Core Habitat
We modeled Core Areas for each target the same way, using an approach developed for the Washington Connected Landscapes project and implemented in the Gnarly Landscape Utilities ArcGIS toolbox (Shirk et al. 2010). First, a moving window average was applied to the habitat model, with a threshold to identify local areas that have high average local habitat quality. The radius of the moving window and the thresholds varied by target and are listed in the table below. For each target, two different moving window thresholds were used to create a set of higher quality areas (the Core class in our spatial priorities hierarchy) and lower quality areas (our Growth Opportunity Areas; GOAs). In this way, Core Areas are nested within GOAs. Within each Core Area or GOA, all pixels with a resistance greater than 5 and all pixels with a habitat quality less than the threshold were removed.
Next, we used the target’s resistance model to calculate the cost-weighted distance to the nearest valid pixel in the Core Areas and GOAs, and then applied a threshold to that cost-distance at a distance approximating a home-range type movement (see table below for the distance threshold, which varied by target). This links nearby patches together unless they are sufficiently far apart in cost-distance that animal movement would be considered dispersal (that is where we map corridors). Finally, all Cores and GOAs that were below the minimum size threshold (also target specific; see table below) were removed.
Because we used moving windows and local movement neighborhoods to define Cores and GOAs, it is possible they contain pixels that are not currently classified as habitat for the target. This is intentional, as animals within Cores and GOAs are likely to move through these areas within their home ranges. These non-native habitats within Cores and GOAs are critical locations to restore or manage in a way that promotes or increases habitat quality within that area.
Last updated