| |

Inside Scientific Webinar: Ecophysiological Impacts of Climate Change: Performance, Fitness and Extinction

Please join Caroline Williams, PhD and Eric Riddell, PhD as they discuss their research involving climate change and the ecophysiological effects of changing global temperatures on organismal biology and survival.

To watch the full archived webinar, click here.

VIDEO NOT LOADING? Viewing videos requires acceptance of all cookies. To change consent settings, open the MANAGE CONSENT tab at the bottom right of this window and click ACCEPT.

Sarah: Good morning, good afternoon, and good evening, everyone, and welcome to our webinar titled “Ecophysiological Impacts of Climate Change, Performance, Fitness, and Extinction.” This webinar has been sponsored by Sable Systems International, so a big thank you to them for helping to make this event possible. Joining us today, we’re fortunate to have Dr. Caroline Williams, an associate professor at UC Berkeley, and Dr. Eric Riddell, an assistant professor at Iowa State University. Their presentations will discuss climate change and the ecophysiological effects of changing global temperatures on organismal biology and survival. All right. And with that, I’m very pleased to welcome Dr. Caroline Williams. Caroline, thanks so much for joining us today. And the floor is yours whenever you’re ready.

Dr. Williams: Great, thank you. I’m glad to be here. My name is Caroline. I’m a professor at University of California at Berkeley, which is the land of the Ohlone tribe. My lab works on the winter climate change and one of the reasons we focus on winter is because winter is actually, winter climate is changing faster than summer. So, this shows the degrees of warming that we’d expect over the next hundred years, and you can see that some of these dark red colors indicate up to sort of six degrees of warming, which is much faster than we’re seeing in summer over the corresponding period. And these changes are having really widespread impacts on the distributions and abundance of animals. So, all of these little pictures show organisms that are moving northwards in North America from tropical and temperate regions due to the increases in winter temperatures, temperatures, just allowing them to expand their distribution.

And another thing that winter climate change is doing is decreasing the snowpack. So, the picture on the left here shows the amount of decrease in the spring snowpack between 1980 to 2020, and the darker brown colors show these really widespread decreases in spring snowpack due to warming temperatures in spring. And this is particularly pronounced in the Sierra Nevada’s, which is one of our study sites. So, you can see here these red circles indicate decreases in snowpack. Blue would indicate increases. And this region here is the Sierra Nevada Mountain range in California where we work. So, the Sierra Nevada’s are really striking in their interannual fluctuation in snow cover. So, the picture on the left, the middle here, at the left of the pair of pictures shows the conditions in a wet year on June 22nd. And on the right, a dry year, just one year earlier on the same day of the year. And you can see in the wet year, there’s still a lot of snow on the ground. You can see the willows in the foreground have not leafed out, so the phenology, the timing of the season is very different, much later in a wet year than a dry year. And the plants and animals in the Sierra have evolved under these big fluctuations between years, between wet years and dry years.

Many ectotherms overwinter beneath the snow, all sorts of organisms from invertebrates to vertebrates, many important components of the terrestrial ecosystems, including pollinators and organisms that form the base of the terrestrial food web, and so the snow is going to really determine the conditions in the ground for these multitude of organisms that spend their winter beneath the snow pack. And the snow really impacts the soil microclimate. Here we’ve got a figure showing the snow depth at one of our sites in the Sierras, and you can see this blue period shows a big snowfall at this point, and the snow is, through that blue period, buffering the soil through that deep insulating layer. And when we look at the soil temperature profile underneath that layer of snow, when the snow is not there, the temperatures are very variable, and you can see the soil will be exposed to cold snaps. This one’s down to just below minus 5 degrees, it can get even colder. And when the snow is present, the temperatures are very stable and right around zero or one degree Celsius. So, snow really strongly modifies the soil microclimate.

And so, this leads to this maxim, colder soils in a warmer world, that when you lose that insulating snow cover, the soil underneath counterintuitively becomes colder. So, one of the consequences of winter climate change can slightly counterintuitively be colder soils rather than warmer soils. So, for ectotherms or cold-blooded organisms in winter, two of the main sources of mortality are cold mortality caused by injuries that organisms incur if they go below a lethal temperature threshold, but also energy depletion. So many overwintering organisms overwinter in dormancy, so they have a fixed amount of energy to get them through the winter. For all ectotherms, metabolic rate is temperature dependent, so any increase in temperature will increase the rate of depletion of those stored energy reserves and can cause mortality or can also lead to decreased energy being available for reproduction in the spring. And these, you can see these two sources of mortality are altered differently by a change in temperature. So, if it gets colder, cold mortality will increase. On the other hand, if it gets warmer, potentially energy depletion can increase and lead to decreases in fitness.

So, when we go back to this maxim of colder soils in a warmer world, a loss of snow cover we might expect to increase cold stress but decrease energy stress. But all of these responses are going to depend on the traits of the organism in question. So, one of the important traits is the metabolic rate. And so, here’s an example of how metabolic rate typically increases with increasing temperature, even in dormant organisms which have an overall suppressed metabolic rate. And the sort of shape and position of this temperature metabolic rate curve can be thought of as an organismal trait that will determine energy use. On the other hand, when we think about cold tolerance often for insects and other ectotherms, cold tolerance responses are really nonlinear; that everything will be fine up until a threshold is reached and then mortality will accumulate very quickly survival will drop. So, we often measure cold tolerance by exposing an organism to, say, a one-hour exposure to a given temperature and measuring mortality. And we talk about the LT50 as the lethal temperature at which 50% of individuals will die from that exposure. And so, this is another sort of organismal trait that’s going to determine the response to temperatures in the soil. And, of course, if the exposure time is longer, then the temperature that can be tolerated will be less extreme, so we always need to think about the interaction between the time that they’re exposed and the temperature that they can tolerate.

So, we work in the Sierras, which obviously is a large mountain range, and as we move up mountain, the air temperature decreases, but conversely, the snow cover increases, and so we’ll buffer the soils from those cold air temperatures to a greater degree at high elevations. And given those two interacting gradients in air temperature and snow cover, we really don’t know, going into this study, how cold mortality and energy depletion might differ along this elevational gradient. And you can see if we base our predictions of responses to climate change on air temperature alone, we might get very misleading predictions. So, we don’t know how thermal stress on overwintering ectotherms that live in the soil layer in snowy mountains changes along this elevational gradient.

So that was the purpose of the study that I’m going to tell you about today. To answer the question “How does changing snow cover impact cold and energy stress for organisms that overwinter in the soil along these snowy mountain elevational gradients?”

So, the hero of our story today is this cute little beetle, the Chrysomela aeneicollis. It’s a willow leaf beetle. It’s about the size of a lady beetle, and it’s really an extremely winter-adapted organism. So, it spends eight months of its 12-month life cycle overwintering in the soil at the base of the willow plants at about sort of six centimeters depth from what we know. And then it has just a few months to complete its life cycle in that short, high elevation growing season. And this beetle is a cold-adapted Arctic relic species. Our study sites here in the Sierra Nevada’s are at the southern range edge of the species range, so they’re really pushed to the edge of their physiological tolerance at these sites. And the reason why I moved into this system and started working on them when I started my faculty position at UC Berkeley is because there’s this really rich history of study on this particular beetle and this series of populations in the Sierras that makes them a good sort of canary in the coal mine, if you like, for responses to climate change in these sort of range-edge populations. So, my collaborators Nathan Rank and Elizabeth Dalhoff have worked on these beetles for many years. There are microclimate monitoring stations distributed across these mountains. You can see these five different elevation gradients that Nathan and Elizabeth have monitored. This is over about a 40-mile stretch of the Sierra Nevada’s just out of Bishop, California, the eastern Sierras. And Nathan and Elizabeth have these amazing long-term data sets on the beetle abundance from surveys that they run every summer. They’re out there right now. and a lot of detailed information on the summer physiology and genetics. So, Kevin Roberts in this picture is a PhD student in my lab and for his PhD work he’s been really filling in the gaps on the winter side of this system.

So, Kevin has been combining the environment temperatures that we get from the data loggers through this time series that gives us the interannual variation in snow cover and the spatial variation along the elevational gradients. We’ve also got a lot of information from the lab on the physiology and limits of these beetles. So, Kevin has measured metabolic rate temperature curves. We have done some measurements ourselves on the cold tolerance limits, as well as using published data from Brent Sinclair’s lab on the tolerance of these beetles to both acute and chronic cold exposures. And we do a lot of energy reserve assays to sort of measure the energy stores through time. We then have the field data, so from beetles in the field, the phenology of the natural population, so the timing of the spring emergence, and we do snow manipulation experiments in the field where Kevin’s been digging holes either sheltered beneath a shed or out in the open where snow can accumulate naturally and measuring the response of the beetles in these semi-natural conditions.

So, we take all these different sources of data, and we use these to build ecophysiological models of the energy and cold stress that the beetles are under, and we then take the results of these models, and we test them both in the lab and the field. So I’m going to show you some of these data today.

So, in terms of the environment, we have these five replicated elevational gradients shown in blue, yellow, purple, or green and red. These are all monitored by microclimate loggers in the soil, which gives us hourly soil temperatures from 2009 until the present in these five replicate elevation transects. And this is what the nine years in our dataset that we’re using look like in terms of the days of snow cover. So we’ve separated these into sort of the three snowiest and the three driest years, and that’s what I’m going to show you today, and we’re going to have snowy always in red and because snow is kind of warmer as we thought of it going in at least, and the dry years will be shown in blue. So, we’ve got those two sorts of aspects of variation the spatial variation along the elevation transects and the inter-annual variation between these snowy and dry years, you know, like those pictures I showed you earlier of the wet and dry years. And this sort of difference between snowy and dry years, we can think of as sort of simulation of what’s going to happen with climate change as droughts become more frequent and more extreme in the Sierras. And we’re actually just coming out of an extremely dry year in the Sierras this year with very little snowpack. Some of the lakes are quite dry. And that’s going to be our reality in the years to come. So, we can use this beetle as a sort of example of how these changing conditions might impact insects and other ectotherms.

So, the first thing we found is that the cold exposure is much higher in the soil in these dry years in blue than it is in snowy years in red. And across the elevation gradient, the cold exposure is highest at mid-elevation. So, you can see in the middle elevations, there’s these really high blue points that show that the soils are getting much colder at these mid-elevation sites than they are either at high or low elevation. So that was really interesting to us because it shows that in these high elevation sites, the persistent snow cover, even in dry years, sort of buffers the soil from these extremely low air temperatures. So, we found that really interesting because it suggests that the snow is actually altering the elevational gradients in cold exposure, rather than it getting sort of colder and colder as we might expect if we looked at air temperature. There’s actually this humped relationship and that’s exacerbated in drought years when the snow cover is so much lower that it’s not buffering the ground.

So, the next thing that Kevin did is he constructed an ecophysiological energy use model. So, he took these metabolic rate temperature curves that he measured in the lab using Sable Systems respirometry equipment. So, he used a FoxBox and did stop-flow respiratory to measure the metabolic response of these beetles to a range of temperatures that cover the range of temperatures that they experience during overwintering. He then input the environmental data, the microclimate data. So, these are the soil temperatures from the data loggers which we can use as a proxy for beetle body temperatures during the winter. And we input those microclimate temperatures into that temperature metabolic rate relationship to give us oxygen consumption over every hour of those nine winters that I showed you. And this is showing the resulting transformed data, which is our oxygen consumption per hour throughout the whole winter from sort of whatever date range you want to use. And then the next step is you select a start date and an end date and sum the oxygen consumptions between that start and end date. And we can run the whole range of start and end dates that we observe in the field to estimate whole winter energy consumption of those beetles. And then we take the beetles out into the field.

Kevin has performed many now of these snow manipulation experiments, so we convert those oxygen consumptions into predicted lipid used, using stoichiometric equivalents of assuming two liters oxygen for every one gram of lipid. And we then compare that to what the beetles really use in the field. So, you can see here some data from a snow manipulation experiment where snow was present or excluded. We measured the amount of lipid that the beetles from these two conditions had at the end of winter. The first line shows the actual measured lipid amounts. The second line shows the model estimates of how much lipid these beetles should have had according to our model. And the bottom line shows the model accuracies are actually really high; that these models are predicting with greater than 95% accuracy the energy use of real beetles in snowy and dry conditions. So, we’ve been quite satisfied with the performance of these models. Kevin’s also done a really extensive lab validation of these energy use models. I’ll only sort of touch on this very briefly, but I’m happy to talk more about it, and Kevin is preparing this paper for publication now, but you can see in both constant temperatures that simulate conditions underneath the snowpack and in variable conditions that sort of fluctuate in a similar way to snow-free soil, we can really accurately recapitulate, you can see these colored lines here, the energy use of beetles throughout the winter compared to the empirically measured lipid results in grey. So, the models are very accurate under a range of lab and field conditions.

And so, when we use these models to predict total winter energy use across elevation, you can see that generally, energy use declines with increasing elevation. And that decline is actually steeper in snowy years than it is in dry years, which is due to a very high energy use at these low elevation sites. So, if you look at the red line, you can see that at low elevations on the left of the plot, the energy use is quite high there, and that’s because both the soil temperatures are relatively warm and also the winters are longer in these snowy years. If you remember the two pictures of a wet and a dry year that I showed you on June 22nd in the wet year, the snow was still on the ground and so the beetles would still have been in dormancy at that time. So that sort of accounts for a big portion of the higher energy use in the snowy years.

So, you can see that the snow is going to impact the elevational gradients and energy use as well as in cold stress, and that snowy years tend to be longer and more energetically demanding. And again, we can see this pattern that the energy use sort of converges between dry and snowy years at these high elevation sites, and that’s again because even in dry years, the snow is still persistent at those high elevation sites, which reduces the interannual variation.

So, these are the same data that I just showed you, just plotted out in a slightly different way. So, you can see on the left, the cold exposure between snowy and dry years, cold exposure will increase in dry years as we move towards increased drought, particularly at these mid and high elevation sites like this site, Long Lake at Bishop Creek.  It will experience an increase in cold exposure, whereas in the right-hand figure you can see that total winter energy expenditure will decrease as we move from more snowy to dry conditions, and that those decreases will be particularly pronounced at the low and low-mid elevation sites because of the big impact of snow on energy use at those low elevations.

So now in order to sort of understand the organismal consequences of these changes in the soil conditions, we need to compare these data to the thresholds that the beetles have. So, I talked about the LT50 or the lethal temperature at which 50% of individuals die. So, we wanted to use this sort of measurement from the lab to compare the microclimate data to the beetle tolerance thresholds. So, what you can see here is that the minimum winter temperature actually never falls below that LT50 for a one-hour exposure. The beetles are very cold hardy. They can survive minus 15 for an hour in the overwintering stage, and we never see temperatures in the soil that approach that threshold. However, the beetles also die if they are frozen below their freezing point for more than 12 hours. So, this species is freeze-tolerant, but they can’t be frozen indefinitely. So, when we look at the longest period below freezing in the right-hand plot, we can see that in many sites the longest time that they’re below that freezing point is exceeded at many of the sites, particularly mid-elevation sites. So, we would expect that many of these beetles would die from just being frozen in the soil for too long.

So, we can use these data to calculate the probability of experiencing lethal cold for a population. And you can see this hump-shaped relationship again, that at the mid-elevation, mid-high elevation sites, there’s an almost 50% chance of populations experiencing a lethal cold event in a given winter. But in our beetles, that probability of experiencing a lethal cold didn’t change with interannual variation and snow cover. For some species, that might. That will depend on that organismal threshold. We can calculate probability of mortality from energy depletion by comparing the energy use data to the amount of energy that a typical beetle starts the winter with. So, if the predicted energy use exceeds what we would expect the beetles to have on board, we would say that they were likely to have died from running out of energy. And this is what I’m plotting on the right here, the probability of energy depletion as a function of elevation. And you can see that in snowy years, particularly at low elevation, the probability of energy depletion is very high, that the beetles are often predicted to use more energy than we would expect they have. So probably a large amount of energy stress in these low elevation populations.

So, to wrap up, we think that what we’re seeing is that there’s a trade-off between cold mortality and energy depletion that’s modulated by this inter-annual variation in snow cover, where snowy years are less cold but a higher probability of energy depletion, and along the elevational gradient, the snow really alters the cold and energy stress. So, as you move up in elevation, the high elevations are buffered due to that persistent snow cover, and the mid-elevations will experience an increased cold stress with drought because they’re not always buffered by that persistent snow cover.

So together, the environment and the organismal traits will determine the stress exposure. And if we can use ecophysiological models to sort of predict the impact of the environment on particular organisms, we might be able to form some sort of general principles of how winter climate change will alter stress exposure. So, the big picture take-home message is that climate change is not so simple as global warming. This maximum of colder soils in a warmer world is really useful. And along environmental gradients like elevation gradients and also latitudinal gradients, we really need to account for the impact of snow cover in determining responses to climate change because it can really change our predictions relative to air temperature.

So, thank you very much for listening, and I’ll be glad to take any questions.

Sarah: Great. Thank you so much for that presentation, Caroline. We really appreciate it. Eric, are you with us? I believe so. Perfect. OK. So, everyone, this is Dr. Eric Riddell. We’re very excited to have him with us. And, Eric, the floor is yours whenever you’re ready.

Dr. Ridell: Great. Thank you so much. So, I’m Eric. I’m an assistant professor at Iowa State University, which sits on the ancestral land of the Sauk and Meskwaki nations. And today what I want to talk about are some of the really fundamental questions that my lab tries to answer. And in order to do that, we’re going to go to two very different ecosystems. One, the temperate rainforest of Southeastern North America, as well as the deserts of the Southwest and North America. So, two very, very different regions we’re going to talk about, and we’re going to talk about two very different types of organisms. But the guiding philosophy that I have is really the same. And one of the big questions that I try to ask is, how does physiology determine the niche? To really answer this question, we need to think across large spatial scales and understand how the environment varies, as well as how organisms actually experience that environmental variation and how physiological performance changes in response to different environments.

To answer these questions, I rely on this field called ecophysiology, which I define very loosely as understanding how organisms function in the context of their environment. And when I say function, we can talk about function in all sorts of ways. We can think about it at the whole organismal level, at the organ level, tissue, cell, and all the way down to the genes. And the way that I really like to think about function, though, is really thinking about integrating across the hierarchy of biological organization. So, understanding how the combinations of all of these different types of higher, the levels in the hierarchy ultimately produce phenotypes that we’re interested in measuring in nature. But that’s only half of the puzzle piece, right? So, the other half that we need to think about is the environment, which we know varies across space and time. Temperatures, for instance, vary with time of day, as the sun moves through the sky, as well as with elevation and latitude, and as a consequence of that temperature variation, we also see variation in humidity.

So, organisms living on this landscape experience that environmental variation, and as a consequence of the biological hierarchy of organization, have specific responses and constraints that ultimately determine where we find individuals across the landscape. And that gives us a sense of the organism’s ecological niche, and more specifically, the geographic range of where we might actually find these organisms.

But we really have a fundamental problem in trying to link these two disparate pieces here of the organism and the environment. And that’s that they’re really collected at different scales, one at the level of the organism and the other at often it can be very broad global continent-wide scales. To solve this problem, I use biophysics, which helps us link physiology to the local environment through a series of calculations based in physics that help us downscale broad climatic variation to the level of the organism and what they’re actually experiencing, so that we can predict how performance varies in environmental conditions where these animals live. So, biophysics for my lab plays this really integral role in linking these two components.

 I began asking this question with a really special salamander for me. This is the southern gray-cheeked salamander. They’re super abundant and they’re pretty cute, obviously. And some of the other traits that I think make them really interesting is that they’re fossorial, so they live pretty much completely underground, except for when they surface, when they’re active on the forest floor to forage. They’re nocturnal, so they only come out at night to forage and find mates. They’re fully terrestrial, so they don’t reproduce in water like a lot of other amphibians, and they’re lungless, so they actually don’t develop lungs. They don’t have any lungs, so they have to breathe almost entirely across their skin. So, they’re basically like a walking lung. Now, as a consequence of being this walking lung, you need to have very thin skin so that gas exchange can actually occur. And that skin needs to be wet, again, to promote gas exchange. And as a consequence of not having lungs and having this thin, wet skin, that influences where we can find these salamanders.

So, the global hotspot of salamander diversity is in the temperate forests of the Southeastern North America in the Appalachian Mountains. So, what we’re looking at here is a heat map of salamander species and the species range – that’s in black outlined there – that is depicting the species of interest here that I focused on, really a species complex. And these mountains are pretty cool. They’re very wet and it’s a perfect environment for an organism that is highly susceptible to desiccation from high water loss rates. So, it’s a very cool, very moist environment and it really helps them avoid drying out. And so, one of the ways that we study how quickly they dry out is by using flow-through respirometry. So, we use really tightly controlled combinations of temperature and humidity, and we put these organisms into these environmentally controlled chambers and monitor the amount of water vapor and that’s being added to the air stream, as well as how much oxygen they’re consuming in order to really understand metabolic rates and energy use, as well as water loss rates to get at how quickly they dehydrate.

And so, once we have this type of data, we can start building ecophysiological models that help us predict the geographic range of these organisms. So, we start with biophysics that helps us really downscale that environment, and we incorporate these animals’ physiology, which we measure in the lab, to predict the amount of time that the organism can be active across space, to estimate how much energy they can bring in through foraging for finding insects. They’ll eat almost anything, even other salamanders, which also helps us estimate how much energy they have to allocate towards reproduction, which is our surrogate here of fitness. And once we can do this, we can do this all across the landscape here, and we can identify regions where they’re capable of reproducing. And then look at the geographic distribution of presences where we actually find these individuals to see how well our model did.

So, this is an example of one of the first sort of species distribution models – ecophysiological models – that I generated for the salamander. What you’ll see here are two maps. The one on the left is for activity, the average amount of activity on a particular night in July in the Appalachian Mountains. And then on the right is similar, except it’s for net energy balance. So, if you look at activity time, you’ll see that the regions in the very bright colors are associated with high amounts of activity. And we can see here that each one of these tiny little white coordinates here, points, is a capture. So, we see that those capture locations are concentrated in areas that the model predicts should be high for salamander activity. Similarly, we see that those same points are associated with high energy balance. So, areas in green and yellow, these really bright colors, are capable of supporting the salamander’s biennial strategy of reproduction. And 83% of these points are estimated to be capable of reproducing.

One thing you’ll notice is that the model is actually predicting some areas where we don’t find this particular species up here in the top right corner, that other bright color where there aren’t any points. Well, it so happens that there are salamanders there. They just happen to be a very closely related species. They can actually interbreed with the species that I’m studying. So, what you’re noticing here or seeing is that this is a general salamander model that sort of predicts suitable habitat for all species with similar physiology, I would add.

So, this framework really helps us understand how we can integrate across the organism to environment relationship to produce fitness across landscapes. Now you can imagine why that might be helpful for answering something like climate change because we know that environmental conditions are not expected to be the same. Not only do they fluctuate on a daily and seasonal basis, but over the next century they’re expected to become much warmer due to climate change, generally.

So, part of the goal that I wanted, the next question that I wanted to ask is how climate change might actually influence these patterns of energy balance and possible extinction across the landscape, with the specific question being, will climate change threaten salamanders with extinction? Now, just like the environment isn’t predicted to change or is predicted to change, we also know that animals can change as well. With fluctuating conditions, animals can adjust metabolic rates to burn less energy or adjust water loss rates to lose less water in a process called phenotypic plasticity, which I define here as the expression of multiple phenotypes from a single genotype.

So, one of the first questions that I sought out to answer was how plastic these salamanders were. So how flexible were they in water loss rates and metabolic rates? To answer this question, I deployed a bunch of environmental data loggers across the landscape where I was studying these salamanders to look at how humidity, which here is a measure of vapor pressure deficit, and temperature changed. So, vapor pressure deficit here is basically as it gets closer to zero, the air is becoming more saturated. So over time, on the x-axis here is from spring to fall. So as this season progressed here it became wetter and wetter in this particular year. And temperature over that same time period exhibited that very characteristic sort of, you know, hump shape where these warmer temperatures in the middle of the summer and cooler temperatures in the spring and the fall. So physiologically what we saw, what was happening is that animals were, these salamanders, were reducing their resistance to water loss. And resistance to water loss is the inverse of water loss rate. So, the higher the water loss rate, the lower the resistance and vice versa. So as the air was becoming wetter, these salamanders were becoming leakier, if you will. Water loss rates were going up or resistance was going down. For oxygen consumption, we found a similar type of response where in the hottest part of the year, particularly in the summer, we found the lowest metabolic rates measured at one particular temperature, suggesting that these salamanders were shifting how much energy they were burning at any particular temperature. And roughly these coincided with about a 25 to 40 percent change in physiological rates depending on the rate that we’re talking about here, so a lot of flexibility.

So, the next stage that we wanted to do was integrate this plasticity to see how much it mattered for predictions of extinction risk to climate change. So, we started with estimates of climate change, which we downscaled to the microclimates that salamanders actually live. We combined that with the physiology we measured in the lab, including that plasticity, with biophysics to calculate activity, to estimate how many insects they could eat, and ultimately allocate towards reproducing eggs, reproduction and egg production. We also added another component, which was extinction risk, which we estimated by whether or not the salamander burned through all of their lipid reserves at any given point throughout the time of the year. So given a particular metabolic rate, if the salamander burns through all their fat, we consider that location to be functionally extinct or locally extirpated.

So here you’re looking at a much bigger map of the southern Appalachian Mountains with the species range here in white. And then inside of the species range, we have the ecophysiological model illustrating net energy balance throughout the range. So, what you’re looking at here is energy balance and extinction without plasticity. Regions here in yellow and green are capable of reproduction. Regions in blue are not capable of reproduction, are actually negative energy balance. And anything in grayscale, which is the vast majority of the map, is predicted to be locally extirpated. Now, if we fast forward to the end of the century without plasticity, without this flexibility, what we see is that pretty much all of the salamander suitable habitat is gone by the end of the century. So, it doesn’t look good for the salamanders without plasticity.

But let’s go back in time and actually add in some of this flexibility that we know exists. So, this is without plasticity. And if we incorporate that plasticity, we see a really large increase in the number of sites that are capable of reproduction throughout the species range. Roughly around 85 percent are capable of reproduction. And as we fast forward to the end of the century, what we see is a contraction of the species’ range, but the core of the range is still maintained, even under the worst-case warming scenario, which is being depicted here. So, their core of the range maintains enough energy to reproduce by the end of the year if we incorporate the flexibility that they’re exhibiting today. We don’t need to, you know, even without adaptation, the salamanders that live today have a high degree of flexibility that could be beneficial in the future.

I became really interested in these regions that were sort of predicted to change, given the plasticity of a particular organism at that location, and really wanted to know across the biological hierarchy of organization, what was driving these changes in habitat suitability at a genetic level? So, what are the genes that are actually responsible for this flexibility? To answer this question, we conducted a weighted gene co-expression network analysis, as well as some other differential gene expression analyses, to draw associations with the plasticity and resistance to water loss. And one of the things that we found here, so this is the resistance in water loss on the x-axis here, our estimate of flexibility. One of the genes that popped out was this gene that’s responsible for encoding ceramides in the skin, which are lipids. And when you blast these lipids to understand their function, their primary function is helping the skin resist water loss. So, this was really cool, because we actually found, we identified this particular gene that was associated with greater flexibility in water loss rates.

We also found suites of other genes that were involved in vasoconstriction, as well as angiogenesis, that play really important roles in regulating how flexible these particular organisms were in losing water. But ultimately what this gave us was a really comprehensive understanding of how the niche was really structured for these organisms across space and helped us to predict extinction risk throughout their species range and what actually might happen under climate change.

So that gave us this framework for asking questions about the future. But one of the big questions that I had was, well, do we know that this even works? Is physiology a good framework for actually predicting how communities and populations respond to climate change? More specifically, can physiology predict responses to climate change? We can forecast all day. But what really matters is we want to know that it works. So, this took me to a really different ecosystem. This was in the Mojave Desert in the Southern region of California. And as you can tell, this is not a temperate rainforest anymore. It does not receive very much precipitation and temperatures can become extremely warm, particularly during the summer. But despite these really extreme conditions, we have around 140 species of birds that we studied. in addition to around 30 to 40 species of small rodents. And one of the questions that we were really interested in, this was during my postdoc at UC Berkeley and Steve Bisinger’s lab, was we were really interested in how these two different organisms responded to climate change, these birds and desert rodents.

This story actually really begins over 100 years ago with Joseph Grinnell, who was one of the first directors of the Museum of Vertebrate Zoology at UC Berkeley. And what he did is he went out into California to dozens and dozens and dozens of sites and took impeccably detailed notes on all of the birds and small mammals that he saw during those surveys along with his team. So, you can see him taking notes up there in the top left image. And this is a band of his motley crew during that time period, looking for all of these organisms. And his notes were so detailed that, you know, he recorded everything, what they had for breakfast and even, you know, what those conditions were like. The wind came up again about nine o’clock and has blown disgustingly all day. So really incredible detail that allowed us to reconstruct not only the experience, but what these communities looked like more than a hundred years ago.

So fast forward to contemporary times, my postdoc advisor Steve Bisinger and his colleagues returned to these sites to actually tell, you know, which species were still there, and which ones were not there anymore. These were conducted through a series of surveys by many, many folks that I’ll acknowledge at the end here and a really incredible effort of resampling Grinnell’s historic sites in in these protected areas, particularly in Death Valley, the Mojave and Joshua tree. One thing that we sort of were curious about is understanding how these organisms, these two different organisms, desert rodents and birds, how their exposure really changed over the last century. Temperatures, air temperatures had warmed by about two degrees Celsius over the last 100 years. And we were looking at birds and small mammals all captured in the essentially exact same locations. So, it was a really great opportunity to see how different communities might be responding to the exact same type of environmental change.

So, what did we find? Essentially, what we found was that we lost many birds and the small mammals stayed, their whole community stayed remarkably stable. So, in the top left here, we’re looking at the change in species richness. And you can see that the number of species that we lost for birds in red was, you know, much higher, roughly around 20 to 15 species on average. And we actually lost around 43 percent of the desert birds and essentially none of the small mammals. Similarly, if we look at change in occupancy, which describes the probability that a particular species occupies a site, we see that the birds are shifted over to the left, indicating a reduction in occupancy values, whereas these small mammals are sitting quite squarely on zero.

So, one of the big questions, well, what actually explains this? Why did the birds do so poorly, and the small rodents actually appear to be stable during this time period in the same habitat with similar amounts of climate change? To answer this question, I built a biophysical model that’s capable of simulating the physiological experience of these two desert animals, these birds and desert rodents. And it takes into account all sorts of radiation and inputs to the heat load, as well as microhabitat use and specific behaviors that are used by these particular animals. It simulates the day and the nighttime over these conditions, and we ran these simulations over the last hundred years to see how their physiological experience changed over the last century. And what this model is really doing is estimating the amount of energy that they need to burn to stay warm or evaporate to stay cool, very similar to how you can think about it in the same way of how you pay your bill to heat your house. When it gets really cold, you’ve got to increase the amount of energy to heat it. And when it gets really hot, you need to increase energy as well to keep it cool.

So, what you’re looking at here is a landscape that’s sort of illustrating the different physiological experiences that birds and small mammals have had over the last century. And these are, in particular, the change in cooling costs in kilojoules per day that a bird has experienced over the Mojave Desert. So, this dotted line here is the boundary of Death Valley. And so, for the bird and the small mammal. And what you’ll see is that birds have experienced a substantially higher cooling costs indicated by these red temperatures relative to these small mammals, where there are some locations for small mammals that have experienced high cooling costs, but they’re pretty isolated and associated with really shallow soil depths where it was difficult to hide from the warming temperatures. And if we look at distributions of the change in cooling costs and cooling costs in general under contemporary climates, we see that for most small mammals across most sites, cooling costs are essentially zero, that they’re able to avoid these really warm temperatures that require a high degree of evaporative cooling in order to keep body temperatures stable. Whereas for birds, the amount of cooling that was required increased by about 30 to 40% over the last century.

What was also interesting was that the species-specific cooling costs for birds was associated with the degree of decline for each species, suggesting this physiological basis for decline in birds. But what was sort of the final result and really interesting was that there was this interaction specifically with their ecology and whether or not they were more animal-based or plant-based diets. And that’s typically because these birds with animal-based diets get all of their water from their food. So, they actually need to go out and expend more water and more energy to find their food and take their water, whereas with birds, the plant-based diets are known to regularly drink and stick closer to desert oases. So not only do you have to take into account their physiology, but their ecology and life history is also really critically important to predicting the impact of climate change.

So I hope now that you can understand how we can use the same framework, no matter what type of organism it is, whether a bird or a salamander, to really understand how these organisms experience climate change in very different ways and what that means for not only population dynamics, but ultimately their ecological niche. So, with that, I’d like to thank the following people and take any questions.

Sarah: Great, thanks so much for that presentation, Eric. And I’m going to bring back Caroline. There we go, you’re both with us, great. Okay so we are going to kick off this Q&A with our very first question. This question is for you Caroline. Does burrowing depth vary with elevation?

Dr. Williams: Yeah, that’s a great question. So, the short answer is in our system we really don’t know because it’s very, very hard to find these tiny little beetles. So, Kevin has spent many hours, and Nathan as well, I’m sure, digging around in the soil, trying to find the beetles, and it’s just very hard to find them. So, we have pretty limited information from one experiment that a collaborator, John Smiley, did where he sort of caged beetles around a willow and found them subsequently in the soil, which is our only information on where they overwinter. But more broadly, I think that is a really interesting question. Ray Huey and Mike Kearney have been doing some work on this question showing that you know the deeper you burrow, the more you’re protected from extreme cold temperatures. So, we would expect that in colder habitats, say you know at dry years or elevations where it’s particularly cold, burrowing deeper would be an effective way to escape the cold. So, I think that’s a wide-open question. I’m not sure if the beetles are the system to address it in since it’s so hard to find them, but definitely important

Sarah: Great Okay, this next question is for you Eric. This question is what would be the advantage of the salamanders becoming leakier during wet? Periods, why not just be non-leaky year-round?

Dr. Riddell: Right. This is a really great question, and I asked the same one myself And I think it’s important to remember that salamanders are a walking lung, and even for you as a human, in order to breathe, your lung actually needs to be wet have moisture on it. So that oxygen can dissolve into the moisture and then diffuse into your bloodstream. So that moisture on the skin is really important for being able to breathe and we actually find this really, really tight association with the skin’s resistance to water loss and metabolic rate. And specifically, as organisms become more watertight and less leaky, we see a simultaneous decline in their ability to breathe and respire metabolically. So now, you know, okay, well, then why don’t you just like keep a low metabolic rate too, because then, you know, you’re not burning anything. Well, that’s also a big problem if you want to do things like defend your territory and reproduce. If you actually want to get work done, not being able to breathe is not a good strategy for doing work.

Sarah: Yeah, no kidding. Awesome. All right. This next question is for you, Caroline. This question is, do you see any evidence for evolutionary shifts in thermotolerance within this species?

Dr. Williams: Yeah. So, Nathan and Elizabeth have done a lot of work finding some genes that are impacted in lots of aspects of thermal tolerance. So, heat tolerance, even cold tolerance, some of these genes of central metabolism. So particularly phosphoglucose isomerase or PGI, succinate dehydrogenase. There’s lots of linkages between genes and these important metabolic and thermal tolerance phenotypes. And these genes sort of change over latitude and they change over elevation, and they change over the season. And so, we’re sort of, our next step is to start to understand how these genes impact winter survival. But we really think that there is genetic variation that determines the response and that that variation might be under differential selection in snowy years and dry years. So that’s one of the hypotheses that we’re testing at the moment. But we do think that there is going to be the potential for evolutionary change in these traits. But if the snow is sort of creating these inter-annual Fluctuations that might actually maintain some of this genetic variation in the population.

Sarah: Okay, so next question here Eric this question is for you This question is aren’t birds also known to upregulate ceramides and other waterproofing molecules when subjected to dry conditions, similar to what you would find with the salamanders?

Dr. Riddell: Yes. And if I were smarter, I would have predicted that before I started the gene expression studies. But I think when I first got started with amphibians, one of the things that everybody talks about with amphibians is aquaporins. And so, I was like, oh, maybe it’s aquaporins, how they’re delivering water to different tissues, but I kind of just got, you know, whacked over the head with the ceramide thing. I was like, wow, that totally makes sense. I didn’t realize that it would occur over such a short time scale. So, for salamanders, we saw this change in ceramide production in about three weeks. From my knowledge of the literature in birds, I think a lot of those comparisons come from seasonal changes in ceramide production, so comparing something like winter to summer, which is, of course, takes place over a longer period of time. So, to see not only this potentially conserved mechanism of waterproofing occur in salamanders as well as birds, and there’s also other evidence in rodents as well, that ceramides in the skin are important, but to see it happen over such a short period of time was really exciting as well.

Sarah: Great. Okay. This next question is for you, Caroline. This person has asked, did your team collect data on humidity gradients?

Dr. Williams: Is that for me or for Eric? So, we did not. Yeah. I wonder if that was for you, Eric.

Sarah: It came in during your presentation. So, I’m assuming it was for you, Caroline, but I mean, Eric, you’re very welcome to answer.

Dr. Riddell: Caroline, I don’t know, do you want to say anything? Go ahead.

Dr. Williams: Yeah, maybe, so the snow, in addition to impacting the cold exposure and energy use, one of the major things about snow is it’s really wet. So indeed, we’re only looking at sort of the thermal properties of snow, but there are all of these other hydric properties of snow which are going to be really important. And we do see, so we haven’t measured the humidity gradients, but we have seen different impacts of like the beetles and the dry plots will actually literally dry out in some cases. So that’s a whole other aspect of snow that we haven’t looked at in detail but is probably going to be important both for the beetles and for the host plant as well, because the willows, after a snowy year, there’s plenty of snow melt. They’re very well hydrated. The plants come out in very good quality, so there will be sort of flow-on effects of the host plant quality in snowy compared to dry years. So, yeah, definitely the water content of the snow is an important aspect that we’re interested in in the future.

Sarah: Great. Eric, this question is for you. Do you think your results showing a difference in energy expenditure in response to climate change between birds and mammals might explain why dinosaurs went extinct, whereas mammals survived the KT extinction event?

Dr. Riddell: Yeah, that’s a really cool question and idea. I think, you know, it is kind of funny too, because birds are our modern-day dinosaurs here. That’s all we have left, and we still have our small mammals around. So, I think it seems pretty logical to make that argument. But one of the things I would just focus on is really understanding how beneficial it is to live a fossorial life, to live underground. And these certain conditions, as Caroline talked about too, it’s a completely different environment that helps buffer you in certain cases from environmental variation. And the small mammals, you know, have done pretty well so far over the last century. And even the salamanders as well that live these really cryptic underground lifestyles, you know, there’s a sort of, you know, there is some evidence suggesting that salamanders are, you know, doing quite well and can respond well to environmental change as well because of their ability to live underground. So, it’s something that we really need to think a lot about and to say that you know I think that living underground would certainly play some sort of role in small mammals making it through the KT extinction.

Sarah: Okay cool very awesome question and this question is actually for both of you, so you can maybe start first, Caroline. If a very snowy year is followed by early snow in the next year, can that limit the number of animals who make it to adult overwintering?

Dr. Williams: Yeah, so definitely those snowy years when the beetles and other organisms emerge from overwintering later, that means the growing season is squished into a shorter period, so they have less time to accumulate energy and potentially could fail to reach the correct overwintering stage, which in this species is the adult stage, by overwintering. And in other systems, in ground squirrels, in the Arctic, we know that this can actually impact juvenile recruitment. You can see mortality because they’re just not making it through. And this is really interesting, I think, because there could be different genotypes that develop faster. And so, we really need to consider the integration of the whole life cycle, I think, from the wintering stage to the growing season. And that’s one of the things that we’re interested in doing in this beetle system, is sort of resolving the whole life cycle, understanding the consequences of genetic variation across the whole life cycle, including the wintering stages and also the growing season. So yeah, definitely potentially important.

Sarah: Okay, perfect. And I just really wanted to thank you both for being here with us today. It was really a pleasure to have you with us.

Dr. Williams: Thank you.

Dr. Riddell: Thank you.

Sarah: In closing, thank you so much for taking part in this Inside Scientific webinar sponsored by Sable Systems International, so a huge thank you to them as well. And we look forward to having you with us again soon.