Dr. Sarah Yarnell is Associate Project Scientist at the Center for Watershed Sciences, and her presentation was drawn from research she and her colleagues have done regarding the relationships between aquatic native species and instream flow regimes.
Specifically, she will be talking about research they conducted in the northern Sierra mountains draining into the rim dams, some of the research on the restored floodplain on the Cosumnes River, the ecological cues that native species use in relation to the flow regime, the effects of those flow regimes in regulated and unregulated systems, some of the research of climate change impacts on flow regimes, and then talk about some of the potential implications for management in our Sierra and rivers here in California.
California’s Mediterranean climate is particularly driven by wet flood regimes in the winter and drought in the summer. “What happens in the transition between those two, particularly the snowmelt recession transition, becomes a real key time that’s an indicator for our species in these systems,” she said. “It’s the one time where we do have high flows in the river, there’s high resources available, but the flows are predictable. They are not flashy; they are not changing erratically, and so because of that, there’s a moment of opportunity for species who are adapted to this flow regime to be able to cue their life history and their reproduction around this predictability and time. This then results in actually a relatively high diversity for our systems here in California.”
“We know that flow regimes affect both the geomorphology and the hydrology, and how much discussion has been going on about how these two are intimately linked. The flow does drive the abiotic processes that happen in our rivers, but the purposes of my talk here today is to talk about these ecological cues and ecological connections, and in particular these flow regimes that provide a lot of cues in terms of timing and behavior that are particular important for our species.”
Ms. Yarnell focused on two species to illustrate this: the foothill yellow-legged frog, whose breeding and reproduction is completely tied to the spring snowmelt recession in the Sierras, and the outmigration of juvenile chinook salmon on the Cosumnes River floodplain.
Example 1: Foothill yellow-legged frog
The foothill yellow-legged frog is one of seven native frog species in California. Its range extends from Oregon to California, occupying the mid-elevations. “What’s unique about it is that it is a completely lentic species,” she said. “It breeds in streams so it uses the riparian and tributary habitat which is unlike other amphibian species which focus on ponds or stillwaters, so as a result, this species is kind of unique because it’s had to craft an entire life history around these fluctuating flows that happen in the river.”
Because of its sensitivity to the river regime, it’s actually had a range contraction and it’s disappeared over the last 100 years from roughly about 50% of its historic range. She presented a map, noting that the black dots represent historic observations dating back to about 1910 from the biologists at the time that were working for the USGS; the current yellow dots represent current locations that are still extant.
“Pretty much in the southern Sierras, the species is almost entirely extinct; in fact it has been proposed for listing in the southern Sierras, and there’s actually been a range contraction to the north in Oregon as well,” she said. “It is currently a California species of special concern and so it does require consideration and management, particularly in regulated river systems in the mountains and the FERC relicensing process, for example, or other regulatory requirements along that line. There has been a fair amount of research to try to figure out impacts of flow on the species.”
The life history of the foothill yellow-legged frog is unique because it is completely in sync with California’s annual average flow regime.
“Their strategy for this species is to lay their eggs in the bottom of a river in a sheltered location on the downstream side of a rock in a spring, at the tail of the recession as those flows are predictably receding down into the summer,” she said. “The low warm flows in the summer then actually provide a great rearing habitat for the tadpoles. There’s a lot of algae production that happens, the flows are low and slow, there’s not a lot of scouring flows so the tadpoles are actually free to migrate around the bottom of the rivers and streams eating algae and growing big and fat. Within about three months, they then metamorphize, transition into small little froglets about 1” big, and they actually hop away from the river and leave the stream, moving to overwintering refugia and tributaries before the high flows come back in spring. So in this way, their entire life history is completely cued into these predictable flow regimes.”
The North Fork American River is one of the largest drainages in the Sierra that is still unregulated. Ms. Yarnell clarified that she would be using the term ‘unregulated’, not unimpaired; there are still a fair number of smaller drainages in the Sierra that are not actually reregulated and reoperated by dams, she said. The graph on the left shows the general pattern of having winter high flows and summer low flows, but it’s very variable from year to year. There are wet years with large magnitudes, a late snowmelt and a late spring recession; and there are warmer, dry years where the rains come early and the snowmelt is small so it happens early and quickly.
“But what you’ll notice in the graph on the bottom is that the snowmelt recessions are quite predictable,” she said. “They tend to have a similar shape, regardless of the water year type. If you were to translate that into the river to stage and think about the actual changes that are happening in stage through time during that snowmelt recession, what is really interesting is that regardless of the water year type and regardless of how much flow is in the river, these stage changes are completely consistent from year to year. The recession rate in a wet year versus a dry year is roughly 10 centimeters per week, regardless of the year. It’s this timing that the frog has really cued into in their reproductive cycle.”
Research studying frogs and breeding requirements across all the rivers in the Sierra have found that on average, frogs tend to lay their egg masses at roughly 20 to 40 centimeters in depth at very low velocity locations in the springtime. It takes about 2 weeks for the egg mass to hatch out, and another week for the tadpoles to get big enough to follow that receding flow line.
“Right there that’s 30 centimeters on average over three weeks, that’s the same rate of stage that we saw at those breeding locations in the North Fork American,” Ms. Yarnell said. “They are genuinely tied to this recession rate. If the flows increase during the spring time, they can be scoured away and conversely, if the recession rates are too quick or the flows drop too quickly, it can leave the eggs stranded.”
The frog’s ability to use the ecological cue of the receding recession does provide some plasticity for them and an ability to fluctuate with the different water year types.
“They are cuing into a receding recession rate and an increase in temperature, so in wet years such as 2011 on the North Fork American, we had a very wet late spring, the snow pulse didn’t even come until June, and the recession didn’t actually start until July. Correspondingly, frog breeding began in the last week of June and extended into the middle of July. This was the latest breeding that has ever been observed anywhere in the Sierra in the last twelve years of records.”
Conversely, 2012 was a drier year with an earlier snowmelt recession. “The cues in the trigger are the receding flows along with an increase in warming temperatures and that caused the frogs to breed in roughly mid May, ending in the first week of June. So, although the water year type is changing and the conditions are changing, by being ecologically cued by these flow regimes, there is actually some plasticity that results in these species that allows them to be somewhat adaptable year to year and being able to tie into that.”
Example 2: Juvenile salmon on the Cosumnes River floodplain
The Cosumnes River floodplain is a lower elevation floodplain; like the North Fork American, it is also unregulated. The floodplain area is on the edge of the valley floor. It’s a mixed rain snow system so it has the characteristic high winter flows with a spring snowmelt recession coming down. The Cosumnes is a slightly smaller; it does not quite extend to the crest of the Sierras so it has a slightly smaller snowpack. Its recession typically ends more in late May; as a result, there are native spring spawners and salmon that utilize this floodplain habitat on the river down in the lower basin who are cuing off of the ecology of the floodplain.
The splittail (pictured on the left) breeds out on the floodplain and the juvenile salmon (pictured on the right) uses the receding flows as a cue for outmigration out to the river.
So what is it about the floodplain that makes it so important and dynamic? “There’s a whole food web that goes on out in the floodplain that is similar but slightly different to what goes on in the channel,” Ms. Yarnell said. “There’s typically a flood pulse that will come down and wet the floodplain. There will be a bloom in phytoplankton that occurs; there’s a subsequent bloom in zooplankton that occurs as they feed on the phytoplankton, and then following that then, of course the zooplankton is just fantastic fish food so there’s a whole dynamic that can occur if there’s enough time for the water to be out on the floodplain for this dynamic to occur.”
Work has been done at UC Davis to explore how long a residence time is needed for these functions to occur. She presented a graph from 2005, a wet year, that shows a series of flood pulses go out onto the floodplain and roughly one week later, there’s a corresponding pulse in chlorophyll or phytoplankton that results. “This pattern is repeated throughout the winter as each flood pulse comes down,” she said. “Even when the flows disconnect from the floodplain and then reengage and then subsequently when the spring snowmelt pulse coming down, there’s a large pulse in phytoplankton again and then this pulse at the very end is resulting from ponded water that stays out on the floodplain into the summer.”
For zooplankton biomass with a similar flow regime, not much happens with zooplankton, but then subsequently two and three weeks later, there is a peak in the zooplankton that then recedes down following that flood pulse.
“So the timing of these two combine together,” said Ms. Yarnell. “So with a flood pulse coming down, roughly one week later, we see a bloom in phytoplankton; two to three weeks later you see a bloom in zooplankton, and then the fish benefit really kicks in after that. This is giving us an idea of this idea that there needs to be enough duration and ramp down of the flows following a flood pulse to really make it ecologically beneficial. In the case for the Cosumnes, it’s roughly three to four weeks at a minimum is really where you get the most bang for your buck in terms of growing bigger fish.”
There are additional cues that happening out on the floodplain, similar to the channel. She presented a graph showing the percent of native fish leaving the floodplain for year 2012, noting that the red in the pie charts represents percent native fish, the white is the nonnative fish percent of that sample.
“We can see that during the wintertime as these pulses are coming in and out, roughly a third to a half of the native fish are going to be leaving the floodplain to head out. Following the spring snowmelt, these are diurnal snowmelt pulses, this is a rain event that happened on top of the snowmelt pulse, but during this time you’ll note there’s a bit uptick in the percent of native fish leaving, particularly on the receding end, and by the time the floods have receded off the floodplain, and there’s no longer any connection with floodplain, there are also no longer any native fish out there on the floodplain.”
“These cues of warming temperatures and receding flows are providing the signal to the fish that it’s time for them to leave. They are adapted for this because they are used to this predictable annual cycle that comes through that has allowed them to adapt to these types of cues.”
Implications for managed flow regimes
Ms. Yarnell then presented a graph showing an example of three hydrographs from the American River system in the upper Sierra. The red line is a hydropeaking reach on the middle fork American; the blue line is the North Fork American which is unregulated, and then the yellow line down is a bypass reach. “At first glance, you can see that the Middle Fork American does provide a lot of hydrologic function, even though it’s a regulated river. The high flows in winter are still allowed to spill downstream, the rain events and larger flows are still creating a more natural signal. However, when you get to the springtime and the summer, you can see that there’s a loss of some of the spring recession and a cue into hydropeaking which is what happens in the summer.”
“On the bypass reaches, you get some loss of peak flows. You do get some of the spring pulse, but because they stop the spill over the dam and just completely shut it off, you’ve lost entirely the spring recession flows and you just go down to the base flow in the summer. So in both of these systems, although there’s some hydrologic function, the loss of the spring flows has really decreased and removed these ecological cues from the system.”
If we want to restore these flows to the system, this can be quantified, Ms. Yarnell said. “Some of the simplest ways to think about quantifying the spring recession are to think about the timing of when it occurs, the rate of change, and how quickly it draws down. This is a really important factor in cuing our ecological species and the magnitudes, and this combination of rate of change in magnitude is what really lends to the duration of how long these flows are going to last.”
There are eight unregulated basins that are gauged in the Sierra, ranging from the Feather in the north down to the Kern in the south; the graph shows the accumulated mean daily hydrograph for each one of those basins which shows the variability from basin to basin. The blue line is the Kern, which is snowmelt driven. The San Joaquin is quite a bit later with its pulses because the snow isn’t even melting there until June and July. Conversely, the green line is the North Fork American which is a mixed rain snow system with high winter pulses and again a predictable spring recession.
“What you’ll notice in all of these unregulated systems is this really predictable spring recession that’s occurring,” she said. “If we quantify that recession in terms of a daily decrease in flow or a daily percent change, what we find is that they are extremely uniform across all these basins. In fact, they average 4-5% per day. At the very beginning of the snowmelt recession, they might be on average as high as 8% per day, but typically they are decreasing down to 4 to 5% per day. Even in extreme events, there is no time where these unregulated rivers decrease more than 25% per day.”
So what happens in regulated systems? Ms. Yarnell presented a slide with three hydrographs from the Yuba. The north Yuba, shown in blue, is unregulated; the middle fork (yellow) and the south fork (red) of the Yuba are both regulated. They are bypass reaches, so they sit at low flows through the year. They spill in spring with the high flows, and typically their spill is completely curtailed, so there’s a loss of the spring recession.
“What happens in these basins is that on average, the duration of the recession is only 17 days on average, compared to 70 days on the north fork Yuba, and the rate of recession is extremely high,” she said. “On average, they are averaging 17 days, but what’s really important is to look at the maximum. These rivers have a recession rate of 80-90% so that’s going from 2000 cfs to 50 cfs in two days, and so you’ve just completely lost any ability for any kind of ecological cue to be happening in the river, much less allowing any duration to translate downstream.”
“So if you want to think about remanaging and reoperating our dams, we can think about using these recession rates as a cue,” said Ms. Yarnell. “They are completely consistent, they are easy to describe and talk about, and they can easily be modeled in a regulated system. You simply need to pick a starting percent daily percent decrease and a starting discharge that works for your managed system. You can even turn it into steps that are appropriate for the infrastructure that allow for working with the operations that you have. We’ve been working over the last couple of years trying to get some of these downramp flows from spills happening in some of the hydropower dams that are upstream, for example on the middle fork American and the upper Yuba Rivers. We have been in discussions within the FERC relicensing process to try to restore some of these downramp rates using this type of approach.”
They have done work trying to model the impacts of climate warming on Sierra rivers, as climate warming will add further complications. Ms. Yarnell presented a graph showing inflows to the Yuba River, noting the dark line is the current historic condition in terms of weekly flows.
“As climate warms and we get an increase in air temperature, what we will see in the Sierra Nevada is a shift from the snowmelt dominated system to a more rain dominated system, as more of that snowfalls as precipitation. What’s going to happen is that initially there’s not going to be in some systems quite as much of a drop off in annual runoff; it’s just the timing and pattern of when that runoff is going to occur is going to change.”
“If you look at an unregulated system, however, if there are not interruptions to the general pattern, although the climate is going to shift, you can still see some of these ecological cues will still be in place, so there will be a shift to earlier timing,” she said. “There might be a slight change in the rate of change; there might be a slight change in the duration, but overall some of those general predictable patterns will still be there. I think this is an important thing to note about because as we think about moving forward, even though there will be things that will be outside of our control, there are some aspects of the flow regimes that we can be able to control, like maintaining some of the patterns for example.”
There are economic tradeoffs, she acknowledged. “We’ve had a lot of discussion about this in some of our research, and one approach that we’ve taken is to try and think about creating a way to quantify how much those tradeoffs occur,” she said. “If you look at the panel all the way to the left, this is a graph showing that if you wanted to increase the minimum instream flows that are happening in a regulated river, what would be the associated cost of that increase in terms of hydropower change?”
“Similarly we could also impose a downramp rate. There’s a cost to downramping from spill in terms of lost water. 25% per week would be comparable to a 4 or 5% per day, so that would be a full natural downramp rate, and then something lesser than that. You can then create these tradeoff curves of how much it costs for each of those to happen.”
There are two ways to look at this, she pointed out. “The environmental managers might say, ‘we really want to maximize and have the most natural downramp rates that we want, and we would really like to have a 50% increase in minimum instream flows and it really only costs 3.5% loss in terms of hydropower generation,’” she said. “Conversely, an operator might say, ‘I’m not willing to give up more than 2%. You can have 2% but you can pick where you want it to go, and you can go to a curve like this and be able to say, we either would like to maximize the downramp rate, increase our minimum instream flows or some combination of the two. Then as you move forward with increased climate warming, we can think about how those tradeoff curves might change and how those costs might change through time.”
So lastly, there are cumulative impacts downstream to be considered. “As our demands on the water resources increase both through climate change and through increased population growth and ag demand, it becomes really important to think about how the resources we have can be better managed.”
She presented a schematic of an idealized Sierra Nevada river with rivers draining from the east down into the west. “Currently what happens is that each dam is operated completely independently. You might have a dam at the upper end that is a bypass reach and it could get base flows all year, and then there’s some sort of spill recession; you might have other dams that are completely bypass, they never spill because all the water is taken, and what happens as you work your way downstream is that the timing of these spill events are completely disconnected. The large events might be connected to the climate in terms of a really big rainstorm or a really big wet year, but a lot of the smaller spills are disjointed from each other.”
“You might get the spill from one dam translating to the next spill downstream in an extreme event, but on average, there’s a lot of disconnection,” she continued. “By the time you get to the bottom and you’re thinking about trying to reduce flows down the valley floor, it’s become almost completely discoupled potentially from what is going on in the actual climate, unless you’re in those really big wet years where those storms are driving what’s going on, and then particularly if you are thinking about needing to manage flows at certain times for water deliveries, you just end up with this really disjointed hydrograph.”
“There’s an opportunity in thinking about this at the bigger picture to be smarter and to use a smart hydrology approach, and to think about how we can use the cues from the actual climate to better create our flow regimes,” she said. “If we ramp down from that spill and we even create perhaps a small spring spill or an ecological pulse that has a ramp down, whether or not that dam is actually going to spill, and the timing of those is corresponding to the timing of wet years, similar rivers are around, you can end up with a more cumulative effect downstream so that the water is allowed to pass from dam to dam to dam down to the valley floor and giving you the most benefit throughout the system and trying to really maximize the amount of water that’s hitting the valley floor at that one time.”
In conclusion …
“In conclusion, I just want to reiterate the fact that these spring flow regimes are really providing key ecological cues for our native species here in our climate,” she said. “It’s really important to consider what it is about the flow regime that is important for the species, not just the shape of it, but what is it about it that is really key that we really need to focus on and think about. I would argue that these spring recession flows are really quite easily modeled and manageable in these systems. We started working our way towards that in working with FERC relicensing, and I think then there’s great opportunity to think about then how that can be managed further down and down on the valley floor.”
“There’s a real opportunity to think more smartly about coordinating management from these upstream reservoirs and how they can propagate down to the valley floor so that we can really make the most use of our water,” she said. “Then lastly, a lot of this work has been able to be done in these unregulated systems, and although we’re not really going to be able to restore these systems, but the ability to be able to look at processes that are happening in other rivers and think about what those cues are in those relationships are, it can be really valuable when we come back to our managed rivers.”