Multiple scale analysis of whooping crane habitat in Nebraska

Geographic Information System (GIS) and remote sensing technologies were used to evaluate whooping crane stopover habitat in Nebraska. The goal of the research was to investigate habitat selection at multiple spatial scales and to determine if habitat selection varied over the time period under investigation. The GIS database consisted of all confirmed whooping crane sightings reported in Nebraska from 1975–1996 and land cover information delineated from color infrared aerial photographs and Landsat Thematic Mapper data. Results suggest that whooping cranes select roost habitat by recognizing local- and larger-scale land cover composition. Wetland is the most strongly selected habitat type at all spatial scales examined. There were similarities and differences in landscape structure among the areas most commonly used by cranes (Central Platte River, Western Rainwater Basin, Central Table Playas). Whooping crane habitat selection was influenced by social group (family, non-family), season (spring, fall), and landscape pattern. Temporal change in habitat within all study areas included increases in grassland and decreases in woodland. Percent wetland habitat was different between time periods studied, but the differences noted were attributed to variation in response to climatic conditions rather than actual habitat change. Results of two approaches for investigating habitat selection (i.e., differences between habitat use and availability) are compared. It is recommended that “availability” measurements of stopover habitat for cranes include designation of random or non-use sites, especially if analyses are conducted at more than one spatial scale.