Erin Muths

Aims: Data from participatory science projects are widely used to understand species distributions. However, participatory science data often contain biases that can obscure real patterns. We address two sources of sampling bias: location uncertainty of individual records and spatially heterogeneous sampling effort. A variety of data processing and modelling strategies have been used to address these biases, without consensus on their relative effectiveness. We test several strategies using data from iNaturalist, a large repository of participatory science data. Location: Twenty‐four amphibian ranges of varying sizes and locations within the continental United States. Method: We fit species distribution models with iNaturalist data for 24 amphibian species to compare the effectiveness of strategies intended to improve distribution estimates. We examined three categories of strategies: (1) filtering out records with high location uncertainty, (2) thinning to reduce heterogeneous effort, and (3) observation models that account for heterogeneous effort without removing data. We evaluated the predictive accuracy of each approach using validation data from systematic surveys with constant effort and known locations. Results: The strategies we tested generally had little effect on predictive accuracy as measured by AUC, with no consistent pattern across species. Instead, we found that the minimal change in performance was influenced by the species‐specific search process, the distribution of records, and the resolution of the analysis. Main Conclusion: None of the strategies we tested consistently improve model performance, but our results provide a foundation for a species‐ and analysis‐specific approach. Further improvement in model performance may require additional information or novel methods.

Emerging infectious diseases affect wildlife populations around the world, making effective conservation strategies essential to the long‐term persistence of threatened species. Understanding the drivers of protective host responses to pathogens can help guide conservation efforts. Species that vary intraspecifically in their susceptibility to pathogens are often best for identifying significant drivers, as this reduces potentially confounding effects of highly divergent genetic and adaptive backgrounds. Boreal toads ( Anaxyrus boreas boreas ) exhibit such intraspecific variation in disease susceptibility. This species ranges across western North America and declines due to the fungal pathogen Batrachochytrium dendrobatidis (Bd) have varied in severity across populations. Previous experimental work shows that some populations of boreal toads are better able to tolerate Bd than others, though the mechanisms are currently unclear. We sequenced the immune locus MHC IIβ exon 2 across tolerant and susceptible populations of boreal toads to determine if host genetics is correlated with Bd susceptibility in this system. We found that tolerant populations differ from susceptible populations at this locus. Specifically, tolerant populations had several distinct MHC alleles, higher observed heterozygosity, and a higher proportion of alleles belonging to one of two functional supertypes. Using these data, we predicted the Bd response of additional boreal toad populations with unknown susceptibility. Boreal toad conservation efforts will benefit from the ability to identify at‐risk populations and individuals with genetic signatures of Bd tolerance for use in captive breeding and reintroduction efforts.

Managing species in an uncertain future is a reality for natural resource decision makers. Climate change is expected to exacerbate threats such as habitat loss and disease, and cause phenological mismatches, but there is uncertainty in the magnitude of these effects. Amphibians are among the most threatened taxa on earth, and most species in North America are uniquely tied to water availability for breeding, larval development, thermal refugia, and food availability. Changes in water availability and temperature may result in phenological mismatches with one or more of these processes. Thus, quantifying the dependency of amphibians to water on the landscape is critical to understanding how species may respond, as well as understanding the interplay with other threats, such as disease. We developed a dynamic co‐occurrence occupancy model to explore the effects of climate change on the breeding occurrence of boreal toads ( Anaxyrus boreas ) and the amphibian chytrid fungus ( Batrachochytrium dendrobatidis , Bd) in the southern Rocky Mountains (SRM). We derived novel covariates to test hypotheses related to multi‐generational impacts of climate on the dynamics of both boreal toad breeding and Bd. We report estimates of current (2001–2019) and future (2055–2069) occupancy under a range of plausible climate scenarios. The probability of boreal toad breeding occurrence at a site in the SRM declined > 40% from 2001 to 2019, and further declines are likely under future scenarios, particularly as active season length increases. To help integrate this information into management, we developed a web‐based decision support tool to summarize predicted future hydrological and occupancy conditions.

Life history strategies vary widely among species and play a vital role in extinction risk, especially in a rapidly changing environment. For many taxa, information on life history such as longevity, lifespan, and generation time is incomplete. This is especially true for amphibians, which have experienced large‐scale declines in recent decades. The mountain yellow‐legged frog ( Rana muscosa ) is a California endemic recognized as a state and federally endangered species. We evaluated a 23‐year dataset of six wild R. muscosa populations in southern California. We calculated the average lifespan of individuals in these six populations to be approximately 9.5 years, with a mean generation time of 7.4 years. We did not detect a difference in longevity between sexes or a difference in apparent survival across various ages of adults. We also documented the longest‐lived ranid frog ever recorded from a wild population: a male R. muscosa that was at least 21 years old. Our results suggest a relatively long generation time for this species, a characteristic that may benefit them because reproduction is regularly challenged by drought, fire activity, and disease. This information is important for understanding the complex life history of this endangered ranid frog and can help guide efforts to manage and recover the species.

Disease, alone or combined with other stressors such as habitat loss and contaminants, affects wildlife populations worldwide. However, interactions among stressors and how they affect demography and populations remain poorly understood. The amphibian chytrid fungus (Batrachochytrium dendrobatidis; Bd) is a sometimes-lethal pathogen linked with population declines and extirpations of amphibians globally. Laboratory evidence shows ubiquitous contaminants like methylmercury (MeHg) can reduce vigor and survival of amphibians, but population-level effects remain unclear. We used non-lethal sampling to assess how Bd and MeHg affected survival of juvenile and adult amphibians in 20 populations across the USA. Survival of several species declined with increasing Bd loads, including some species previously considered resistant to Bd (e.g., eastern newt [Notophthalmus viridescens]). Although our sampling for MeHg was less intensive than for Bd, we found MeHg can both directly reduce survival and synergistically magnify the effects of Bd infection. For a population of foothill yellow-legged frogs (Rana boylii), the estimated reduction in survival from MeHg exceeded that from Bd. Although effects varied widely among populations and species, our results help clarify the potential for synergistic effects of disease and contaminants and emphasize the complexity of identifying and quantifying the population-level effects of interactions among stressors.

Determining where animals are, and if they are persisting across protected landscapes, is necessary to implement appropriate management and conservation actions. For long‐lived animals and those with boom‐and‐bust life histories, perspective across time contributes to discerning temporal trends in occupancy and persistence, and potentially in identifying mechanisms affecting those parameters. Long‐term data are particularly useful in protected areas to quantify indicators of change that may be less obvious or occur more slowly. We used long‐term amphibian data from Rocky Mountain National Park (RMNP) in a Bayesian occupancy modeling framework to estimate changes in occupancy, colonization, and persistence of amphibians over three decades and to explore the effects of climate, landscape change, and visitor use as mechanisms behind observed changes. Our results indicate that colonization and persistence are low and/or declining for Pseudacris maculata, Lithobates sylvaticus, and Ambystoma mavortium, and that occupied catchments are increasingly isolated. We found visitor use to have a consistently negative effect on occupancy and persistence of amphibians in RMNP, and that all species are more likely to occupy catchments with more complex habitat and a higher proportion of wetlands. While these results are sobering, they also provide a way forward where mitigation efforts can target identified drivers of change.

Batrachochytrium dendrobatidis (Bd) is a globally distributed fungal pathogen of amphibians that has contributed to one of the largest disease‐related biodiversity losses in wildlife. Bd is regularly viewed through the lens of a global wildlife epizootic because the spread of highly virulent genetic lineages has resulted in well‐documented declines and extinctions of multiple amphibian species. However, the current state of Bd occurrence, host range, host impacts, and ecological drivers remains poorly understood outside of the most negatively affected amphibian species and regions. Our objective was to describe the macroecology of Bd occurrence and infection intensity on caudates (salamanders) across the United States and to compare these patterns with better‐studied anurans (frogs and toads). We collected swabs from 11,183 amphibians at 609 sites from 54 species across the United States from 2015 to 2017. We analyzed the prevalence and intensity of Bd infection jointly using a Bayesian hurdle model with covariates of site‐level temperature and precipitation, as well as individual characteristics and species identification. Bd was distributed widely across sites and species sampled across the spatial extent of the conterminous United States. We found that Bd prevalence and intensity were most strongly influenced by temperature in the month preceding sampling and by differences among taxon groups. We estimated that temperature had a strong and nonlinear influence on both Bd prevalence and intensity with peak infection at intermediate temperatures and lower infection at low and high temperatures. We found Caudate hosts tended to have higher prevalence than Anuran hosts and Anuran hosts tended to have higher intensity at optimal temperatures for Bd infection. Our findings suggest that Bd has an amphibian‐wide host range, temperature gradients exert a strong influence on Bd, and enzootic transmission likely encompasses a much larger spatial and species distribution than previously recognized across North America.

The Arizona Toad (Anaxyrus microscaphus) is restricted to riverine corridors and adjacent uplands in the arid southwestern United States. As with numerous amphibians worldwide, populations are declining and face various known or suspected threats, from disease to habitat modification resulting from climate change. The Arizona Toad has been petitioned to be listed under the U.S. Endangered Species Act and was considered “warranted but precluded” citing the need for additional information – particularly regarding natural history (e.g., connectivity and dispersal ability). The objectives of this study were to characterize population structure and genetic diversity across the species’ range. We used reduced-representation genomic sequencing to genotype 3,601 single nucleotide polymorphisms in 99 Arizona Toads from ten drainages across its range. Multiple analytical methods revealed two distinct genetic groups bisected by the Colorado River; one in the northwestern portion of the range in southwestern Utah and eastern Nevada and the other in the southeastern portion of the range in central and eastern Arizona and New Mexico. We also found subtle substructure within both groups, particularly in central Arizona where toads at lower elevations were less connected than those at higher elevations. The northern and southern parts of the Arizona Toad range are not well connected genetically and could be managed as separate units. Further, these data could be used to identify source populations for assisted migration or translocations to support small or potentially declining populations.

Due to the ubiquity of disease in natural systems, hosts have evolved strategies of disease resistance and tolerance to defend themselves from further harm once infected. Resistance strategies directly limit pathogen growth, typically leading to lower infection burdens in the host. A tolerance approach limits the fitness consequences caused by the pathogen but does not directly inhibit pathogen growth. Testing for intraspecific variation in wild host populations is important for informing conservation decisions about captive breeding, translocation, and disease treatment. Here, we test for the relative importance of tolerance and resistance in multiple populations of boreal toads ( Anaxyrus boreas boreas ) against Batrachochytrium dendrobatidis (Bd), the amphibian fungal pathogen responsible for the greatest host biodiversity loss due to disease. Boreal toads have severely declined in Colorado (CO) due to Bd, but toad populations challenged with Bd in western Wyoming (WY) appear to be less affected. We used a common garden infection experiment to expose post-metamorphic toads sourced from four populations (2 in CO and 2 in WY) to Bd and monitored changes in mass, pathogen burden, and survival for eight weeks. We used a multi-state modeling approach to estimate weekly survival and transition probabilities between infected and cleared states, reflecting a dynamic infection process that traditional approaches fail to capture. We found that WY boreal toads are highly tolerant to Bd infection with higher survival probabilities than those in CO when infected with identical pathogen burdens. WY toads also had higher probabilities of clearing infections and took an average of five days longer to reach peak infection burdens. Our results demonstrate strong intraspecific differences in tolerance and resistance that explain why population declines vary regionally across the species. We used a robust, multi-state framework to gain inference on typically hidden disease processes when testing for host tolerance or resistance and demonstrated that describing an entire species as ‘tolerant’ or ‘resistant’ is unwise without testing for intraspecific variation in host defenses.