TABLE 1.
List of covariates considered for inclusion in MaxEnt species distribution models for American and Pacific marten.
| Covariate | Layer processing | Source |
|---|---|---|
| Climate | ||
| Precipitation as snow (normals 1991–2020) | – | Climate WNA (~800 m) |
| Summer heat to moisture index (normals 1991–2020) | – | Climate WNA (~800 m) |
| Topography | ||
| Elevation | – | NASA SRTM GL3 DEM downloaded from Open Topography (90 m) |
| Slope | Calculated from elevation with 8 nearest neighbors | NASA SRTM GL3 DEM downloaded from Open Topography (90 m) |
| Topographic Position Index – 2000 m | Calculated from elevation (scale = 23 with a rectangular window) | NASA SRTM GL3 DEM downloaded from Open Topography (90 m) |
| Vegetation | ||
| Distance to forest edge | Distance from any pixel classified as EVT_LF = “Tree” | LANDFIRE EVT (30 m) |
| Forest Height | – | LANDFIRE EVH (30 m) |
| Lodgepole pine | Vegetation class 7050 converted to proportion within 1 km moving window. | LANDFIRE EVT (30 m) |
| Mixed conifer | Vegetation classes 7045, 7046, and 7166 converted to proportion within 1 km moving window. | LANDFIRE EVT (30 m) |
| Subalpine Spruce‐fir | Vegetation classes 7055 and 7056 converted to proportion within 1 km moving window. | LANDFIRE EVT (30 m) |
| Mesic Forest | Veg classes 7047 and 7056 converted to proportion within 1 km moving window. | LANDFIRE EVT (30 m) |
| Dry Forest | Vegetation classes 7045, 7053, 7055, and 7166 converted to proportion within 1 km moving window. | LANDFIRE EVT (30 m) |
| Riparian | Vegetation classes 9022 and 9019 converted to proportion within 1 km moving window. | LANDFIRE EVT (30 m) |
| Other | ||
| Distance from water | Distance from any pixel classified as water in “swnet” raster product. | NHD HR Plus (10 m) |
| Burns | Any burns since 1984 converted to a proportion within a 1 km moving window. | MTBS 2022 Burned Areas Boundaries (30 m) |
Note: Only climate and topographical variables were included in models used to predict future marten distribution. All vegetation layer processing was done in the R package SpatialEco. Mean annual precipitation, mean summer precipitation, precipitation as snow in the spring, mean coldest month temperature, and mean warmest month temperature were also considered as covariates, but were excluded from the final model because they were highly correlated with other variables.