U.S. Geological Survey Fisheries and Wildlife Science for a Changing World: Part II


8:00AM Isotopic Insights into the Nutritional Ecology of Salmon, Sea Lions, and Polar Bears
  Craig Stricker
Stable isotopes are a powerful tool for making inference on animal nutrition. Consumer tissue selection dictates the dietary integration window and can be used to gain insight into specific life history traits. Slower turnover tissues integrate nutrition more broadly, but may be equally powerful for cryptic or logistically challenging study species. Inference on animal nutrition using isotopes generally requires assembling prey data, which can be difficult for broadly distributed species. Three case studies that include Pacific salmon, Steller sea lions, and polar bears will be explored to illustrate the utility of stable isotopes in the study of animal nutrition. Our studies of juvenile salmon have demonstrated that adult spawners provide an important resource subsidy that scales with escapement and bolsters overall productivity. For generalist sea lions, we have shown that diet estimation is feasible, have developed a modeling framework that can accommodate time series data, and that seasonally these animals shift trophic levels. Polar bears are novel carnivores that specialize on the consumption of blubber when available, changing the traditional paradigm for isotope-based investigations where the focus is on dietary protein assimilation. As a result, model development that includes both dietary protein and lipid is currently in progress.
8:20AM New Approaches Linking Trout Abundance to Headwater Stream Flow
  Nathaniel P. Hitt
Effects of stream flow on fish populations can be spatially structured within stream networks, thus requiring spatially intensive observations in addition to downstream river gages. Here we describe on-going research in Shenandoah National Park (Virginia) to understand the spatial properties of native brook trout abundance in relation to stream flow. We estimated flow at multiple sites within small watersheds (~HUC12 scale) and collected spatially-structured fish abundance data from stream pools using three approaches: underwater video, electrofishing, and dive counts. Video data yielded similar abundance estimates of adult trout as electrofishing but consistently lower abundances than dive counts. Regression-tree models indicated pool geomorphic features that affect method performance for estimating adult fish abundance (pool volume and riffle crest depth). We demonstrate that video sampling provides a rapid, cost-effective sampling strategy to assess fish-flow relationships for adult fish at the necessary spatial scale in stream networks.
8:40AM Capture-Recapture Meets Big Data: Integrating Statistical Classification with Ecological Models of Species Abundance and Occurrence
  Andy Royle
Advances in new technologies such as remote cameras, noninvasive genetics and bioacoustics provide massive quantities of electronic data. Much work has been done on automated (“machine learning”) methods of classification which produce sample class designations (e.g., identification of species or individuals) that are regarded as observed data in ecological models. However, these data are actually derived quantities (or synthetic data) and subject to various important sources of bias and error. It is well-known that misidentification of individuals in capture-recapture or false positives in occupancy models produce biased inferences about population size or occupancy probabilities. Thus, if the derived classifications are used to make ecological determinations without consideration of these errors, then inferences which inform monitoring, conservation, and management will be flawed. We have developed methods of “coupled classification” in which statistical classification models are linked to ecological models of species abundance or occurrence. In this new framework, classification (e.g., species identification) takes into account the local structure of populations, communities and landscapes and does not assume that where a sample is collected is independent of the class structure of the population, as most current classification methods do.
9:00AM Development of a Coordinated Regional Program to Monitor for Dreissenid Mussels in the Columbia River Basin
  Timothy D. Counihan
Efforts to monitor for the presence of dreissenid mussels in the Columbia River Basin (CRB) have increased over the past decade in response to concerns that these invasive mussels may be introduced and become established. The motivation for the increased monitoring is to improve the probability of detecting infestations and, optimally, to provide early detection capabilities so that new arrivals are detected soon after their introduction. To increase the probability of detecting invasive mussel infestations in the CRB, scientists from the U.S. and Canada have come together to explore options to increase the efficiency of existing monitoring programs. For example, reallocating existing monitoring efforts to areas of higher infestation risk, such as areas where propagule pressure from areas already infested is highest and where dreissenids are most likely to become established, could increase the probability of detecting infestations. We will provide an overview of regional efforts to 1) convene an international forum to coordinate dreissenid mussel monitoring in the CRB, 2) assess the extent and magnitude of current monitoring efforts within the CRB, 3) better define infestation risk for water bodies in the CRB, and 4) develop a strategy to coordinate and focus dreissenid mussel monitoring activities.
9:20AM Water Storage Decisions in Response to Drought in the Colorado River Basin will Drive Aquatic Ecosystem Dynamics
  Kimberly L. Dibble
Managing the world’s freshwater supply to meet societal needs is one of the biggest challenges of the 21st century as human populations expand and precipitation patterns change globally. Dams and their associated reservoirs capture spring snowmelt and provide more reliable water supplies to society, but dams also fragment rivers and create thermal discontinuities that profoundly influence aquatic communities. Here, we quantify thermal discontinuities created by dams across western US rivers, investigate direct and indirect drivers of temperature throughout 2560 km of river network using the Colorado River basin as a model system, and predict future Colorado River temperatures under various scenarios of air temperature, flow, and water allocation decisions. While river temperatures downstream from shallow reservoirs will primarily be driven by atmospheric warming and declining runoff, policy-level decisions on water storage among large Colorado River reservoirs will have a far greater effect on water temperature than atmospheric warming or declining runoff. Under some scenarios, Colorado River temperatures may warm to pre-dam conditions, however, it is unclear whether a return to natural water temperatures will facilitate recovery of native, endangered fishes or lead to further native fish declines owing to invasion of piscivorous warm-water non-natives.
09:40AM Break
10:10AM Climate Change and Plague Dynamics: Implications for Prairie Dog Management
  Robin Russell
Climate change has the potential to affect both the vector and host involved in the plague transmission cycle. Flea reproduction and growth is dependent on suitable weather conditions, while prairie dog health and body condition has been shown to be associated with vegetation condition (also dependent on weather). Though burrow conditions may insulate fleas and prairie dogs from fluctuating conditions above ground, in the United States climate change is expected to shift conditions for plague northward and eastward. We present results of 5,024 flea samples collected from 4,218 individual prairie dogs during a large-scale field trial of an oral sylvatic plague vaccine. These results indicate that in general, flea abundance (number of fleas on hosts) is higher during plague outbreaks, lower when prairie dogs are more abundant, and reaches peak levels when climate and weather variables are at intermediate levels. We incorporate these results into a spatially-explicit simulation model of plague transmission and demonstrate how these models can be used to simulate the effects of climate change and plague mitigation tactics (vaccination and dusting) on prairie dog populations and the likelihood of plague outbreaks.
10:30AM Are Trout in Hot Water? Understanding Climate Change Impacts for Conservation of Native Salmonids in the Northern Rockies
  Robert Al-Chokhachy
Salmonids are a group of fishes with high socioeconomic and ecological value. The narrow thermal tolerances and strong ties to historical hydrologic regimes hint at the sensitivity of salmonids to changing climatic conditions and justify concerns. We combined long-term biological monitoring data with high-resolution climate predictions to evaluate how climate change and other human stressors influence the vulnerability of native salmonids in the northern Rocky Mountains, USA and Canada. Our findings illustrate that a shifting climate is exacerbating interactions with non-native species, leading to increases in hybridization and declines in native salmonid abundance and distribution for species such as Bull Trout Salvelinus confluents and Cutthroat Trout Oncorhynchus clarkii subsp. Our results also illustrate considerable variability in exposure of a changing climate at finer spatial scales (i.e., population level). Despite the negative effects of non-native species, our research highlights the benefits of proactive control programs in tempering the impacts to native populations. Together, these results indicate that climate warming will inhibit cold-water salmonids, yet other stressors – especially invasive species and habitat loss – are immediate threats, and identifying frameworks to effectively prioritize management actions is an important step in the conservation of salmonids.
10:50AM Mapping Old-Growth Forest Vulnerability to Hotter Droughts: Lessons from California’s Extreme Drought
  Nathan Stephenson
Hotter droughts – droughts in which unusually high temperatures exacerbate the effects of low precipitation – are expected to increase in frequency and severity in coming decades, challenging us to identify which parts of forested landscapes may be most vulnerable. However, standard niche-modeling approaches to species vulnerability mapping suffer three problems that can render their results useless to land managers: (1) key data layers (like temperature and precipitation) often contain substantial errors, (2) other critical environmental variables (like subsurface hydrology) are invisible to us, and (3) standard modeling and validation approaches can lead us to believe we have created an accurate model when we have not. We thus sought to develop a new, empirical approach to vulnerability mapping, taking advantage of California’s 2012-2016 hotter drought to let the trees themselves reveal which parts of the forested landscape are most vulnerable. Our “leaf to landscape” approach uses strategically co-located ground-based monitoring of leaf physiology, leaf chemistry, and tree population dynamics during drought to calibrate and validate state-of-the-art LiDAR and hyperspectral remote imagery. Early results, including those for the iconic giant sequoias, appear promising and have already provided some unexpected insights, but some clear challenges remain.
11:10AM Interdisciplinary Approaches to Addressing Ecological Drought in the U.S. Geological Survey
  Jason Dunham
Growing awareness of the magnitude and extent of drought across much of the Nation has led to increasing concerns over water availability and effects on humans and ecosystems. To respond to this issue, the U.S. Geological Survey (USGS) published an Integrated Drought Science Plan in 2018. The Strategy is built on the need for interdisciplinary collaboration, leading to a scientific and institutional alignment of major science divisions within USGS (Mission Areas). The first steps toward interdisciplinary collaboration involved two recent publications that define ecological drought and articulate specific plans for scientific integration. Key points from these papers and results of ongoing efforts to implement ideas from these publications are detailed. Lessons learned to date highlight both challenges and opportunities to adapt drought science to address an increasingly urgent issue.

Organizers: Clint Muhlfeld, Melanie Steinkamp, Steven Hilburger, Jean V. Adams, John Thompson, Mark Wimer, Paul Wagner
Supported by: U.S. Geological Survey, Ecosystems Mission Area

Location: Reno-Sparks CC Date: October 2, 2019 Time: 8:00 am - 11:50 am