Correcting spatial bias in wildlife citizen-surveys: Integrating manatee sighting reports with GPS tag data

Poster created and presented in 2012 Abstract Species occurrence data collected by citizen volunteers or mined from historical records can supplement species distribution studies by adding dimensions of spatial and temporal survey coverage that may not be cost-effective or even possible for researchers to obtain otherwise. In the absence of rigorous survey or quality assurance methods, most citizen-collected or historic occurrence data may be biased in space to favor human population centers: species in areas of high human activity may be oversampled, while species in relatively unpopulated areas may not be adequately accounted for. We gathered citizen-reported occurrence data for the West Indian manatee (Trichechus manatus) in coastal Alabama (AL), USA from 2007 to 2012. We found significant differences in spatial distribution between citizen-reported sightings and reference locations derived from GPS tag data from six manatees in the study area. To correct for potential spatial biases in the citizen-reported dataset, we weighted each sighting report using a) population density of closest US Census block (2010), and b) aquatic distance to closest boat ramp. After correcting for spatial bias, the spatial distribution of citizen-reported sightings was closer to reference GPS locations. These results provided a foundation to enhance our knowledge of manatee distribution in AL by supplementing GPS tag data with unbiased sighting report data not explicitly detected by GPS tags, such as manatee group size and behavior. These results demonstrate that species occurrence data collected under different methodologies may be corrected and integrated to form a more robust depiction of location and movement patterns for the species being surveyed.