USGS - science for a changing world

Northern Prairie Wildlife Research Center

  Home About NPWRC Our Science Staff Employment Contacts Common Questions About the Site

Accuracy of Aerial Telemetry Locations in Mountainous Terrain

Glen F. Gantz, InterWest Wildlife and Ecological Services, Inc., 11255 North 2000 East, Richmond, UT 84333 USA and L. Charles Stoddart, Denver Wildlife Research Center, Animal and Plant Health Inspection Service, U.S. Department of Agriculture, Utah State University, Logan, UT 84322-5295 USA

Aerial telemetry is commonly used to locate wildlife in remote areas (Gilmer et al. 1981, U.S. Fish and Wildl. Serv. Resour. Publ. 140; Gantz 1990, M.S. Thesis, Utah State Univ., Logan). Typically, aerial locations are subject to 2 sources of error. First, if transmitter-equipped animals are not sighted from the aircraft, their ground location must be estimated. Second, error arises when ground locations are transferred to a map.

We conducted the accuracy tests in the Bear River mountain range in northern Utah and southeastern Idaho. Elevation ranged from 1425 to 3042 m. Topography included steep mountains, deep, narrow canyons, high mountain valleys, and flats. We attached a 3-element yagi antenna aft of each wing strut of a Cessna 182 using custom-made strut brackets. Antennas were mounted symmetrically with the main beams perpendicular to the fuselage, elements oriented vertically, and the front of the beam angled 8 toward the ground. Coaxial leads were connected to a custom-made, solenoid-activated coaxial switch on the plane's steering yoke. Common output from the switch was coupled to a Telonics TR-2 receiver (Mesa, AZ, USA). The observer received telemetry signals through a portable intercom (Sigtronics Transcom II, Covina, CA, USA) and David Clark H10-30 headsets Worcester, MA, USA). The observer (principal investigator) was also the pilot.

For our test, an assistant placed transmitters on the study area and marked the locations on a USGS 7.5-minute topographic map using topographical features as references. To locate a transmitter once a radio signal was detected, the observer first aligned the plane toward the transmitter by orienting the plane so the antenna signal strengths were equal. Equal signal strengths were maintained by adjusting the heading while approaching the transmitter. Approach altitude was kept as low as safely possible (10-300 m), dictated by topography, wind, turbulence, and pilot experience. Passage over the transmitter was noted by a pronounced peak in signal strength. Subsequent approaches were flown from different directions as described by Mech (1983, Univ. Minnesota Press, Minneapolis, MN). The estimated ground location was recorded on a USGS 7.5-minute topographic map using available topographical features as references.

Mean error was 89 m (SE = 13, n = 39). Among terrain types, only the mean errors for ridge (Mean of X = 45 m, SE = 19, n = 8) and canyon (Mean of X = 114 m, SE = 13, n = 9) locations were significantly different (P = 0.01). The number of passes per transmitter ranged from 3 to 7. Mean numbers of passes among terrain types varied <1 pass and were not significantly different (P > 0.23), nor was there a significant correlation between number of passes and error (P = 0.42). Observer prediction of how well he estimated a transmitter location was significantly correlated with error (r2 = 0.97, t = 8.04, df = 2, P < 0.01). Flights ranged between 2.2 and 2.8 hr during the accuracy test. We found no correlation between error and flight time (r2 = 0.002, P = 0.45) in this range of flight times. Accuracy was generally poorest for transmitters located in canyons and on side hills.

Because we flew in good weather for this accuracy test, we think our mean error of 89 m (SE = 13, n = 39) was a good estimate of the maximum capability of our aerial telemetry system and observer in mountainous terrain.

When we first tested the accuracy of our aerial telemetry system during a pilot study conducted in December 1987, the observer had only 12-17 hr of radio locating, mountain flying, and Cessna experience. The mean error for the pilot study was 214 m (SE = 38, n = 12). The significantly improved performance in the accuracy test was after an additional 250 hr of locating transmitters in mountainous terrain.

Previous Section -- An expandable telemetry collar for calf elk in Colorado
Return to Contents
Next Section -- Foraging and reproductive behavior of Indian flying fox (Pteropus giganteus)as determined by radio telemetry

Accessibility FOIA Privacy Policies and Notices

Take Pride in America logo logo U.S. Department of the Interior | U.S. Geological Survey
Page Contact Information: Webmaster
Page Last Modified: Saturday, 02-Feb-2013 07:28:45 EST
Menlo Park, CA [caww55]