The 8th Biennial Bay Delta Science Conference brought together over a thousand scientists, managers, and policy makers to Sacramento to hear the latest advances in scientific information and ideas on water resource management in the Delta, its watershed, and the San Francisco Estuary.
As the fifth and final plenary speaker, Delta Independent Science Board member Stephen Brandt began by saying he would be talking about fish habitat quality. “Understanding habitat quality is perhaps a real core of our ability to predict a species persistence in the environment, and today I’m going to provide examples from outside the Bay Delta area to explore some of the key elements of thinking about habitat quality,” he said.
“So for the next few minutes, I want you to make believe you’re a fish, and the reason for that is you need to look at habitat quality from the perspective of the fish. The fish’s physiology and behavioral capabilities will determine what the value of that habitat is to that individual species or life stage of fish. A fish habitat quality is actually formally defined by NOAA as essential fish habitat, and that’s habitat that affects survival or reproductive success, growth or production.”
A habitat is more than a structure, such as a coral reef; it’s clearly more than that, he said. “Think about the habitat quality from the perspective of a pelagic fish, or a fish that lives in midwater,” he said. “How do we define it, how do we measure it, and how do we quantify it to give us the tools that allow us to compare habitat quality across species or ecosystems, across both physical and biological gradients, time and space, and also how it might respond to various environmental stresses.”
Dr. Brandt said to think about the growth rate potential for a fish as a measure of habitat quality. If the temperature, salinity, oxygen and prey density are known for a relative small uniform volume of water (about a tenth of the size of the conference room), the growth rate potential for that particular species can be calculated. He noted that this depends on the species; another species placed in the exact same environment might grow very differently.
“Growth rate is clearly something that’s been used throughout ecology as sort of a measurement of integrated response of a fish performance to its environment as it has direct impact on its survival, and its reproductive capabilities. It is based on the fish’s requirements and physiological needs in the context of the prevailing environmental conditions,” said Dr. Brandt. “It’s going to differ among species and life stages, it’s going to be varied in time and space, and it’s going to be nonlinear.”
The models are fairly well developed and are based on a mass balance that says when a fish eats more, a portion goes to growth and a portion to waste products, and for various species, these kinds of models are fairly well established under known environment conditions, he said.
He then presented a slide of the model results for striped bass. “This is based on a laboratory experiments and looking at what the growth would be for an adult striped bass under different conditions of oxygen and temperature,” he said. “There are areas where no matter how much food is given, the striped bass will still lose weight under high and under low temperatures, and under low oxygen; that the relationship is clearly non-linear. There are peaks and valleys and it’s a non-linear integration of two parameters: temperature and oxygen.”
“What does that nonlinearity mean?,” Dr. Brandt said. “In very simple terms, growth at two different temperatures is not the same as the growth at the mean temperature. You cannot take mean temperatures either across space or time and use that to evaluate how well an individual fish might grow under that particular temperature condition.”
“Another way to look at it is if you have a complex environments as shown by a spatially complex grid and you start averaging those conditions, either average across space or across time, what in essence you do by taking an average is you eliminate the very highs and lows, and those may be the conditions that are the best or the worst for that individual fish under those conditions. So that’s just one aspect of looking at the habitat.”
“What our approach has been is to provide the habitat in very small spatial or temporal scales, and then measure things like prey density, temperature, and environmental conditions in those cells and then run it by the bioenergetics models to calculate how well that fish would grow if placed in that environment,” he said. “The cells we’re talking about that are about a tenth of the size of this room, so we have to do run those same models 10,000 times and look at it from the perspective of that individual species of fish, how well it would do in that environment.”
Dr. Brandt said he would be giving some examples of work being done in the Great Lakes, the Chesapeake Bay, and the Gulf of Mexico, that focuses on hypoxia, a stressor that impacts some of the fish.
He then presented the results for striped bass growth potential for the middle of the Chesapeake Bay, noting that the first two panels are the monitoring data for water temperature and dissolved oxygen over an annual cycle, and the third panel is from the viewpoint of the striped bass. “This is how well could you grow under those conditions given unlimited food supply,” he said, noting that the red areas indicate high growth rate.”
He pointed that that growth rate ‘sort of’ follows oxygen but not exclusively because temperature plays an important role. “It’s really the combination of temperature and oxygen filtered through the striped bass physiology in a nonlinear way to say what that habitat would look like,” he said.
“That’s just one year. If you look at the growth rate potential for three different years for that striped bass, you can see there’s a lot of interannual variability of how well that striped bass will do that is purely based on interannual variability and the physical conditions of that system.”
“Another way to look at it is to say within one year, if we look at different locations in the Chesapeake Bay, because hypoxia and the temperature structure differ, then you get a wide degree of geographic variability in any single year at different points,” he said. “So in an ideal world, what you need to do is put together the full three dimensional structure of what that habitat quality looks like and how that is quantified across the Chesapeake Bay in time and space.”
Dr. Brandt noted that the results he has presented so far have been based strictly on temperature and oxygen and giving unlimited food supply; however, food supply needs to be measured as well. “One of the ways we do this is to use an underwater acoustic system,” he said. “Basically it’s like sonar that we use to sample the food distribution continuously through the water.” He noted it can be towed from vessels large and small to conduct in essence continuous spatial measurements of the underlying habitat structure.
The Chesapeake Bay was sampled at this intensity during the spring, summer, and fall for over five years. He then presented a slide of some of the results for October. “This is an example of how the prey field looks like, and in most systems, the prey that are very highly distributed and in some cases in layers, depending on the underlying physical conditions,” he said. “But from the point of view of a predator, what does that look like.” He noted that the top figure is temperature taken in October and shows very little temperature variability; the second figure is food distribution, and the third and fourth figures are what that might look like to a bluefish and to a striped bass. “The take home message here is that it looks pretty similar to a bluefish and a striped bass. It may not be as good of habitat quality but there’s a high degree of overlap; one could argue that in this system at this time, it may be the bluefish or the striped bass, their habitat quality as defined by growth rate is highly overlapping, maybe there’s even some competitive interaction occurring between these two species.”
He then presented the results for the bay anchovy. “The bay anchovy is very similar to the Delta smelt,” he said. “It’s an annual species and it lives its one year entirely within the Chesapeake Bay so in this case, so one might expect conditions of how well it grows might be correlated to how well the next year’s crop might look like.”
Dr. Brandt said they took all of the monitoring data for zooplankton, food measurements, temperature, oxygen, and salinity over the course of a number of years and plotted that out. He noted that the black line is the mean growth rate over the summer, and the other two lines are independent measures of fish abundance. “What this suggests is that at least the growth rate potential does mimic the annual abundance of this fish. It’s not based on temperature, oxygen, or prey density; it’s based on all of those things combined through the physiology of that particular species.”
High nutrient loading in the Gulf of Mexico leads to a very extensive dead zone or hypoxic area. “Our questions have to do with how hypoxia affects the fish so we ran a series of sampling to show what the hypoxic structure looks like if you take a series of sections off the edge of the shelf,” he said, noting that the red areas indicate those areas of low oxygen. “That’s what it looks like to us, but what does it look like and what does it mean to a fish?”
He then presented a slide with four panels, noting that the upper left is oxygen, the upper right is temperature, the lower left is food density, and the lower right is what it looks like to an adult bluefish. He noted that the red areas are those with a high potential growth for that species if placed in that environment.
He next presented a slide of what those conditions look like to a striped bass. “Striped bass has very poor growing conditions because the temperature is just slightly higher,” he said. “Even though the oxygen is okay and the food level is okay, the temperature is just slightly higher which means you cannot feed in those areas, so unlike the Chesapeake where these two species may have been competitive at least in terms of their habitat quality of overlap; here they have no habitat quality overlap. There’s also very few striped bass in the Gulf of Mexico in this area and this may explain in part why they are not there.”
“We were fortunate enough to be out during Tropical Storm Edouard,” said Mr. Brandt. “Before the storm, there is a lot of structure, both the physical and biological structure, and then after the storm, everything got mixed up. How frequent and how important are these events when they disrupt and mix the system, and how long does it take for it to get back to normal conditions? When we think about the Bay-Delta and a flooding event, how do those sorts of things impact the habitat quality, and how long do those impacts survive and permeate throughout the system?”
Dr. Brandt then turned to the Chinook salmon in the North Pacific, noting that it’s the same species that we have here. “A large portion of that species’ life is spent out in the open ocean. What is the habitat quality like for that species out in the open ocean? And can we begin to quantify that?” he said.
He presented a slide of the surface temperature and noted that it’s quite complex. “What does that look like to a fish? We’ve been able to do some preliminary looking at a station 200 meters deep and a station 600 meters deep. … using the same kind of plots I used before, on an annual cycle, and the top figure is the temperature at the shallow depth over the course of a year, the second one is oxygen, and the bottom one is what that looks like in to an adult Chinook salmon.”
“As we saw in the Chesapeake, there’s a lot of interannual variability. These are four different years, from the perspective of what the growth might be for a Chinook salmon.”
This can be quantified in a number of ways, he said, presenting a frequency histogram of the growth rate potential. “We don’t really know what this means yet, because do fish respond to average conditions? Or is the fact that in 2003, you had areas of very high growth rate, is that something good? Does that result in higher growth rate for Chinook salmon? Higher growth rate may mean better supply and better returns. That’s one of the areas we’re exploring.”
The results for the deeper station are quite a bit different. “You only see growth in the very near surface areas, if you translate that to a frequency histogram, it looks something like this, and clearly the shallower areas have a higher proportion of good growth areas.”
Dr. Brandt then turned to climate change and its impact on the Great Lakes, noting that Lake Michigan has three times the surface area or 66 times the volume of Chesapeake Bay. “When we think about fish habitat, clearly it’s volume,” he said. “They have a big system, but still one could wonder how would climate change impact that.”
He then presented a plot of depth versus the annual cycle under baseline conditions, 2030, and 2090. “As you get warmer, clearly there are surface areas where the Chinook salmon won’t do well. But the Chinook salmon likes 15 degrees centigrade and that’s always there, even under the warmest conditions because we have the depth to give deeper water and cooler water. In the case of the Chinook salmon, the lake warms up sooner and stays warmer longer, so the actual growing season over the course of a year is longer, so we concluded in this paper that climate change was actually beneficial to the Chinook salmon in terms of its growing season, so the duration of events is something to really consider.”
Dr. Brandt had one other point to make. The model conditions for 1996 coincided with the measurements, but for 1998, it was much warmer. “It looks like it was a very warm year, and in fact, 1998 was somewhere between 2030 and 2090, and what that suggests to us is that the current years that we are experiencing now and might experience more frequently in the future suggest that maybe these unusual years might tell us something about what might happen under climate change. The drought now being a good example of that.”
Dr. Brandt said that so far he has talked about what habitat quality might look like to the fish, but it’s independent of where that fish is. “So when you really want to calculate observed growth, you really need to look at how the fish respond to those conditions and do they actually go to areas where things are good.”
“Another way of looking at it is looking at growth rate potential and habitat availability,” Dr. Brandt said. “The top figure says here is where the habitat is available. You might have conditions where fish might be very effective at using that habitat quality, only going in those areas where there’s good growth. On the other hand, you might have fish that don’t do very well under those conditions. Maybe they are restricted by flow, by changes in the canals, or by other things that allow them not to be able to use the habitats that are most valuable to them.”
Dr. Brandt then related how his presentation applies to the Bay Delta. “One is that many of the responses of fish to food in their habitat are clearly nonlinear, and that raises questions about how good correlations might be, particularly when you’re given multiple factors that interact, both biological and physical factors,” he said.
“You need to look at correlations across species that have different physiologies and correlations across things that are nonlinear to begin with. You need to consider both the spatial and temporal scales. Averaging across habitats and averaging across times can be dangerous. And there are things like events, long term changes and also the duration of conditions. It’s not just that you might have good conditions at any one point when you measure it, but how good of growth conditions for how long of a time period is really important when you think about a fish and its growth perspective.”
“Clearly the fish physiology and perceptive capabilities will vary among species and life stages. And in terms of habitat choice options, just having good habitat quality is only the first step. You need to make sure that the fish can actually use and do use those habitats for it to be effective in the environment.”
“And all of this argues that [from] a whole ecosystem perspective, I know doing Bay Delta wide studies are difficult and expensive, but clearly they’ve been done on much larger systems and I think those are the kinds of things, particularly with the spatial and temporal complexity that you have, it would give you insight into fish habitat quality.”