Habitat Selection Scenarios for Molting Waterfowl in the Goose Molting Area of the Teshekpuk Lake Special Area, for NPR-A Integrated Activity Plan/Environmental Impact Statement (2020)

Vector polygons of lakes in the TLSA were combined into molt-unit polygons based on observations of usage patterns by flocks of molting waterfowl during aerial surveys (USFWS Teshekpuk Lake Molting Goose Surveys 2011-2016). Molt units were ranked according to population density of target species (Pacific Black Brant or Canada Geese), and selected in order of decreasing population density until a specified percentage of the molting population was included. Population density estimates were derived from the average number of individual birds seen using a molt unit during aerial surveys between 2011-2016 (USFWS Teshekpuk Lake Molting Goose Surveys 2011-2016).Once selected, molt units were bounded by a half-mile (800 m) or 1-mile buffer (1600 m), depending on the scenario. The half-mile minimum buffer width was determined based on the area of terrestrial habitat around molt unit lakes that was used by foraging Pacific Black Brant equipped with GPS transmitters (Lewis et al. 2011). The one-mile buffer was included based on the expectation that molting geese can respond to visual or auditory disturbance at a distance. However, the minimum buffer distance required to eliminate disturbance effects is unknown, and the one-mile buffer width should not be interpreted as having eliminated the potential for disturbance within selected areas. All selected molt unit polygons and their buffers were dissolved into a single layer, and merged with two additional pre-defined polygons (a 3-mile wide buffer along the shore of Teshekpuk Lake, and a 1-mile wide buffer along the Beaufort sea coast, including coastal features such as bays and estuaries. The merged polygons representing each management scenario (target species + population threshold + buffer width) were combined into a single shapefile using GIS tools in the R statistical environment.