Premium
Evaluation of a Density Estimator Based on a Trapping Web and Distance Sampling Theory
Author(s) -
Wilson Kenneth R.,
Anderson David R.
Publication year - 1985
Publication title -
ecology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.144
H-Index - 294
eISSN - 1939-9170
pISSN - 0012-9658
DOI - 10.2307/1939171
Subject(s) - estimator , statistics , confidence interval , range (aeronautics) , trapping , sampling (signal processing) , monte carlo method , variance (accounting) , mathematics , population , distance sampling , sample size determination , ecology , computer science , biology , materials science , demography , accounting , filter (signal processing) , sociology , business , composite material , computer vision
A new density estimation technique using a trapping web and distance sampling theory was evaluated using Monte Carlo methods. A sophisticated computer algorithm simulated movement of small mammals according to four different home range utilization distributions. Trapping of small mammal populations was simulated with four spatial patterns, two population densities, and three realistic average capture probabilities; in total, 8151 replications were evaluated. Percent relative bias of the estimator of density ranged from —4 to 29% for average capture probabilities (°) of .16 and .24, and from —17 to 30% at ° =.09. Average coefficients of variation were 7.8—9.8% at a density (D) of 100 individuals/ha and 15.1—19.1% at D = 25 individuals/ha. Achieved confidence interval coverage for the density estimator ranged from 32 to 94%, under the assumption of a 95% nominal level. An improved variance estimator for the trapping web was developed that should improve the confidence interval coverage. The trapping web approach appears very promising in aiding the manager or researcher to obtain reliable density estimates for small mammal populations. The statistical theory underlying this estimation method is well developed and is based on relatively few assumptions.