NOAA Fisheries Rachel Johnson opens salmon symposium with a look at the outmigration and life history diversity strategies that have allowed salmon to evolve and survive on California’s highly variable landscape
Warming temperatures, changing hydrographs, drought, and other impacts of our changing climate present challenges to the survival of salmon in California. Is there anything that can be done to conserve these iconic species? The answer, perhaps surprisingly, is yes. Earlier this fall, the UC Davis Coastal and Marine Sciences Institute Center for Coastal Ocean Issues and the Delta Science Program convened the California Salmon and Climate Variability Symposium to explore how variable and changing ocean and hydrologic conditions affect Central Valley salmon and their management. Scientists shared their research focused on topics such as hatchery practices, predation, floodplain restoration, ocean conditions, and management practices that can help salmon survive on California’s changing landscape.
Starting off the conference as the keynote speaker was Dr. Rachel Johnson, Research Fisheries Biologist with NOAA Fisheries, Fisheries Ecology Division specializing in applied ecology of freshwater and anadromous fishes. In her presentation, she discussed some of the strategies that salmon have employed to adapt to California’s highly variable climate.
Dr. Rachel Johnson began by recalling that in 2007, very few salmon returned to California’s rivers, and it was devastating for a lot of communities. “It was also devastating to the scientific community, and it was devastating to us because we were surprised at how unresilient this workhorse stock of the fall run really was,” she said. “They were not as predictable or as knowable as we thought that they were.”
There was a lot of finger pointing. “All of a sudden, no fish return and the question was, who was to blame? Is it that all the salmon got exported down to LA with the water? Did the commercial fisherman take too many? What caused this collapse? So Steve Lindley and a cast of many looked at the data available and concluded that the proximate cause for this collapse was poor oceanographic conditions when the juveniles in 2005 entered the ocean.”
“The report goes on to say that it’s death by a thousand cuts,” she said. “The stock is not resilient. One of my favorite quotes at that time is one of Peter Moyle’s, on the complex interaction between freshwater conditions and ocean conditions: ‘Blaming ocean conditions for salmon declines is a lot like blaming the iceberg for sinking the Titanic, while ignoring the many human errors that put the ship on course for the fatal collision.’”
It was an important moment where we learned a lot, Dr. Johnson said. “Both that the stock isn’t as resilient as we thought it was, and that the ocean is incredibly powerful but it’s not the only thing that’s eroding this resiliency,” she said. “We also learned the importance of managing salmon across both of their aquatic ecosystems, and really starting to shape the way that we think about how these stocks will fare in the face of climate change. So what I’m hoping today in this symposium we’ll be able to do is talk about some of the emerging science that really links these freshwater influences and marine influences and then tie that together to look at how the population as a whole is responding to these influences and then how we can with our partners, managers, and stakeholders ensure that we have viable populations into the future.”
How salmon evolved on the landscape
Dr. Johnson started by talking about salmon in the past. “The sabertooth salmon were beasts that roamed the Sacramento and the San Joaquin rivers,” she said. “It’s no surprise that extinctions happen with climate. When climates shift regimes, things go extinct, and that’s exactly what happened to the sabertooth salmon, which are thought to have gone extinct during the Pleistocene.”
She noted the map on the right that shows the extent of where the populations of salmon used to live on the landscape, noting that it extends from the Central Valley all the way up to Alaska. She noted that the condition of the salmon varies, with the areas shown in red being where salmon are at risk of extinction; the areas shown in green which are mostly in Alaska are where salmon still remain and are thriving on the freshwater landscape.
“One of the amazing things about salmon is that they are incredibly resilient, so the salmon that have evolved to be in California in our historical climate have survived it,” she said. “[The early 1900s] were a time when there was hope and promise for Californians, that we could actually irrigate and raise crops on this arid landscape and there was hope that Reclamation was going to reclaim this arid west, so that we ourselves could actually live on this landscape. Salmon were presumably doing just fine.”
There were floods as well, Dr. Johnson noted. “There was the flood that happened in 1862 that happened before the Yolo Bypass and yet another engineering solution for us to be resilient on this very variable climate landscape,” she said. “One of my favorite quotes that came a decade later was one of John Steinbeck’s, ‘And it never failed, that during the dry years people forgot about the rich years, and during the wet years, they lost all memory of the dry years. It was always that way.’ And I ask, as a rhetorical question, is it still that way?”
She presented a figure from a paper by Mike Dettinger in 2011 paper, explaining that it is showing not annual precipitation but rather that variance in precipitation. “This is literally the coefficient of variation in precipitation, and on the scale of this darkest color here is the highest variation, so it’s that flood-drought dynamic that we’ve seen,” she said. “Here we are with Central and Southern California being one of the most variable precipitation landscapes on the U.S. What this tells us about the organisms that live on this landscape is that they mirror and have coevolved with this variation, and so we see this reflected in the salmon stocks that we have here in the Central Valley.”
California is unique in having all four runs of salmon geographically located in the Central Valley. “These four runs of salmon are in different listing statuses,” she said. “They have variation in adult spawn time that contributes to the genetic diversity between these groups. They are locally adapted largely due to the success of these strategies and the reproductive isolation in spawn time, either in space or time, between these runs.”
Ms. Johnson noted that at the time of this presentation (Sept 2015), 92% of the area of the state is either in severe or in exceptional drought. “We know that salmon in the past have been able to deal with mega-droughts and mega-floods, and yet we’re concerned about them presently and we’re concerned about their future in terms of this new climate regime that they might be experiencing.”
“If you think about the salmon toolbox and what allowed them to persist on the landscape, they have a lot of tricks in their bag of tricks and tools,” said Ms. Johnson. “This is the variation in life history diversity and traits with adult migration timing, spawning, incubation, rearing, across these four runs. There are salmon at life stage in the Valley year round, which is incredible. And so this is how salmon persist on the landscape with that environmental variation. And the question is, are we eroding their tools in their toolbox with the way we are managing them today?”
So how do we maintain this life history diversity in these salmon stocks if this is one of the primary tools that they have? “Well, it’s all about the base,” she said. “This is the salmon population viability pyramid, and at the base of this pyramid is habitat carrying capacity in all of their aquatic ecosystems. It’s this habitat mosaic, this environmental variation that facilitates the phenotypic expression and the success of these life history strategies over time.”
Winter-run Chinook salmon
Ms. Johnson then discussed the winter-run salmon and their functioning habitat, noting that it’s important for all life stages, but she will be focusing on the adult spawning and early juvenile rearing.
She then presented a slide showing the historical extent of habitat of salmon in the Central Valley prior to construction of the dams, and then afterwards, noting that the black indicates loss of habitat on the landscape. “What happened in this case is that there were strong selection for different runs with this one action,” she said. “Winter-run and spring-run were the ones disproportionately relied on these higher elevation habitats, so when the dams went in, we selected for fall run, which relied heavily on the Valley floor. “
“What’s regrettable is that it’s this high elevation habitat during droughts are probably very important for salmon,” she said “Winter run and spring run during long drought series have the reprieve of spring-fed river systems and this colder water, and so we’ve eroded that tool in their tool box.”
She then showed a graph showing temperature at Keswick Dam in 2014, and the temperature in the McCloud River above Shasta Dam for the same period of time. “A temperature of 56 degrees or lower is thought to be suitable egg and juvenile rearing thermal habitat, and you can see that it exceeds that in 2014 throughout,” she said. “Yet in the McCloud River, where they historically spawned, you can see that the temperature is much more conducive to supporting winter-run Chinook salmon, so again, this is higher elevation habitat that is no longer available to winter-run.”
There are four populations of winter-run: Battle Creek below Shasta Dam and three above Keswick and Shasta Dam on the Pit and the McCloud and the upper Sacramento, she said. “Now they currently spawn and rear in this upper stretch just below Keswick Dam, where they historically never spawned, and they really are on life support. That water that they get from Shasta Reservoir, that cold water in that deep water pool is like their elixir and life support line to be able to occupy this habitat that they hadn’t previously used. It is life support, because their eggs are in the gravel and juveniles are there in the summer time. … There are juvenile salmon in this habitat, and so they are literally relying on this cold water that’s below Shasta Dam.”
She presented a graph of the mean temperature coming out of Keswick Dam, noting that the gray shows the range of temperatures, the black line is the mean temperature, and the red line is the temperature in the river in 2014. “You can see the temperature that the winter run experienced in the upper Sacramento River and then you see this temperature spike where we lost all control of cold water going to that population,” she said.
“So what life stages might that have impacted?,” she said. “We have the carcass estimate and we can calculate, based on degree days and temperatures when eggs were going to hatch and fry emergence … it impacted this tail end of the egg stage and early fry rearing.”
Ms. Johnson then presented a temperature landscape map showing the temperature in the river downstream from Keswick Dam for 2013, which was also a drought year but not the year when they lost temperature control. “You can see the 56 degree line within the temperature set and the eggs largely being up here in this cooler water with the exception of this unfortunate red that was sitting below the optimal temperature,” she said.
She then presented the thermal landscape map for 2014. “We have our winter run eggs which are already so constrained in space, their historical spawning distribution was much lower … you can see this impact to winter-run occurring,” she said.
She presented a chart showing the percentage of egg to fry survival down to Red Bluff Dam, noting that the impacts could also be seen in the monitoring. “Over time, and you can see that the longer term average is 26% survivorship, and you can see that we have this 95% mortality occurring for winter run in 2014.”
So with lessons learned from last year, other strategies are being employed this year, she said. “Such as can we make that cold water last a little longer and make everybody a little warmer a little bit early on to stretch it out, and so that’s what going on presently, so stay tuned for what that might actually result in for egg to fry survival.”
News update: Ms. Johnson gave this presentation in September of 2015. About six weeks later, the Sacramento Bee reported that for the second straight year, huge numbers of juvenile winter-run Chinook salmon appear to have baked to death in the Sacramento River due to loss of temperature control. Read more from the Sacramento Bee here: Feds: Winter salmon run nearly extinguished in California drought (October 28, 2015)
“So the take home for the winter-run story is that this spatial diversity key,” she said. “This is not a sustainable strategy, eeking out a living below Keswick Dam and Redding where they historically didn’t occur. Again with climate change, while the ambient temperatures might not be a great thing for salmon, really it’s this elixir, this cold water and the extent to which climate variation is going to deplete this cold water.”
“In 2015 there was more water in the reservoir than in 2014, but less cold water, so that dynamic is really important and it’s going to be highly variable in the future,” she said. “They would do better in their historic habitat which is more conducive temperature-wise to what they are currently experiencing.”
Juvenile outmigration strategies: Supporting life history diversity
Ms. Johnson then turned to juvenile outmigration strategies. “Salmon are amazing,” she said “They leave their natal rivers as fish that are small; they also leave their rivers as fish that are larger; that’s incredible variation of size of fish leaving their natal rivers. The smaller fish is not sea ready, so they need to rear downstream, prior to entering the ocean.”
“When salmon leave their natal rivers at different sizes and times, it allows them interface, in theory, with the mosaic of complicated habitats that allow for habitat-specific growth rates and variation in phenotype that would result in variation in timing and size at ocean entry,” she said.
“This is important for resilient salmon because the way we manage our hatcheries isn’t necessarily in line with optimizing or showing that diversity,” she said, presenting a figure from a recent paper on changes in release hatchery practices at the Nimbus hatchery. “They recently published a paper coining this term, ‘advanced smolt’ for the evolution of how we manage our hatchery fish, which are like these big fish that hadn’t commonly been found in hatchery practices or on the landscape before, so these uber-advanced smolts. At Nimbus in the 1950s, they were releasing on site in the American River; then they started toying around with releasing in the Bay, then this diversity of strategy, all the way to 2000 where they are releasing very uniform sized fish, in one location over a very narrow window of time.”
This is important because our harvest relies very heavily on hatcheries. “One of the challenges is that we might be overleveraged in this one strategy,” she said. “We are really reliant on hatchery production for our fishery. Work that I published in 2000 showed that 90% of the fish that we caught in the fishery were of hatchery origin using natural markers, and then we have the constant fractional marking program confirming that in general, we are really reliant on our hatchery production.”
One challenge associated with this narrow release strategy is the size and timing of ocean entry, Ms. Johnson said. “If you think about prey availability in the ocean as a normal distribution, and you think of migration date, if you have a broad window of out-migration strategies, some individuals will interface with the ocean at the optimal timing and some of them will not, but in general, the mean of the fish that are leaving will coincide with the optimal habitat when they arrive.”
“If you have a very narrow window and they migrate earlier than the ocean is producing good upwelled water with a lot of forage, then it’s a mismatch,” she said. “So it’s this match-mismatch associated with migration and ocean entry and when spring transition comes that would make these stocks very vulnerable with this narrow strategy.”
Ms. Johnson then presented a slide showing return rate of 3-year old coho adults. “Historically at the hatchery on Scott Creek, they would release fish early in the season with the vernal equinox and the new moon,” she said. “What they opted to do was to release 4000 fish each week over this broader window of time, and what you can see here is that in this particular year, the fish that were released later had higher survival.”
“We as scientists are asked this question a lot – what’s the right time to release fish, and we say its March or it’s May … as though there is a singular solution set,” she said. “We never know what Mother Ocean’s going to do, our science is uncertain. Potentially we might want to think about more of a diversification strategy than an optimization strategy.”
The role of life history diversity
The last story Ms. Johnson had was about the role of life history diversity given the hydrologic variation in our system. “One of the questions is, salmon produce fish at a different size that leave their natal rivers, but do these guys really survive and contribute? Which of these life history strategies are viable?”
The story takes place in the Chinook salmon’s most southern distribution for this species range, the San Joaquin, she said. The work of Stephanie Carlson along with others shows that the portfolio of stocks is very weak in California, she said.
“We are not well diversified, and so our populations both on the Sacramento and San Joaquin side fluctuate in a similar way, and one of the take home messages from her paper is if we can get the San Joaquin back online, that’s probably one of the smartest things we can do to buffer our system.”
Ms. Johnson then presented a slide of adult returns to the San Joaquin. “One of the cool things about the San Joaquin is that flow really matters – it’s a pretty dewatered system so you add a little water and fish respond and you don’t always see that in systems,” she said. “What I am showing you here now is adult returns and the flows in the springtime, two and a half years prior to when the adults returned, so flow seems to really be an important driver of adult returns. Here, I have an asterisk here is because here is that poor return year due to oceanographic conditions, again this points towards this lack of resiliency that we have in our system. They are very vulnerable to ocean predation.”
This study takes place on the Stanislaus River where researchers were investigating a better understanding on outmigration size and timing and who is contributing to the adult population.
There are two types of hydrographs on the Stanislaus: One for flood control release where water is released and fish respond and a lot of fish leave; the other is a managed spring pulse because conceptually, they think conceptually bigger is better so the limited water on the smolts rearing in the natal river until it’s time to go to the ocean, she said. In dry years, there isn’t a flood release, so there’s a bit of a managed pulse where a few fish leave but not an expression of small fish (55 mm) leaving.
“So the general pattern is in wet years you’ve got a bunch of fish leaving and they are small; in drier years, you don’t’ have these early leaving fish or small fish, you see these smolts that are leaving the system.”
Ms. Johnson explained that each of the salmon rivers has a unique geology that is incorporated into the fish’s otolith (or earbone) on a daily basis, much like tree rings, and so the origin and movement of fish on the landscape can be tracked by looking at the chemistry of their otoliths. By studying the otoliths of the adults that returned, they can determine if the fish left their natal river as small fish or as large fish.
Some of the fish were found to have spent more of the time rearing in the San Joaquin and the Delta, while others reared for a longer duration in the Stanislaus. “What surprised us is that in the first wet year that we looked at, over 20% of the adults that returned to the Stanislaus left the Stanislaus left as small 55 millimeter sized fish,” she said.
“One thing to then think about, if these fry are actually important demographically in these wet years is how we’re managing that hydrograph,” she said, presenting a graph of the one day maximum flow before and after New Melones Dam went in. “You can see that variation in flow that cues this juvenile outmigration has been suppressed over time.”
“The fry not only are potentially being suppressed from expressing this life history diversity, but we all know that the habitat that they are arriving at in the Delta is much changed from the historical habitat mosaic versus the agriculturally-dominated diked landscape of today, which is simply to say that because we see fry thriving in the population, that restoration efforts intending to increase their survivorship might actually be a productive population-level benefit,” Ms. Johnson said.
She then displayed a set of graphs showing how this data has been applied over multiple years on the Stanislaus. “You can see this bimodal distribution in the size that juveniles leave at the rotary screw trap in these different years – these are the fry that are leaving and these are the larger smolts.” She then displayed the adult reconstruction. “So in the adults, they left their natal river at this intermediate size. What we think this means is the potential for stabilizing selection, where if you leave too small, you’re survivorship is not that great due to predation and if you stay too long, you cook in the San Joaquin; and so what we see is that despite very few fish leaving at this size range, they tend to be the ones that inherit the survivorship distribution in the adults across a variety of different hydrologic years. This isn’t to say that diversity isn’t important – we see that in wet years, frys do contribute, and there are so many that leave as fry that they are potentially demographically important.”
“What I think is very interesting here is that the juveniles are expressing that variation which when we think about climate change and traits that might be selected for and traits that might be at the tail of a distribution, we’re poised pretty well in terms of what the juveniles are expressing in traits that are heritable,” she said.
Ms. Johnson said that they are embarking on some work looking at the return of spring-run fish to the San Joaquin, which are considered extirpated from the basin. “We see spring-running fish coming into the Stanislaus River during spring-run timing windows, and when you look at the climate projections from the Lindley paper, although there are more recent climate projections, the Stanislaus River is one of these rivers that would potentially be kind of a safe haven for spring run in the climate future,” she said. “One of my favorite take home messages from Ray’s work in 2003, is ‘tomorrow’s most important populations might come from populations that are relatively unimpressive today.’”
So in conclusion …
“So just to end, in our state that is extensively plumbed as California is, the state of fish can be more dependent on water managers and water operators than on nature in many ways,” Ms. Johnson said. “Our management actions can act to support, maintain, or erode life history diversity and the very tools that salmon may need to respond to future climate changes and to locally adapt to those changes. And so I want to end by saying that salmon populations require a full tool box to thrive in an uncertain climate future.”
Coming up next …
Tomorrow, the impacts of the changing climate and predation on California salmon; on Thursday, floodplain restoration and specific steps that can be taken to preserve salmon in the Central Valley.