Analyzing Animal Movement Characteristics From Location Data

When individuals of a species utilize an environment, they generate movement patterns at a variety of spatial and temporal scales. Field observations coupled with location technologies (e.g. GPS tags) enable the capture of detailed spatio-temporal data regarding these movement patterns. These patterns contain information about species-specific preferences regarding individual decision-making, locational choices and the characteristics of the habitat in which the animal resides. Spatial Data Mining approaches can be used to extract repeated spatio-temporal patterns and additional habitat preferences hidden within large spatially explicit movement datasets. We describe a method to determine the periodicity and directionality in movement exhibited by a migratory bird species. Results using a High Arctic-nesting Svalbard Barnacle Goose movement data yielded undetected patterns that were secondarily corroborated with expert field knowledge. Individual revisits by the geese to specific locations in the breeding and wintering grounds of Svalbard, Norway and Solway, Scotland, occurred with a periodicity of 334 days. Further, the orienta-tion of this movement was detected to be mostly north-south. During long-range migration the geese use the north-south oriented Norwegian islands as “stepping stones”, Short-range movement between mudbank roosts to feeding fields in Solway also retained a north-south orientation.