With projected decreases in biodiversity looming due to changing environmental conditions, it is important for conservation managers to have accurate predictions of species’ distributions. Species distribution modeling (SDM) and mechanistic modeling approaches that account for biophysical factors are important tools for predicting potential distributions and range limitations for many species. Recent advances in microclimate modeling have allowed for the incorporation of microclimate models into SDMs and other mechanistic approaches, a critical step for developing biologically relevant models for the variety of organisms reliant on microclimatic regimes. However, there remains a need to integrate microclimate-derived SDMs and mechanistic models at fine spatio-temporal scales to quantify and limit model uncertainty. In this study, we developed microclimate SDMs and mechanistic models of dispersal resistance for two plethodontid salamanders at a fine spatial resolution (3 m2) across Great Smoky Mountains National Park, USA during the 2010, 2030, and 2050 time periods. We determined the spatio-temporal agreement between microclimate SDMs and dispersal resistance to assess model uncertainty. We also modeled microclimate dispersal corridors and assessed spatio-temporal variability within these pathways. We found that agreement between microclimate SDMs and dispersal resistance models was generally poor. Importantly, model agreement varied across temporal periods and at differing spatial extents and resolutions. Furthermore, dispersal corridors varied temporally and demonstrated increased habitat fragmentation under future projections. The findings from this study highlight a potential contradiction in which we have a need to model species distributions with microclimate data at finer, more biologically meaningful resolutions but model agreement between correlative and mechanistic approaches may be weakened at these fine scales. Further quantifying model uncertainty and working on alternative methods for integrating SDMs and mechanistic models at fine-scale resolutions will be an important step towards accurately predicting species distributions under changing environmental conditions.

Understanding how 3D habitat structure drives biodiversity patterns is key to predicting how habitat alteration and loss will affect species and community-level patterns in the future. To date, few studies have contrasted the effects of three-dimensional (3D) habitat composition with those of 3D habitat configuration on biodiversity, with existing investigations often limited to measures of taxonomic diversity (i.e., species richness). Here, we examined the influence of Light Detecting and Ranging (LiDAR)-derived 3D habitat structure–both its composition and configuration–on multiple facets of bird diversity. Specifically, we used data from the National Ecological Observatory Network (NEON) to test the associations between eleven measures of 3D habitat structure and avian species richness, functional and trait diversity, and phylogenetic diversity. We found that 3D habitat structure was the most consistent predictor of avian functional and trait diversity, with little to no effect on species richness or phylogenetic diversity. Functional diversity and individual trait characteristics were strongly associated with both 3D habitat composition and configuration, but the magnitude and the direction of the effects varied across the canopy, subcanopy, midstory, and understory vertical strata. Our findings suggest that 3D habitat structure influences avian diversity through its effects on traits. By examining the effects of multiple aspects of habitat structure on multiple facets of avian diversity, we provide a broader framework for future investigations on habitat structure.

Movement and demographic rates are critical to the persistence of populations in space and time. Despite their importance, estimates of these processes are often derived from a limited number of populations spanning broad habitat or environmental gradients. With increasing appreciation of the role fine-scale environmental variation in microgeographic adaptation, there is need and value to assessing within-site variation in movement, growth, and demographic rates. In this study, we analyze three years of spatial capture-recapture data collected from a mixed-use deciduous forest site in central Ohio, USA. Study plots were situated in mature forest on a slope and in successional forest on a ridge but were separated by less than 100-m distance. Our data showed that the density of salamanders was less on ridges, which corresponded with greater distance between nearest neighbors, less overlap in core use areas, greater space-use, and greater shifts in activity centers when compared to salamander occupying the slope habitat. However, these differences were moderate. In contrast, we estimated growth rates of salamanders occupying the ridge to be significantly greater than salamander on the slope. These differences result in ridge salamanders reaching maturity more than one year earlier than slope salamanders, increasing their lifetime fecundity by as much as 43%. The patterns we observed in space use and growth are likely the result of density-dependent processes, reflecting differences in resource availability or quality. Our study highlights how fine-scale, within-site, variation can shape population demographics. As research into the demographic and population consequences of climate change and habitat loss and alteration continue, future research should take care to acknowledge the role that fine-scale variation may play, especially for organisms with small home ranges or limited vagility.

Climate change is expected to systematically alter the distribution and poEurycea pulation dynamics of species around the world. The effects are expected to be particularly strong at high latitudes and elevations, and for ectothermic species with small ranges and limited movement potential, such as salamanders in the southern Appalachian Mountains. In this study, we sought to establish baseline abundance estimates for plethodontid salamanders (family: Plethodontidae) over an elevational gradient in Great Smoky Mountains National Park. In addition to generating these baseline data for multiple species, we describe methods for surveying salamanders that allow for meaningful comparisons over time by separating observation and ecological processes generating the data. We found that Plethodon jordani had a mid-elevation peak (1500 m) in abundance and Desmognathus wrighti increased in abundance with elevation up to the highest areas of the park (2025 m), whereas Eurycea wilderae increased in abundance up to 1600 m and then plateaued with increasing uncertainty. In addition to elevation, litter depth, herbaceous ground cover, and proximity to stream were important predictors of abundance (dependent upon species), whereas daily temperature, precipitation, ground cover, and humidity influenced detection rates. Our data provide some of the first minimally biased information for future studies to assess changes in the abundance and distribution of salamanders in this region. Understanding abundance patterns along with detailed baseline distributions will be critical for comparisons with future surveys to understand the population and community-level effects of climate change on montane salamanders.

The field of landscape genetics has been rapidly evolving, adopting and adapting analytical frameworks to address research questions. As landscape genetic analyses have shifted away from Mantel-based analytical frameworks, studies are increasingly using regression-based frameworks to understand the individual contributions of landscape and habitat variables on genetic differentiation. This paper outlines appropriate and inappropriate uses of multiple regression for these purposes. Of concern is the prevalence of studies seeking to explain genetic differences by fitting regression models with effective distance variables calculated independently on separate landscape resistance surfaces. When moving across the landscape, organisms cannot respond independently and uniquely to habitat and landscape features. Therefore, independent resistance surfaces and their effective distance measures have no mechanistic meaning or relevant statistical interpretation. There are also tremendous challenges to fitting and interpreting regression models that include ‘independent’ effective distance measures as predictors, including statistical suppression. As such, regression analyses seeking to understand how landscape resistance affects gene flow should be univariate models, with the creation of a single resistance surface being a necessary precursor to the regression analysis. There are, however, important statistical advances underway that explicitly model the covariance of allele frequencies or genetic distances as functions of spatial landscape variables. The growth and evolution of landscape genetics as a field has been rapid and exciting. It is the goal of this opinion paper to highlight past missteps and to ensure that future use of regression models will appropriately consider the process being modeled, which will provide clarity to model interpretation.