An inexpensive and open‐source method to study large terrestrial animal diet and behaviour using time‐lapse video and GPS

Abstract

Abstract The behaviour of free‐ranging animals is difficult to study, especially on the large spatial and temporal scales relevant to long‐lived large species. Animal‐borne video and environmental data collection systems (AVEDs) record behaviour and other data in real time as animals conduct daily activities. However, few studies have combined systematically collected, long‐term AVED foraging data with environmental and movement data to test hypotheses on animal foraging. Additionally, AVEDs are often either prohibitively expensive, or require extensive fabrication and programming knowledge. The video and coordinate animal‐mounted system (VACAMS) is an animal‐mounted data collection system based on a modified GoPro® action camera platform that records short, first ‘person’ perspective videos of animal behaviour on an automated time‐lapse schedule. As most videos are georeferenced, researchers can return to the locations of specific behaviours and collect accurate, fine‐grained data on non‐woody vegetation and other habitat characteristics that may influence animal behaviour. Moreover, VACAMS are inexpensive and easy to use. This study describes VACAMS preliminary data on cattle foraging and a hypothesis exploring free‐ranging cattle browsing habits throughout the rainy season in the tropical dry forest of Sonora, Mexico. I generated a database of vegetation types consumed by cows each month (annual, woody and leaf litter) and compared actual vegetation type frequencies to a priori assumptions based on seasonal patterns of forage availability. During the monsoons, when palatable vegetation was abundant, frequencies of annual and woody perennial vegetation in cattle diets did not differ from month to month. When the rains ceased and palatable vegetation became scarce, cows switched to leaf litter, dead annual vegetation, twigs and dried leguminous fruits. Open‐source software and commercially available hardware make VACAMS financially attainable for many researchers, land managers, students and other user groups. VACAMS could be used on a range of domestic and semi‐domestic free‐ranging animals, particularly in dense forests where conventional observations are impossible. With improvements to GPS battery life and durability, the weakest points of the system, VACAMS could also potentially apply to studies of other large terrestrial animals.