they had four pilings that went all the way underwater and a sheet of plywood that acted as a dock.  For the submerged aquatic vegetation, they used a fake garland which allowed them to standardize the amount of habitat so that it was the same in both the dock and the submerged aquatic vegetation enclosures.


For the first question, they found that predator density does not affect salmon survival.  Ms. Collins presented a graphic which showed salmon survival in the low, medium, and high predator density treatments and pointed out the proportion of fish that survived is very similar among the treatments.

So we’re seeing that the number of predators did not change how many salmon were eaten,” she said.  “Remember we are focusing on this work because we want to understand how to better manage stressors that are negatively influencing salmon.  There’s been a lot of talk about efforts in the Delta to remove predators or reduce the number of predators, but this research suggests that it doesn’t really matter how many predators you have; it’s not changing the survival of salmon.  So this suggests that this might not be a management action that we want to invest a lot of resources in.”

Ms. Collins did give a few caveats.  The first is that they used largemouth bass for this experiment, but the results might not transfer to other predator species that are out there, as other predator species have different hunting strategies and behavioral requirements.  The second is that they used the enclosures for the study which is the only true way to manipulate predator density, but the enclosures could be having some sort of effect that they are not able to observe.

She also noted that there have been other research efforts in the Delta that have actually removed predators and looked at salmon survival and the results were similar.  “These efforts and ours are saying that reducing the number of predators does not increase salmon survival.  There are a few exceptions at very particular hotspot locations where we have human made structures that are aggregating predator and prey species, but in general as a whole, we’re not observing some sort of detectable effect here.”

Turning to the results of the habitat experiment, they found that submerged aquatic vegetation or SAV did decrease salmon survival.  She presented a graphic of the results, noting that the survival in the no habitat and dock habitat treatment is significantly higher than in the submerged vegetation treatment.

The proportion of fish that survived in the submerged aquatic vegetation was 66% compared to 94-95% in the other treatments.  So this is suggesting that removal of submerged aquatic vegetation might be a feasible management action that can be implemented to increase salmon survival.”

Ms. Collins then reviewed the results of this study.  “We observed that changes in predator density did not affect salmon survival, so this suggests that efforts to reduce the number of largemouth bass may have limited success to increase salmon survival.  However, we did observe that submerged aquatic vegetation decreased salmon survival, so there’s some potential that this management action removal of this type of vegetation could really help to increase salmon survival.  We’re currently developing a manuscript that summarizes the study that we did and the results which is being submitted to a scientific journal this summer that can go through the peer review process.”

Ms. Collins noted that there is a lot of submerged aquatic vegetation in the Delta.  She presented a slide showing submerged aquatic vegetation in Franks Tract in the years 2004, 2015, and 2018.

What we’re seeing is that the amount of submerged aquatic vegetation is growing through time.  In fact in the Delta as a whole, we’ve seen a large expansion of submerged aquatic vegetation and floating aquatic vegetation, and there are some areas in the Delta where submerged aquatic vegetation has tripled just in the last three years.  So there’s a lot of this stuff to remove, and that’s why there’s potential that this management action could have a really big effect.”


So the researchers next wanted to know if there are other habitat features out there that could be addressed or if there were ways to modify habitats where predators and prey interact that could benefit salmon, and they wanted to understand what is it about those locations or the features around those locations that increase the vulnerability of a juvenile salmon being eaten.

So in 2019, they are looking at what habitat features are out there that can be addressed, so the general question for the study is can predator contact points be modified to reduce predation?  Ms. Collins explained that a contact point is a physical or hydrological feature that’s a result of human alteration to the riverscape, things such as diversion pipes and pilings, artificial lights, scour holes, and current breaks.

We think that these contact points are increasing the chance that juvenile salmon are eaten, but there is still a lot of scientific uncertainty around what these contact points are, why they are increasing predation on salmon, and if they can be modified in a way that actually increases the survival of fish that are passing by these structures,” she said.

To answer these questions, they used predation event recorders (or PERS) which is a three-foot PVC pipe that floats down the river; it is equipped with a GPS and a GoPro camera, and on the bottom, a salmon is tethered to the PVC pipe.  When the juvenile salmon is eaten by a predator, the fish is pulled away from the pipe which pulls out a magnet and that records the location and time of that predation event.

By using this sampling methodology, you can get where and when predation occurred on the landscape, you can identify the predator species because it’s being filmed, and you can get information on the environmental covariates that surrounded that predation event,” Ms. Collins said.  “All of these data give us insight into where and when predation is occurring and the conditions that surrounded that predation event.  From that, we can start to learn and better understand how predation is happening.  So by floating 100s of these PERs down a river reach, you can start to identify when and where predation is occurring and what kind of contact point it is associated with.”

Ms. Collins then presented a map of the Lower San Joaquin that was developed by UC Santa Cruz and NOAA Fisheries. The warmer colors indicate areas where the predation rates were higher.

The arrow is a little off but you can see that there are these dark red lines at the top and the middle of the slide that are showing high predation rates around two diversion pipes, and then there are two scour holes that are present here, so there are higher predation rates around the scour holes,” she said.

Once we know what these contact points are, we can think of ways to modify these contact points in an effort to try and increase salmon survival,” she continued.  “For example, if there was a diversion pipe, we could try to screen the diversion pipe or we could try to modify the habitat that surrounds that diversion pipe in a way that provides salmon some cover, or we could put a deflector on the pipe to try and minimize the flow velocity separation of that pipe.  We can and should be creative in the types of modifications that we’re coming up with.”

The goal is to identify management actions that can be easily implemented on a landscape level and can have a large effect.  “For example, if we found that artificial lights increase predation rates on salmon, we could turn off lights during the period of time that salmon are migrating through the Delta, or you could change the lightbulb so you have a different UV spectrum so it’s not shining so much, or you could point the lights up to the sky instead of facing the water,” she said.  “There’s a lot of things that you could do and that management action would have a big effect because there’s a lot of lights in the system.  This is the first year of the study so I don’t know if lights are a problem, I just wanted to give you an example for context.  And I don’t know if you can modify lights in a way that will increase salmon survival, but those are the types of management actions that we’re trying to identify through this type of research.”

This is the first year of a 5-year project.  There are several funding partners; Metropolitan is one of them.   Ms. Collins noted that they are already seeing some interesting results; there have been really high flows and really cool water temperatures and they have observed that predation rates are really low when the water temperature is really cool.

We think that’s because the energy requirements of the predators are less,” she said.  “That’s just observational at this point but we haven’t had a ton of predation occurring right now this year.  We have a couple more weeks of this project; we started to see an increase in predation and then we had a rain event that happened mid-May.  Water temperature was cool again and we had a bump in flows, so it’s going to be interesting to see how predation changes.

Ms. Collins noted that out of possible management actions for things that can be controlled, water temperature in the Delta is a really difficult one. “If you’re trying to cool down a section of river in an upper tributary where we have a dam, you can use water project operations to help manipulate water temperatures,” she said.  “In the Delta, it’s more difficult.  It’s such a big place and there’s lots of efforts that are going to help with water temperature through habitat restoration and tidal marsh restoration.  But it might be in some years when it’s really hot that the modifications we’re doing aren’t going to be successful, because water temperature is a master factor that we can’t control.”

As part of the study, they are working with partners to ensure the results of the study are used in the model that informs and prioritizes restoration actions as part of the Central Valley Project Improvement Act, or CVPIA.  The CVPIA was passed by Congress in 1992, and it mandates changes in the management of the CVP, particularly for the protection, restoration, and enhancement of fish and wildlife species.   The CVPIA program is benefitting salmon, so whether it’s the CVP or the SWP, it’s an important program and there are a lot of restoration dollars available, she said.

The studies I’ve shown you today are really trying to get at understanding that realm of management actions and the control that we can have over things,” said Ms. Collins.  “This work is part of a larger effort to understand the stressors that are impacting fish and how we can manage those stressors more effectively so we can help to recover our salmon and build resilient populations.  If we can achieve that, there will be more flexibility in our water supply.”


During the discussion period, many of the directors were astonished that the number of predators didn’t affect survival rate, calling it counterintuitive at best.

Ms. Collins said there a lot of potential reasons.  “One could be the amount of time that fish spent in the enclosure, maybe they are just getting out fast.  We did in our experiment see fish spending different amounts of time in the habitats, so some left very quickly and some stayed for a long time.  It could be that even though you have less predators, you still have maximum hunting opportunities, so the less that there are, there’s only a few predators but there are lots of salmon so there’s more for you to eat, that could be part of the reason … the other studies that I’ve looked at of predator removals, part of the issue is that when you take predators out, there are so many in the Delta.  One study found that when you removed predators, it just tripled the number of predators that came back the next day.  So there’s so many predators out there, it’s part of the reason why we think we are observing these results.”

So the bill to decrease the number of striped bass by having no limit on fishing for striped bass, that that really wouldn’t have had any effect on it.  Is that what you’re saying?,” asks Director Glen Peterson.

Ms. Collins noted that the study was done using largemouth bass.  “We should be careful when we interpret that on to a different species, but other studies that have removed predators – they’ve removed striped bass, channel catfish, white catfish, and largemouth bass – and they didn’t see an increase in salmon survival.”

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!

Leave a Reply

%d bloggers like this: