Climate Adaptation to Support Amphibian Conservation in U.S. Caribbean

Video Transcript
Download Video
Right-click and save to download

Detailed Description

Climate change will strongly affect tropical island ecosystems. The risks of significant negative impacts are likely to be higher in these island systems than in many temperate regions of the world because of the limited size of their land masses, high numbers of species that only exist in narrowly defined regions, and expectations that tropical environments will experience more impactful changes in temperature. Since 2013 we have developed an ongoing multi-disciplinary and collaborative project aimed at gaining foundational knowledge to build a robust, climate-informed adaptation strategy to enhance the status of 2 endangered species and prevent the listing of 14 species in the genus Eleutherodactylus, commonly known as the “coqui”.

Our focus for the conservation strategy is consideration and evaluation of two classes of adaptation actions:

  1. translocations or introductions and
  2. targeted use of climate refugia.

This work represents a potential ‘gold standard’ of CASC-supported adaptation science. Rather than a ‘top-down’ or ‘bottom-up’ approach to adaptation science, we have consistently engaged in a ‘middle-out’ strategy, enhancing value for resource managers through actionable scientific knowledge and insights, while also leading with new research that is broad enough in scope to be useful to a wide array of stakeholders.


Date Taken:

Length: 00:53:58

Location Taken: US


[0:00] [silence]

Philip Lu:  [0:05] Hi, everyone. Welcome remotely from the US Fish and Wildlife Services’ National Conservation Training Center in Shepherdstown, West Virginia.

[0:14] My name is Philip Lu, and I would like to welcome you to our webinar series held in partnership with the US Geological Survey's National Climate Adaptation Science Center. Today's webinar is titled, Climate Adaptation Science to Support Amphibian Conservation in the US Caribbean.

[0:31] We are excited to have Adam Terando and Jaime Collazo with us today. To introduce our presenters, we have NCASC Chief Scientist Shawn Carter to lead. Shawn, take it away.

Shawn Carter:  [0:43] Welcome, everyone. Thanks for joining us today. It's my privilege to introduce a few good friends and fellow researchers. First is Adam Terando. He is a USGS Research Ecologist focusing on climate impacts with the Southeast Climate Adaptation Science Center. He's a geographer and climatologist by training. He received his PhD in Geography from Penn State for examining the relationship between climate extremes and future crop yields in the US.

[1:13] At the Southeast Climate Adaptation Science Center, his work focuses primarily on the impacts of global change on the ecosystems in the Southeast and the Caribbean and the complex human‑environment relationships that drive these processes.

[1:28] Jaime Collazo is a research wildlife biologist and unit leader at NC State Co-Op unit, Cooperative Fish and Wildlife Research unit. He's a professor in the Department of Applied Ecology at NC State.

[1:41] Jaime has a BS from the University of Puerto Rico and received his PhD from Iowa State University. His teaching interests lie in hierarchical species habitat modeling of tropical avian biology. His research interests focus on demographic processes and species habitat relationships for endangered species conservation.

[2:03] Their talk today is going to touch on all of these topics. Without further ado, I'd like to introduce Adam and Jaime.

Jaime Collazo:  [2:09] Thanks, Shawn. I will start out by giving a little bit of context and also to thank you for promoting me to unit leader. I'm actually the assistant unit leader. I'll take that as an omen.

[2:23] The project started in, Adam and I put this project together, back 2013, '14. The impetus for it were and the motivation was a couple of research and conservation priorities by our stakeholders which are the Puerto Rico Department of Natural Resources and the US Fish and Wildlife Service Region 4 specifically the Boqueron field office.

[2:48] These two research conservation priorities were improved from the projections and application of dynamically downscaled data for conservation. The other one was to provide these agencies, that I just mentioned, with information to guide decision regarding the [indecipherable] species, and preventing the listing of others.

[3:10] Specifically, we will characterize the sensitivity and ability of amphibians, which I'll mention in a minute here, to cope with environmental stresses, to map the genetic structure of seven representative species, identify centers of genetic diversity, and gain some insight from connectivity.

[3:33] Lastly, assist the agencies and development conservation strategies, centered on two potential adaptation strategies. These are, species relocation to new habitat by either translocations or introductions. Identify climate research and how it affects the climate [indecipherable]. Adam and I recognized early on that carrying out this project would take a number of years.

[3:57] Obviously, it is multi‑disciplinary in nature, therefore we assembled a team of collaborators. Many of them are local from the University of Puerto Rico, and others from across the US, NC State, California Academy of Science, Eastern Illinois University. These are in the areas of amphibian ecology, ecophysiology, genetics and decision analysis.

[4:21] Our approach has been to gradually build towards those long‑term objective, but in the process, deliver some products of values for our stakeholders as well. Our focal group will center around 17 species of frogs of the genus Eleutherodactylus, which are commonly known as coqui, 15 of which are endemic. We have a high degree of endemicity.

[4:49] The species are of high conservation priorities for the local [indecipherable] agencies. Three are endangered, and the remaining are at risk of becoming endangered by virtue of either their patchy distribution, low vagility, or  because they're very specialized in micro‑climatic conditions.

[5:10] Their dependence on these micro‑climatic conditions make them a suitable biological model to assess current and predicted consequences induced by environmental change, including climate change. Adam will, for the next few minutes, give you an overview of both of these project components, and how they come together to address the conservation views of these agencies.

[5:35] Although the work is being conducted in Puerto Rico, it'll be obvious to you that methods and applications can be adapted by stakeholders in the US, or somewhere else elsewhere in the tropics, particularly the Caribbean basin. With that, Adam, take it away.

Adam Terando:  [5:53] Thanks, Jaime. Thanks for laying all the background and context there. I'm going to lay down a little bit more context before we get into the meat of what Jaime and I have been working through these past seven years now. Why are we all here? At least in my shop, we're the Southeast Climate Adaptation Science Center.

[6:21] We're in the Anthropocene, or at least I hypothesize, I believe, that we are. Humans are now the dominant drivers of many global processes, of fundamental bio, geo, chemical processes, such as the amount of greenhouse gases in the atmosphere. Here's a figure showing that.

[6:39] Carbon dioxide in the atmosphere is higher than any point in at least the last 800,000 years, probably more like the last three million years. There's a physical response for that, to that change in the radiative balance, this radiative coursing. The Earth is warming. That has all different sorts of consequences.

[6:57] Given that humans are the driving force behind this large change, as a society and as decision‑makers, decision‑makers have to then grapple with that. I think of this in two main ways. One is, there's the decision up top. Should we put less carbon into the atmosphere? That's not really in my purview in terms of what I work on.

[7:28] Down below, there's the other sorts of actions, or decisions, that we have to think about. That's, what actions are needed to accomplish our objectives, given however much climate change we're experiencing now, and that we could experience in the future? In the top, that's the global decision. That's the decision. At the bottom, that's more the adaptation decision.

[7:50] That's the decision space that we operate in, that our science operates in, in terms of helping our stakeholders figure out what climate change means for the things that they manage and have a responsibility for, and how they can manage that risk. Again, both of these decisions, when we think about climate change or global change, it is a risk management type of decision.

[8:15] What is the risk, what are the consequences, and how do we manage that risk going forward? Adaptation becomes a complicated proposition because it's multi‑dimensional, multi‑scale. You're thinking about all sorts of different decision points, and different objectives and values that we have to weigh the trade‑offs against. Science can help in all those aspects.

[8:40] That's where this is coming in. This is a figure from the National Climate Assessment, the recent one. Thinking about adaptation as iterative risk management, this is the gold star, or the vision, of where we would like to go if we could. This is getting through that adaptive management cycle.

[9:03] We're going from awareness to assessment planning, implementation, and monitoring, and evaluation. These inner semicircles here are trying to describe where past climate assessments have tried to go, in terms of how far through that adaptive management cycle we've been able to go.

[9:25] In general, I like to think of this also in terms of where our science tends to operate, whether that's in USGS, or across our partners and across DOI, Department of Interior. Historically, USGS, and I say in academia, potentially, their comfort zone has been more in these first two circles, and then maybe part of the monitoring and evaluation.

[9:48] It's harder for our science to get effectively entrained into planning or implementation. To have a true adaptive management cycle, we have to be working towards that. We have some great examples of that, particularly in USGS and from folks at different centers excel in that, like the Patuxent Wildlife Research Center. That's still an area where that's an ongoing process to be able to make that happen.

[10:15] Jaime and I, and some colleagues at NC State and USGS, we wrote a paper trying to think through this and simplify that adaptive management type process and what that would look like. Just to simplify it as much as possible. You can think of it in these three big boxes.

[10:31] We've also been overlaying this idea about through all this, you do still need that basic scientific research, that foundational research, to provide the insights and knowledge base that allow you to enter into and build an effective adaptive management cycle. If you want more information on that, that paper was published in "Wetlands" a year or so ago.

[10:59] Getting into the meat of what we're talking about here, these at‑risk species in the US Caribbean, and Jaime set this up nicely, but just so everyone's aware of where we're talking about, this is the US Caribbean. Mostly focusing on Puerto Rico, but part of the project did also include the US Virgin Islands. We're down here in the border of the Caribbean Sea and the Atlantic Ocean.

[11:23] Jaime already talked about the species we're thinking about is Eleutherodactylus genus, and 2 endangered species, 14 at‑risk. There's a strong management concern for, how do we help these species persist or recover through time?

[11:42] As the climate is now in a nonstationary peak period, meaning it's getting warmer, how do those management actions change? How do the risks change that decision‑makers need to consider?

[11:57] One big thing they have to consider is that a warming climate doesn't mean, in the region we're talking about, that things get warmer. Unfortunately, for the most part, there's pretty good evidence that we think there will be drying in this region.

[12:11] This is called subtropical drying, and there's a lot of work that's been done to nail down the theoretical basis for why we think this region in the Caribbean could be more prone to that in the future.

[12:25] That's what we're showing, and that's for a moisture‑sensitive species like amphibians that becomes a real concern then for, how do we adapt? First of all, what is the level of risk that managers are facing and that the species are facing, and then what actions might help mitigate or reduce that risk?

[12:45] Puerto Rico itself is varied in terms of the amount of precipitation received across the island. This is showing annual precipitation averaged over about a 30‑year period.

[12:58] You go from the northeast part of the island in El Yunque, which is the rain forest mountain area here, talking about over 4,000 millimeters of rain a year, down to the southwestern part of the island, less than 1,000 millimeters a year. That's a huge difference.

[13:18] Then corresponding to those sharp precipitation gradients, you get sharp changes in the ecosystems and the ecological, we call here light zones, across the island. Going from rain forest down to subtropical, dry forest across most of the southern part of the island.

[13:35] You have this rain forest type of systems in different parts of the island, here again showing El Yunque, although it extends across also the central mountain range as well, down to Guanica dry forest where you'll have a much different landscape in a different assemblage and mix of species.

[13:54] If overall we expect drying in the region, how will that affect amphibians, specifically on the island? With something like in El Yunque, would it maybe migrate to something that's more like this? Not quite Guanica, but something more mesic, but definitely something drier.

[14:17] Thinking about that across the island, how are these gradients changing over time? How could they change over time and what does that mean for these species? That's the central part of what we're working on here. That's where we combine the climate science and the ecology in this project.

[14:37] Again, to do that, we need this foundational research to help build that knowledge base, where we're starting from a lot of unknowns and uncertainty, both about what the climate response would be in the island and what the amphibian response or the ecological response to that climate change would be on the island.

[14:57] Ultimately, all this is adding up to, I have the same color coding here, thinking about the red or the blue areas here. That's where we're at, that is the space we're operating in. Trying to evaluate candidate strategies that would be useful for managers to consider that they could then adopt to be effective adaptive management as the climate changes.

[15:19] To do that, we need to understand the relationship between climate ecology and the biology of these species that we're working with. Again, back to that foundational knowledge that we're working towards, is we need information about the climate response function of the species and systems of interest.

[15:41] What are those climate sensitivities? How do they affect different parts of the life history, the population parameters of the species? How are all those things sensitive to climate? Are they sensitive to climate? If they are, which climate variables are they most sensitive to and in what ways? Those climate variables, how do we think they might change in the future?

[16:05] One way to think about this, building up why we went about the climate modeling pathway that we did, here's a toy example. Jaime drew this example where you think about on the Y‑axis here is a risk of extinction for a particular species.

[16:23] Let's say that that risk is higher if you're starting off in the dry forest in Guanica because the species needs a wetter and more mesic habitat. That risk of extinction is much lower than in the Maricao, in the wet forests, where the occupancy of the species is more prevalent.

[16:41] If you think about that in current versus future, if the actual climate variable area, or that is to say, the time between rainfall events, maybe this affects the breeding season or something like that, as the region dries, you would expect the net risk of extinction to go up over time.

[17:04] Global climate models, or GCMs, they provide that foundational knowledge for the climate risk characterization. I already brought that into discussion. We're worried about this subtropical drying phenomenon into the future. The problem is that global climate models, their resolution is still very coarse.

[17:23] Here's an example. You can think of this as an example. One glimpse at a global climate model. I'm not going to get into all the ins and outs of global climate models, but just suffice it to say that they can't resolve what's going on over the island.

[17:37] Even though global climate models are very good at understanding the overall climate system and even down to say the continental or subcontinental scale, when you talk about, say, islands in the ocean, they're not going to be able to resolve the complexity that's going on and how this rock that's in the middle of the ocean interacts with and affects the atmospheric response over the island.

[18:06] Here again is that map showing the annual precipitation across the island. Whereas, for a typical climate model, you would get one number of precipitation for the entire island. It's not going to be able to understand how the mountains respond or interact with the atmosphere and cause that differential on precipitation.

[18:27] Not only that, but more importantly, we're not going to be able to understand if there's going to be a different response to climate change across the island. For instance, that regional climate response could manifest in different ways, depending on where you are in the island because of the complexity in the topography and the land surface and how those things interact with and feedback with the atmosphere.

[18:53] That risk of extinction perhaps it goes up even more in the wet forest than it does in the dry forest just as a hypothetical, these are the things where we're basically starting from a complete uncertainty knowledge base and which is why then we decided to use something called downscaling ‑‑ an approach called downscaling ‑‑ to be able to better understand the local manifestation of the global climate change response.

[19:24] Downscaling, put simply, it resolves either explicitly, through numerical modeling, or empirically through statistical models, the physical processes that are not resolved or/are just ignored by GCMs, because they're not operating at the scale that's GCMs operate at.

[19:43] You can take into account things like typography, on the left here, and how that would affect local climates or varied land surfaces, like the one I'm showing here, how a mixed water and land interaction or interface could then affect the climate over that area.

[20:06] We're using something called dynamical downscaling, where we run higher resolution climate models over a smaller domain, we stick in a high resolution, limited area climate model inside the large global climate model, that large global climate model we take the output from that and feed it into the higher resolution climate model.

[20:29] We're able to do that and ultimately, the goal is to achieve a higher resolution climate model, down to 2 kilometer horizontal resolution, whereas a normal global climate model would be around a 100 kilometer resolution.

[20:45] This is showing, in a nutshell, our experimental design on the climate modeling part. The downside of using dynamical downscaling is that it's extremely computationally intensive.

[20:56] If we're used to thinking about climate model ensembles, if you've ever heard that term or looking at climate model results, typically, you see like lots of different climate model output, say, up to 20 models per emission scenario. You might have two or four or five emission scenarios.

[21:16] We were not able to do that in the initial project that came with the funding, just because of computational resource and personnel constraints. What we were able to do is use three global climate models and then two RCMs, or regional climate models, that were then split across these GCMs. We use that one emission scenario. It's the higher emission scenario. We did that over the mid‑century.

[21:41] The reason we chose the higher emission scenario too, is because that gives us a nice bracketing of our uncertainty over time, where you could always, as a manager, cheat in terms of a different scenario. A lower emission scenario would just be assumed that current conditions, that the current climate is maintained through time.

[22:02] By default, you get two scenarios for the price of one. Using the higher emission scenario, you can bracket that emissions’ uncertainty.

[22:13] The original GCM projection would look something like this showing again the precipitation change by mid‑century. You get one number for all Puerto Rico, not very helpful. We don't think that's going to be accurate.

[22:27] Then, you see the before and after using the dynamical downscaling to two kilometer resolution. There's a lot more variation in the change, and you see different processes and patterns showing up across the island. Now, you really see how the island interacts with and affects the local climates with change and affects the local manifestation of the global climate change signal.

[23:01] Also, a nice thing about using dynamical downscaling is that there's many physical variables available compared to, say, alternative approaches to downscaling such as statistical downscaling or empirical downscaling.

[23:15] That gives us a lot more ability for our users and our stakeholders to be able to explore what climate change might mean for species or ecosystems that are sensitive to more than just temperature and precipitation. Of course, there's a cost to that. That cost is that there's time, storage, and processing constraints.

[23:41] This is part of our coproduction aspect of this project is we have a lot of stakeholder workshops with scientists and managers to think about, what sort of climate variables do we think are most important for the work that you're doing?

[23:58] The idea there was to have climate projections that are specific to our decision or decisions that we are most concerned about with the frogs and the amphibians, but, also, that could be relevant to other scientific, or ecological, or management questions.

[24:13] In a sense, we're not only providing information for our specific project and our specific stakeholders but trying to provide public goods as well. The idea being that this information could be used for more than just the specific context here.

[24:32] Here's an example showing some of the output from the climate models. Here, for instance, is showing the projected change and maximum daily temperature. Annual precipitation change, there's that subtropical drying showing up, that regional drying. You see it showing up here across the island.

[24:52] You see there's quite a lot of variation in that pattern. The overall signal is one of drying almost everywhere with a few small notable exceptions which I'll get to in just a minute. With the dynamical downscaling, one really nice feature is that we can look at other variables besides just temperature and precipitation.

[25:14] Things like soil moisture, there's a soil moisture module or sub‑model within the RCMs we were using. Things like low‑level cloud fraction, these become important for different cloud force type systems where that saturated moisture at the surface level can become really important for maintaining some forest systems and species that are sensitive to that.

[25:44] There's a few hints from the results that elevation may provide some buffering against the broader regional projected drying signal that we expect. What I mean by that here is if you look at this image on the right specifically, zeroing in there, this is for one combination of GCMs and RCMs. The CNRM, GCM, and what we call the WRF regional climate model.

[26:12] On the left is the same WRF model but combined with the CCSM global climate model. The blues mean that there's a projected increase of precipitation. That, you notice, is that area that I showed you in the original figure with the precipitation, the really high rainfall areas of the island, the rainforest, El Yunque National Forest.

[26:38] That's a really interesting thing to think about here is, are these elevation areas providing some sort of buffering, potentially against the broader drying pattern? We don't see evidence of that in the CCSM‑WRF combination. Here on the X‑axis is showing elevation. Each circle here is an individual two‑kilometer grid cell.

[27:04] Here on the Y‑axis is the projected precipitation change by mid‑century under the higher emissions scenario. We do see that in the CNRM‑WRF GCM run where we do see this interesting pattern of a shift in the projected precipitation change as you go up in elevation.

[27:27] To get technical here for just a minute, we think we've been able to trace this back to how the broader GCMs are projecting changes in the low‑level flow onto the island. The difference is that the CNRM model is projecting a stronger low‑level flow from the northeast onto the island. That, basically, means you get stronger trade winds.

[27:50] The effect of that is then over the higher mountainous areas on the island, you may get more forced convection and more precipitation that could be counteracting some of that broader drying pattern that we would see across the region.

[28:08] Without that increased low‑level flow, which is what we see in the CCSM, then local land surface feedbacks tend to dominate, which could result in even more drying than in the regional pattern of subtropical drying.

[28:24] In essence, what we're seeing here is we're generating new hypotheses. We're not necessarily narrowing the uncertainty, but we're learning more about what climate change could mean locally on the island by taking this approach.

[28:36] Those new hypotheses that we're generating can then be used to test new ideas and inform our next climate modeling that we do. Then, we can start to narrow down some of those uncertainties that now have been generated by what we've learned so far. Again, back to that foundational research to provide those insights and knowledge base.

[29:04] How does climate then affect these species? That's the biology part of this, it's nice to have these cool projections of climate change, with the fancy climate models and the regional climate models, but if we don't know how these species are sensitive to climate in the first place, that isn't going to give managers much useful information.

[29:27] Broadly speaking, we're trying to figure out what are the environmental limits of these species? What climate variables are they most sensitive to? One way we started out looking at that was we used acoustic recorders to help understand what are the occupancy of three Eleutherodactylus species across environmental gradients, basically precipitation and elevation gradients.

[29:56] We do some cool things. We visualize the audio data using some software from some of our colleagues, and we're able to separate out the different species. We can see those across here where we can see the different patterns in the audio showing up depending on which species are there in the background of the forest.

[30:21] The question then is, are these gradients of occupancy likely to change with climate change here? Once we were able to figure out that base understanding of where these species are living, and what climate variables they're sensitive to.

[30:42] Then, the question is, how would the species potentially react as the climate warms, and as the climate dries, and as that drying takes on maybe a more complex pattern across the island? That's where using those insights gained from the high‑resolution climate modeling can help.

[31:05] This is where we're at now is we want to take that information and, now that we are starting to understand some of the species’ sensitivities to climate, be able to use the climate model information to then build those candidate strategies with translocation work.

[31:25] How about identifying climate refugia. Something that my colleague, Toni Lyn Morelli, is an expert in. How can we develop different ways or actions that managers could use to help ensure the persistence of these species into the future?

[31:46] Some of these next steps, I've been already getting into that, but we've been able to apply the information we've had. Here's one example with this, the paper is in review right now. We have one example where our colleagues are applying the climate model information that we have.

[32:05] They've used a lot of occupancy model data, again using those acoustic recorders, and developing species distribution models to help with strategic habitat conservation. This is showing that the projected ‘always suitable areas’ by mid‑century for selected avian and amphibian species.

[32:26] Then those combined areas and you can see where those match up, then given the climatic changes that are projected. That can be a useful tool for conservation planners and managers. Jaime, I'm going to let you talk to this slide if you want to unmute for a second.

Jaime:  [32:46] Sure. Among those next steps, we’ve moved on from occupancy to abundance, which in this case, can only be done in two‑levels either few or a whole lot, which we call a low risk.

[33:01] We also looked at reproduction when we have not only occupied habitat, but our watch is occupied where do you have reproduction? Is it when there are a lot of them, or where there a few? What are the microhabitat conditions that then lead to higher reproduction?

[33:21] We were able to tie those three things together on the last iteration of work. Another component that we're tackling, we're starting, provided COVID gives us a break, is established genetic population, the population structure and gain through those analysis and concentrating particular areas in Puerto Rico.

[33:45] Insights on connectivity, we need to go through this tool first not only to identify the centers of biological genetic diversity but also because monitoring these animals is not easy. Certainly, their movements are not that big, so connecting, understanding what impedes, or facilitates moving among individuals is going to be a challenge.

[34:14] We hope to get some insight on this with a genetic score. This is another component that we also started to work with. This is tackling the eco‑physiological work of a project. Here, the idea is to understand the sensitivity of these animals to environmental stress, in this particular case, climate change.

[34:41] We're going to establish some basic metabolic performance of these animals, oxygen consumption, CO2 emissions, water loss, those kinds of things given, or for instance, current or projected conditions in the future.

[34:58] We will also gain some understanding of what are the limits? For instance, basic work on safety matters, critical thermal maximum temperatures to understand how far can these frogs go in terms of tolerating extreme temperatures, which would kick in and that's the last thing I'll say about this slide, what we call adaptive capacity of these animals, which in this case is in such a terraria.

[35:29] We will look at the interaction between natural, the climatic or the spread, if you will, and then adaptive capacity through behavior which may be in the high number crevices on the sun, vegetation, debris, and what have you. There are ways that they can deal or buffer, if you will, these harsher conditions as they become reality in the future.

[35:51] [crosstalk]

Adam:  [35:52] Cool. I'll come back to you in just a second, Jaime. The other thing that we had to take into consideration [laughs] was when Maria came through Puerto Rico in 2017. That changed a lot in the work we were doing and, of course, the logistics of working on the island. Here's some photos from our students afterwards showing the absolute devastation.

[36:34] Of course, Jaime, you want to take this one in terms of what's maybe not devastation for some of these frogs.

Jaime:  [36:42] Right. Over the years, I've worked primarily with birds, those of you that know me. Hurricanes are not good news, certainly in the short term, for birds, especially[indecipherable] . For frogs, it's a different story. It seems to be.

[37:01] Unless a place is denuded completely, what you often find, as you saw in the previous slide, is a lot of debris, broken trees and what have you. What it does is create the microhabitat for these individuals. What we found, to our pleasant surprise, was that, from the year before and the year after, is that they do exhibit higher population densities, higher reproductive activity.

[37:29] It doesn't seem to be a function or an artifact of aggregation. There's plenty of habitat around. They just seem to be doing a lot better and cranking up their reproductive rates. Part of it is because there's an explosion of food availability, and they take advantage of it.

[37:48] This has something has been shown very nicely also at the Caribbean National Forest ‑‑ now it's the El Yunque National Forest ‑‑ for Hurricane Hugo. I think it was also being shown, it will be published soon, in another part of the island.

[38:05] These animals seem to be one of those few species that have a positive response to these catastrophic events, in just about every other element of biodiversity.

Adam:  [38:18] Thanks, Jaime. This is wrapping up the last slide here. The big picture, the dream, is to work in adaptive management as a means to manage risk. We know there are risks with climate change, particularly in the US Caribbean and Puerto Rico.

[38:45] Warming in this region, we do think, it's more likely than not, it would mean drying. At least under the greenhouse gas emission trajectories that we are currently on.

[39:02] Working with our managers, and our stakeholders, and our partners to build the base scientific knowledge, and then be able to effectively entrain this into an adaptive management cycle, has been our shining star throughout this.

[39:23] It's a long process. Like I said, this has been seven years now, but we keep building those building blocks. We're learning a lot along the way, both about the science and the species itself, but also how to go about doing this actionable science.

[39:43] With that, thank you. We'll be happy to take any questions.

Philip:  [39:47] Thank you, Adam, Jaime, for that great presentation. We are now moving to the Q&A session. Please feel free to ask any questions on the bottom right‑hand corner, and we'd be happy to call on you.

[40:03] We have one question from Jamie. His question is, "Are managers actually responding to your input, and advice on how to get them to respond to the kind of messages they often don't want to hear?"

Jaime:  [40:18] The answer will be yes. If they're speaking, not only the information but our advice. We are hoping that that is the case. That's what I can say right now.

[40:35] So far, the information we’ve generated has been very much welcomed by management. Both levels at both agencies, the state, and the federal agency, the Fish & Wildlife Service.

Adam:  [40:48] Jaime, I see a question from Sean, "What information is available for potential new diseases in this area, as a result of warming and drying?"

Jaime:  [41:00] That work is being tackled by some colleagues of the University of Puerto Rico, mostly the chytrid fungus. They came out with a recent paper, it was last year, in "Science," where they dealt with it in a global scale. They included the locations in Puerto Rico as increasing with warming. In the article, they've been [indecipherable] . Obviously, with predictions on survival of these species.

[41:37] It's yet another factor that will come, another threat that comes as a result, mediated through the global warming.

Philip:  [41:48] Great. We have one question from Carrie, they ask, "How do you see these strategies you've developed during this project can best be extended to other regions/issues?"

Adam:  [42:00] You want to try it first, Jaime? Then, I can give it a shot.

Jaime:  [42:03] Sure. Yeah, the fundamentals are basic and very...You can explore, you can pick them up. If you think about the way we're using translocation experiments to, for instance, learn what is limiting the distribution of a species. One experiment will be moving these animals from occupied habitat to a seemingly suitable, but unoccupied habitat.

"[42:31] Why are they not there?" is one question. This is something that can be done anywhere. We're also looking at, what we call, the extremes experiment. That is, moving the animals from suitable conditions down to marginal, and then perhaps into extreme conditions. We would see how they manage.

[42:56] Again, when going back to the adaptive capacity of the animal, not only the physiological aspect, which we cannot measure directly, but we can measure movement. We can measure survival of these animals, and to learn what to expect when these changes take place. In the ecophysiological aspect, it's the same thing.

[43:20] We're trying to identify what are the thresholds. Not only the ones that approximate lethal conditions, but those that are more interesting, which are the sub‑lethal conditions that affect, as I was saying, the metabolic performance of these animals.

[43:36] Oxygen consumption, CO2 production, water loss, these things are the subtle, hard‑to‑detect changes and adjustments that animals will have to make before things kick‑in in strong, or harsh conditions. These are things that can be adopted by any other, emulated by any other set of colleagues anywhere in temperate regions or tropical regions.

[44:07] It's basic research to understand what are the limits and capabilities of these animals when they're facing threats? Especially, if you can reproduce those threats in captivity, in lab conditions, as we hope to do, or through experimental translocations, as we're also trying to do. Those are two examples of how it can be transported to other areas.

Adam:  [44:32] The only other thing I would add to that, again more on our strategy in terms of our research approach, Carrie, would be that climate modeling was...I think that ended up being a nice approach we took, in terms of better linking up the ecologists with the climate modelers themselves. Those two groups often don't have in‑depth communication.

[45:09] It's usually the ecologist taking information from climate modelers, taking what they can get. Bringing those two groups together in a true co‑production type of manner was helpful, and can point a way forward. The interest from USGS, if I might say, in funding climate modeling ebbs and flows, if I'm being diplomatic.

[45:44] We've demonstrated that there's definitely a need for it and if we're going to be able to understand the risks that our stakeholders, particularly in areas where there's not going to be off‑the‑shelf climate model information, useful climate model information, that could be applied in this context.

[46:15] Again, by learning what's going on, even though it's probably increased our uncertainty, we have a better idea of what we know and don't know. We have a better idea of those known unknowns now than we had before. Before, they were just unknown unknowns, to use the Rumsfeld terminology from way back when.

[46:35] That's been a big help and is an applicable lesson to other regions and certainly in the CASC network. I see Paige's question. I'll just read it.

[46:46] Do you see any indications so far that these species will be able to either, A, adapt in place to a changing climate or, B, move and track the habitat they need as the climate changes and make previously suitable habitat unsuitable or vice versa? Jaime?

Jaime:  [47:01] Yeah. That is one of the things we're trying to tackle and why we are trying to understand this physiological performance of these animals, and then along with that, some of the behavioral responses to see to what extent.

[47:17] As you probably know, from your own studies, certainly the coquis, the frogs that we're talking about, they are not good movers, especially if there's a lot of resistance, the matrix between the good habitat and the next good habitat.

[47:36] To the best of our knowledge, experimentally, it's been show that they can move as far as 100 meters. Normally, they don't go beyond 10 meters. That's not a whole lot of movement. The so‑called positive tracking will likely not be the best option for them. That's why we're thinking about managed translocations to assist. That's the whole thing that's intended to do, assist these animals with migration.

[48:06] What we're trying to understand a little bit with the physiological work ‑‑ I don't think we can get to it yet with the genetics to understand, at least in this first round ‑‑ what are their adaptive capacities?

[48:22] If you read the literature ‑‑ and I know you are well aware of the literature ‑‑ most people are hanging their hat on the adaptive capacity of these animals but, of course, there are limits to that.

[48:32] By pushing the envelope in terms of physiological responses to these scratches, we can start to understand what those thresholds are and then the need to take action and then implement some managed translocations.

Adam:  [48:48] What is the plan for using the findings from the ecophysiology experiments in terms of translating, say, temperature‑dependent metabolic rate in water loss to environmental suitability?

Jaime:  [49:01] We hope that we can relay these parameters or these responses, because we're going to be creating these conditions, emulating either current conditions or projected conditions.

[49:15] The linkages will hopefully be there, at least, or maybe paying us initially. We hope to strengthen those with time. The idea is to then be able to go out in the field and measure some habitat and then assess their suitability for transportation.

[49:36] Part of what Adam and I would like to start doing as well ‑‑ and we do have a plan to start doing that ‑‑ is either we connect our understanding of these relationships of the species in the field and the macroclimatic conditions that they respond to, most often in terms of occupancy abundance, and then reproduction, and then ecophysiological with the downscale projection.

[50:08] If we can make those connections a little clearer, I think this is where we gain a tool to then look for strategic decisions in terms of what habitat should be protected because they have potential as climate refugees. Adam, I don't know if you want to talk a little bit about what you have in mind. That linkage is very important on this, what you call coproduction.

Adam:  [50:35] Right. Part of the next step there ‑‑ this also gets a little bit back to Paige's question ‑‑ is, we do have, at least from the climate model we've run so far, we can start to piece together scenarios of, based on the projections in the future, which areas on the island now are sort of the analogs of the future.

[51:02] Some areas that are projected there is no analog. If you're in Guanica and it's projected to get drier...In the future, there's no place drier than Guanica, the island. You have to go to ‑‑ I don't know ‑‑ Arizona or something to find the analog.

[51:16] If you're talking about the wettest habitats up in the mountains and the drying that's projected there, at least under one of the climate models, we have some candidate sites in the Karst region, which is just a little bit north of there, that we think might serve as interesting and potentially good analogs for what those wettest mountain areas might look like in the future.

[51:47] Some drying but not completely xeric. One of our collaborators actually have some sites there that we are interested in testing, some potential climate refugia and little microhabitat sites to see how some of these species might fare there now to get a hint at how are they able to then survive in a place that is either a refugia site so it's cooler or more moist than the surrounding area.

[52:24] Or they can find shelter in the broader region there to get a sense of how we think they might be responding in the future as the climate changes.

Jaime:  [52:37] We need to make clear that the ecophysiological work provides us with...We can measure movement. We can measure what they do as survival, but what may be inducing that mortality other than predators, which we know exists, is what we need to understand.

[52:55] There are a number of mediating mechanisms to affect that survival, that behavior and ability to survive in a particular place, which may be stressful. It is through this mechanistic understanding, for instance this ecophysiological work, that gives us some insights of what the animals may be enduring or having to deal with as we put them in different habitats under different conditions.

[53:23] They provide some input as to what are the mechanisms that lead to an ultimate response which is measured, in our case, as survival of these animals. [indecipherable] survival.

Adam:  [53:35] That's it for the questions.

Philip:  [53:43] I want to thank the presenters again for their valuable insight.

Adam:  [53:50] Great. Thank you so much.

Jaime:  [53:52] Thank you.