Skip to main content
ARS Home » Pacific West Area » Burns, Oregon » Range and Meadow Forage Management Research » Research » Publications at this Location » Publication #173915


item Ganskopp, David

Submitted to: Rangeland Ecology and Management
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 12/27/2004
Publication Date: 5/20/2005
Citation: Ungar, E.D., Henkin, Z., Gutman, M., Dolev, A., Genizi, A., Ganskopp, D.C. 2005. Inference of animal activity from gps collar data on free-ranging cattle. Journal of Range Management. 58(3):256-266.

Interpretive Summary: In research of beef cattle grazing behavior, it is important to know the areas of a pasture that cattle use and what the animals are doing that may affect each location. Recently available geographic positioning system (GPS) collars contain motion sensors that can count the number of times an animal's head moves in various directions and commit those counts to memory along with the animal's position. This research evaluated the accuracy of using 2 different statistical methods (linear regression and tree regression analyses) to determine what activity (grazing, resting, or walking) beef cattle were pursuing as they moved about in extensive rangeland pastures over several days or weeks. Motion sensor counts and the distance between successive areas used by cattle accounted for about 85 percent of the animal's activity using linear regression methods. Linear regression methods also performed better when animal positions were determined at longer time intervals (every 20 minutes). Classification tree analyses could successfully identify beef cattle activities about 86 to 88 percent of the time and was more accurate when a cow's position was determined at shorter time intervals (every 5 minutes). These findings will mostly benefit livestock and wildlife researchers who are interested in both the long term movements of the animals they study and the impacts they may have upon the areas they choose to occupy.

Technical Abstract: Global Positioning Systems (GPSs) enable continuous and automatic tracking of an animal's position, but the full value of such information can be realized only if the corresponding activity of the animal is known. We tested the inference of animal activity by means of Lotek GPS collars fitted on beef cattle on extensive rangeland in two contrasting foraging environments. The collars were configured to record GPS fixes at intervals of 20 min (US) or 5 min (Israel), together with counts from 2 motion sensors. Synchronized field observations of collared cows were conducted in 1999 (US) and in 2002 and 2003 (Israel). Grazing, traveling (without grazing) and resting activities were recorded as minutes out of 20 for each category (US), or as a single category (Israel). For the US data, stepwise regression models of grazing, traveling and resting time accounted for 74-84% of the variation, based on the motion sensor counts for the left-right axis and the distances between GPS fixes. Regression tree analysis of grazing time yielded a simple model (four splits) that accounted for 85% of the variation. For the Israeli data, the misclassification rates obtained by discriminant analysis and classification tree analysis of animal activity were 0.14 and 0.12, respectively. In both analyses, almost all grazing observations were correctly classified, but other activities were sometimes misclassified as grazing. Distance alone is a poor indicator of animal activity, but grazing, traveling and resting activities of free-ranging cattle can be inferred with reasonable accuracy from data provided by Lotek GPS collars.