Geostatistical analysis of GPS trajectory data: Space-time densities
Creation of density maps and estimation of home range is problematic for observations of animal movement at irregular intervals. We propose a technique to estimate space-time densities by separately modeling animal movement paths and velocities, both as continuous fields. First the length of traject...
| Main Authors: | , , , |
|---|---|
| Other Authors: | , |
| Format: | Article in Journal/Newspaper |
| Language: | unknown |
| Published: |
2008
|
| Subjects: | |
| Online Access: | https://dare.uva.nl/personal/pure/en/publications/geostatistical-analysis-of-gps-trajectory-data-spacetime-densities(65045eb8-61f3-4651-b783-e8a4ec12c6e4).html http://www.spatial-accuracy.org/2008/PDF/Hengl2008accuracy.pdf |
| Summary: | Creation of density maps and estimation of home range is problematic for observations of animal movement at irregular intervals. We propose a technique to estimate space-time densities by separately modeling animal movement paths and velocities, both as continuous fields. First the length of trajectories for a given grid is derived; then the velocity of individual birds is interpolated using 3D kriging; finally the space-time density is calculated by dividing the density of trajectories (total length of lines per grid cell) by the aggregated velocity at that grid cell. The resulting map shows density of a species in both space and time, expressed in s/m2 units. This length-by-velocity (LV) technique is illustrated using two case studies: (1) a synthetically generated dataset using the Lorenz model; and (2) GPS recordings of 14 individual birds of lesser black-backed gull (Larus Fuscus). The proposed technique is compared with kernel smoother - a technique commonly used to derive home range for species. The results of using a synthetic dataset proved that the LV method produces different outputs than kernel smoothing, especially if irregular observation intervals are used. The main advantages of the proposed technique over a kernel smoother are: (1) it is not sensitive to missing observations; (2) it is suited to analyze fly paths (e.g. it preserves information about velocities and directions), and (3) it allows the movement of birds (velocity, trajectory) to be modeled separately e.g. as function of environmental conditions, wind, day time and similar. The remaining research issues are development of methodology for selection of optimal grid size and optimal time interval between recordings. |
|---|