Geolocation by light: accuracy and precision affected by environmental factors
1.Geolocation by light allows for tracking animal movements, based on measurements of lightintensity over time by a data-logging device (‘geolocator’). Recent developments of ultra-lightdevices (<2 g) broadened the range of target species and boosted the number of studies using geolocators. However, an inherent problem of geolocators is that any factor or process that changes thenatural light intensity pattern also affects the positions calculated from these light patterns.Although the most important factors have been identified, estimation of their effect on the accuracyand precision of positions estimated has been lacking but is very important for the analyses andinterpretation of geolocator data. 2.The ‘threshold method’ is mainly used to derive positions by defining sunrise and sunset timesfrom the light intensity pattern for each recorded day. This method requires calibration: a pre-defined sun elevation angle for estimating latitude by fitting the recorded day⁄night lengths to theo-retical values across latitudes. Therewith, almost constant shading can be corrected for by findingthe appropriate sun elevation angle. 3.Weather, topography and vegetation are the most important factors that influence lightintensities. We demonstrated their effect on the measurement of day⁄night length, time of solarmidnight⁄noon and the resulting position estimates using light measurements from stationarygeolocators at known places and from geolocators mounted on birds. Furthermore, we investi-gated the influence of different calibration methods on the accuracy of the latitudinal positions. 4.All three environmental factors can influence the light intensity pattern significantly. Weatherand an animal’s behaviour result in increased noise in positioning, whereas topography and vegeta-tion result in systematic shading and biased positions. Calibration can significantly shift the esti-mated latitudes and potentially increase the accuracy, but detailed knowledge about the particularconfounding factors and the behaviour of the studied animal is crucial for the choice of the mostappropriate calibration method.