Remote Sensing in Movement Ecology: Scale Dependencies of Space Use

Abstract

"Movement is an essential variable in animal ecology. It provides crucial information on individual and collective behavior and is the basis for efficient conservation planning of e.g. migratory species. To fully understand movement it is essential to consider the underlying environmental conditions. Animal movements are strongly influenced by the presence and distribution of resources. In this context, remote sensing has gained particular attention. Earth Observation (EO) data supports the monitoring of surface conditions providing a larger perspective on regional processes on multiple spatial and temporal scales. While remote sensing has been intensively used in the scope of movement ecology its application can be improved to take full advantage of the information acquired through EO. Existing applications focus on pixel based information aiming for the spatial and temporal matching of satellite and tracking data. However, this fails to provide an accurate description of the individual’s perception of space and of the interactive learning process that guides resource selection. Additionally, the direct integration of remote sensing data disregards the dynamic nature of the landscape. Spatial dissimilarities are often hidden when considering temporally invariant information. Selecting adequate spatial, temporal and thematic scales becomes thus essential. Following this premise this research will discuss the impact of scale on the analysis of animal space use and in the consequent understanding of resource selection. It will provide an assessment of the explanatory power of space use comparing the selection of different spatial, temporal and thematic resolutions. Movement data for a population of White Storks (cicogna cicogna) collected in the scope of the LifeTrack program is used as a reference for this analysis. Relevant remote sensing variables are derived from medium spatial resolution and high temporal resolution optical data as well as existing land cover information. Additionally, the integration of Synthetic Aperture Radar (SAR) based information is also tested in the scope of this research."