Comparison of Different Measures of the Error in Simulated Radio-Telemetry Locations

作者: David Saltz , Gary C. White

DOI: 10.2307/3808917

关键词:

摘要: We used computer simulations to compare 8 potential measures of linear error involved in estimating radio locations by triangulation. Linear is defined as the shortest or Euclidean distance (ED) between estimated triangulation point and actual location. a good measure one that highly correlated 95% quantile all possible ED's around given The major axis confidence ellipse (as determined Lenth's [1981] maximum likelihood estimator) performed best. Regression function parameters correlation coefficients varied with tower positioning standard deviation on bearing. recommend using for location studies. J. WILDL. MANAGE. 54(1):169-174 Heezen Tester (1967) first suggested importance locating an animal through radiotracking. Yet, no standards have been established estimate report these errors. Error provide information about precision (but not bias) data. Radio based result either areal Areal depict area will cover transmitter probability. probability be larger than equal (x^, ) (x, y) locations, i.e., ED = [(i x)2 + (y y)2]/. If habitat use primary objective, estimators are preferable because they specific patch. Alternatively, relocations, movement speed, home-range size influenced true from point; thus, would parameters. 90% polygon (Springer 1979) (White 1985) proposed Both employ sampling inherent bearing determination. By placing limits bearings, arc (Saltz Alkon (i.e., projected +2 SD). A delineated outer boundaries region overlap 2 such arcs. derived estimation (Lenth 1981). estimator (MLE) maximizes observations. limit parameter located 2-dimensional space. polygons ellipses has determined. Saltz (1985) superior recommended did test) large diagonal polygon. They argued (or ellipse) may nearly independent its dimensions. Thus, reduction necessarily accompanied corresponding change length diagonal. Furthermore, largest always represents any points polygon, it most sensitive therefore reliable ED. Our purpose was test set simulation. define high r2) value time. term this I Present address: Nature Reserve Authority, P. O. Box 340, Mizpeh Ramon, Israel.

参考文章(6)
D.W. MACDONALD, F.G. BALL, N.G. HOUGH, The Evaluation of Home Range Size and Configuration Using Radio Tracking Data A Handbook on Biotelemetry and Radio Tracking#R##N#Proceedings of an International Conference on Telemetry and Radio Tracking in Biology and Medicine, Oxford, 20–22 March 1979. pp. 405- 424 ,(1980) , 10.1016/B978-0-08-024928-5.50052-X
Mervyn G. Marasinghe, SAS System for Linear Models ,(1991)
Russell V. Lenth, On Finding the Source of a Signal Technometrics. ,vol. 23, pp. 149- 154 ,(1981) , 10.1080/00401706.1981.10486257
Joseph Tucker Springer, Some sources of bias and sampling error in radio triangulation Journal of Wildlife Management. ,vol. 43, pp. 926- ,(1979) , 10.2307/3808276
G. C. White, Optimal locations of towers for triangulation studies using biotelemetry Journal of Wildlife Management. ,vol. 49, pp. 190- 196 ,(1985) , 10.2307/3801870
Keith L. Heezen, John R. Tester, Evaluation of Radio-Tracking by Triangulation with Special Reference to Deer Movements The Journal of Wildlife Management. ,vol. 31, pp. 124- ,(1967) , 10.2307/3798367