E de la Garza Institute for Goat Research Langston University
Workshops & Field Day Newsletter Newsletter Subscription Demonstrations Demonstrations Langston University Research Building
Goat Menu




Accuracy of calculated distances between consecutive fixes of GPS collars worn by goats

T. A. Gipson, G. Detweiler, and A. L. Goetsch

Journal of Animal Science 88(E-Supplement 2):458-459. 2010

Small ruminants have been fitted with GPS collars to estimate distance traveled in grazing studies; however, accuracy has not been assessed. To do so, a mobile stand was developed to hold 21 Lotek 3300 GPS collars (Lotek Wireless, Newmarket, Ontario, Canada) and was moved a prescribed distance between fixes on four azimuthally different courses (NE at 45?, S at 180?, W at 270?, and NW at 315?). Fixes were scheduled at 5-min intervals. Distances traveled on a course were 0, 10, 20, 30, 40, 50, 60, 70, 80, 90, and 100 m. Distances were replicated three times for each course and the order of the distances was randomized within each replicate. Two courses were run per day and each course was paired with every other course, for a total of six different dates. Fixes were downloaded and distances between fixes were calculated using spherical geometry. The BIAS was estimated as distance calculated from collars minus true distance and was analyzed using a repeated measures design (PROC MIXED; SAS). Dependent fixed effects included true distance (0 to 100 m), course (NE, NW, S, W), and the two-way interactions. Collars and dates were considered random effects. There was no effect (P > 0.10) of course on BIAS. For true distance, BIAS was greatest for 0 m (7.6 m 0.36) and least for 60 m (0.5 m). Other estimates of BIAS were intermediate at 10 (2.9 m), 20 (1.6 m), 30 (0.9 m), 40 (1.7 m), 50 (1.4 m), 70 (0.6 m), 80 (0.8 m), 90 (1.2 m), and 100 m (1.3 m). There were linear and quadratic (P < 0.01) effects on BIAS when all distances were analyzed. However, if 0 and 10 m distances were eliminated, the linear effect disappeared (P > 0.10) but the quadratic remained (P < 0.05). The ability of GPS collars to differentiate between when an animal is stationary or moving only a short distance between fixes appears very limited; however, if an animal is moving more than 20 m between GPS fixes, collar estimates are within 1.6 m of actual distance traveled.


Extension Activities   |   Research Activities   |   Other Activities
Library Activities   |   Quiz   |   Search   |   About Us   |   Contact Us   |   Faculty & Staff
Research Extension Home   |   Top of Page

Copyright© 2000 Langston University   • Agricultural Research and Extension Programs
P.O. Box 730  • Langston, OK  73050 • Phone 405.466.3836