USING A CONSUMER-GRADE UAV FOR A BROAD-SPATIAL SCALE ESTIMATION OF THE UNCERTAINTY IN GNSS-BASED TELEMETRY DATA

摘要

GNSS-based telemetry has been used to track the movement of wild animals. However the location error in GNSS-based telemetry data has been a concern as to the potential effects on subsequent analyses of movement patterns, space use, and resource selection. We examined a UAV-based method to estimate the degree of uncertainty in GNSS-based telemetry data over a broad spatial scale. The Digital Surface Model (DSM) was derived from a set of high-spatial-resolution aerial images acquired from a consumer-grade UAV and the Structure-from-Motion and Multi-View-Stereo (SfM-MVS) techniques. The DSM was used to calculate two variables: (1) the Morphometric Protection Index (MPI); and (2) the Elevation Mask (EM). First the comparability between the MPI and the Canopy Openness (CO), which has been commonly used as the first-order approximation of the GNSS signal interruption probability, was evaluated based on a simple linear regression model, where the MPI was used as predictor variable and the CO as response variable. The result showed that the CO could be predicted from the MPI over a broad spatial range. Second the Geometrical Dilution of Precision (GDOP) and number of satellites (NSAT) were predicted using the EM and GSILIB software. To account for the effect of GNSS signal transmission through an object or gaps among objects on the ground surface, the EM were derived from the normal DSM and the quasi-DSM which was identical to the normal DSM except for tree canopies. To test if both of the GDOP and NSAT have correlation to the uncertainty in GNSS locations, both indexes were compared to the mean error and standard deviation of the GNSS locations observed at several fixed points. As a result, the null hypothesis that the population correlation equals zero was rejected with statistical significance (p-value < 0.0005). We concluded that both of the GDOP and NSAT predicted from UAV-derived geospatial data products would be useful to estimate the degree of uncertainty in GNSS-based telemetry data over a broad spatial scale.