The conservation field is experiencing a rapid increase in the amount, variety, and quality of spatial data that can help us understand species movement and landscape connectivity patterns. As interest grows in more dynamic representations of movement potential, modelers are often limited by the capacity of their analytic tools to handle these datasets. Technology developments in software and high-performance computing are rapidly emerging in many fields, but uptake within conservation may lag, as our tools or our choice of computing language can constrain our ability to keep pace. We recently updated Circuitscape, a widely used connectivity analysis tool developed by Brad McRae and Viral Shah, by implementing it in Julia, a high-performance computing language. In this initial re-code (Circuitscape 5.0) and later updates, we improved computational efficiency and parallelism, achieving major speed improvements, and enabling assessments across larger extents or with higher resolution data. Here, we reflect on the benefits to conservation of strengthening collaborations with computer scientists, and extract examples from a collection of 572 Circuitscape applications to illustrate how through a decade of repeated investment in the software, applications have been many, varied, and increasingly dynamic. Beyond empowering continued innovations in dynamic connectivity, we expect that faster run times will play an important role in facilitating co-production of connectivity assessments with stakeholders, increasing the likelihood that connectivity science will be incorporated in land use decisions.

The impact of landscape changes on the quality and connectivity of habitats for multiple wildlife species is of global conservation concern. In the southwestern United States, pumas (Puma concolor) are a well distributed and wide-ranging large carnivore that are sensitive to loss of habitat and to the disruption of pathways that connect their populations. We used an expert-based approach to define and derive variables hypothesized to influence the quality, location, and permeability of habitat for pumas within an area encompassing the entire states of Arizona and New Mexico. Survey results indicated that the presence of woodland and forest cover types, rugged terrain, and canyon bottom and ridgeline topography were expected to be important predictors of both high quality habitat and heightened permeability. As road density, distance to water, or human population density increased, the quality and permeability of habitats were predicted to decline. Using these results, we identified 67 high quality patches across the study area, and applied concepts from electronic circuit theory to estimate regional patterns of connectivity among these patches. Maps of current flow among individual pairs of patches highlighted possible pinch points along two major interstate highways. Current flow summed across all pairs of patches highlighted areas important for keeping the entire network connected, regardless of patch size. Cumulative current flow was highest in Arizona north of the Colorado River and around Grand Canyon National Park, and in the Sky Islands region owing to the many small habitat patches present. Our outputs present a first approximation of habitat quality and connectivity for dispersing pumas in the southwestern United States. Map results can be used to help target finer-scaled analyses in support of planning efforts concerned with the maintenance of puma metapopulation structure, as well as the protection of landscape features that facilitate the dispersal process.

Under rapid landscape change, there is a significant need to expand and connect protected areas (PAs) to prevent further loss of biodiversity and preserve ecological functions across broad geographies. We used a model of landscape resistance and electronic circuit theory to estimate patterns of ecological flow among existing PAs in the western United States. We applied these results to areas previously identified as having high conservation value to distinguish those best positioned to maintain and enhance ecological connectivity and integrity. We found that current flow centrality was highest and effective resistance lowest in areas that spanned the border between southern Oregon and Idaho, and in northern Arizona and central Utah. Compared to other federal jurisdictions, Bureau of Land Management lands contributed most to ecological connectivity, forming “connective tissue” among existing PAs. Our models and maps can inform new conservation strategies and critical land allocation decisions, within or among jurisdictions.

More consistent and current estimates of forest land cover type and forest structural metrics are needed to guide national policies on forest management, carbon sequestration, and ecosystem health. In recent years, the increased availability of high-resolution (<30 m) imagery and advancements in machine learning algorithms have opened up a new opportunity to fuse multiple datasets of varying spatial, spectral, and temporal resolutions. Here, we present a new model, based on a deep learning architecture, that performs both classification and regression concurrently, thereby consolidating what was previously several independent tasks and models into one stream. The model, a multi-task recurrent convolutional neural network that we call the Chimera, integrates varying resolution, freely available aerial and satellite imagery, as well as relevant environmental factors (e.g., climate, terrain) to simultaneously classify five forest cover types (‘conifer’, ‘deciduous’, ‘mixed’, ‘dead’, ‘none’ (non-forest)) and to estimate four continuous forest structure metrics (above ground biomass, quadratic mean diameter, basal area, canopy cover). We demonstrate the performance of our approach by training an ensemble of Chimera models on 9967 georeferenced (true locations) Forest Inventory and Analysis field plots from the USDA Forest Service within California and Nevada. Classification diagnostics for the Chimera ensemble on an independent test set produces an overall average precision, recall, and F1-score of 0.92, 0.92, and 0.92. Class-wise F1-scores were high for ‘none’ (0.99) and ‘conifer’ (0.85) cover classes, and moderate for the ‘mixed’ (0.74) class samples. This demonstrates a strong ability to discriminate locations with and without trees. Regression diagnostics on the test set indicate very high accuracy for ensembled estimates of above ground biomass ( $R^2=0.84,$ RMSE $=37.28$ Mg/ha), quadratic mean diameter ( $R^2=0.81,$ RMSE $=3.74$ inches), basal area ( $R^2=0.87,$ RMSE $=25.88$ ft ${}^2$ /ac), and canopy cover ( $R^2=0.89,$ RMSE $=8.01$ percent). Comparative analysis of the Chimera ensemble versus support vector machine and random forest approaches demonstrates increased performance over both methods. Future implementations of the Chimera ensemble on a distributed computing platform could provide continuous, annual estimates of forest structure for other forested landscapes at regional or national scales.

Landscape‐scale disturbance events, including ecological restoration and fuel reduction activities, can modify habitat and affect relationships between species and their environment. To reduce the risk of uncharacteristic stand‐replacing fires in the southwestern United States, land managers are implementing restoration and fuels treatments (e.g., mechanical thinning, prescribed fire) in progressively larger stands of dry, lower elevation ponderosa pine (Pinus ponderosa) forest. We used a Before–After/Control–Impact experimental design to quantify the multi‐scale response of avifauna to large (∼250–400 ha) prescribed fire treatments on four sites in Arizona and New Mexico dominated by ponderosa pine. Using distance sampling and an information‐theoretic approach, we estimated changes in density for 14 bird species detected before (May–June 2002–2003) and after (May–June 2004–2005) prescribed fire treatments. We observed few site‐level differences in pre‐ and posttreatment density, and no species responded strongly to treatment on all four sites. Point‐level spatial models of individual species response to treatment, habitat variables, and fire severity revealed ecological relationships that were more easily interpreted. At this scale, pretreatment forest structure and patch characteristics were important predictors of posttreatment differences in bird species density. Five species (Pygmy Nuthatch [Sitta pygmaea], Western Bluebird [Sialia mexicana], Steller's Jay [Cyanocitta stelleri], American Robin [Turdus migratorius], and Hairy Woodpecker [Picoides villosus]) exhibited a strong treatment response, and two of these species (American Robin and Hairy Woodpecker) could be associated with meaningful fire severity response functions. The avifaunal response patterns that we observed were not always consistent with those reported by more common studies of wildland fire events. Our results suggest that, in the short term, the distribution and abundance of common members of the breeding bird community in Southwestern ponderosa pine forests appear to be tolerant of low‐ to moderate‐intensity prescribed fire treatments at multiple spatial scales and across multiple geographic locations.