Calving season habitat selection of maternal and non-maternal female moose in Southwest Alaska, U.S.A

We predicted habitat selection of maternal and non-maternal female moose according to a use-available design with animal paths as the sampling unit. To make predictions while assessing the uncertainty from randomness in our representation of available habitat, we created a meta-model for each group from 50 independent random forest models each trained on a different set of available paths (but with the same used paths). We trained all models relative to 16 raster covariates representing topographic and plant community characteristics that we expected to be important to moose during calving season. All covariate rasters are provided in this data archive. The resulting habitat selection models successfully distinguished between used and available paths with cross-validation accuracies of 88% for non-maternal moose and 77% for maternal moose. In addition to habitat selection rasters, we provide uncertainty rasters as the per pixel 95% confidence interval width and binary significance rasters where pixels were significant (value = 1) if the 95% confidence interval did not cross 0 (i.e., from selection to avoidance or vice versa). The spatial prediction rasters enabled us to interpret selection patterns relative to landscape structure.

Methods
The continuous foliar cover datasets included in this data package are derived from the Continuous Foliar Cover Maps of Plant Species and Aggregates in North American Beringia Version 1.0 (https://doi.org/10.5281/zenodo.4770218). Alpine barrens were derived from the National Land Cover Database 2016 (https://www.mrlc.gov/data/nlcd-2016-land-cover-alaska). A lake raster was generated from lake features in the National Hydrography Dataset (version 2021; https://www.usgs.gov/national-hydrography/national-hydrography-dataset). The topographic datasets included in this data package are derived from the USGS 3DEP 5m IFSAR DEM (https://gis.data.alaska.gov/pages/elevation). We extracted all input and covariate rasters to our Southwest Alaska study area. Vegetation cover and height data were collected in the vicinity of the Nuyakuk and Nushagak rivers in 2019. For maternal and non-maternal females, we provide the following model outputs in the form of raster datasets: habitat selection, uncertainty (95% confidence interval width), and binary significance. Detailed information on the processing of all provided data is available through the code repository linked in the code archive (). Geospatial data are provided as TIF single band rasters with 10 x 10 m resolution projected into Alaska Albers Equal Area Conic (EPSG 3338). The data package contains optional .lyrx files for visualization of model output rasters in ArcGIS Pro.