Exploring the Variance Components of Distance Samping to Simplify Adbundance Estimation of White- Tailed Deer
Gormezano L. 2004. Exploring the Variance Components of Distance Samping to Simplify Adbundance Estimation of White- Tailed Deer. SUNY College of Enrivonmental Science and Forestry
Distance sampling is a statistically robust approach to estimating abundance and its associated uncertainty. Learning the theory and software used to generate estimates, however, requires substantial time and effort. I used data from line transect surveys of white- tailed deer (Odocoileus virginianus) to explore relationships among the variance components of abundance in distance sampling (density, encounter rate, detection probability, cluster size) in order to develop a simplified method of analyzing line transect data. I developed a linear regression model to predict densitybased on the strong correlations observed between (1) encounter rate and cluster density and (2) effective strip half-width and mean perpendicular distance. The model was validated with new distance sampling data from geographically distinct deer and other mammal populations. For 56 of 67 surveys, the regression model yielded density predictions that were within 20% of estimates produced using DISTANCE software. Variance estimation techniques and model limitations are also discussed.
- Type
- Academic
- Authors
- Gormezano, Linda
- Date of Issue
- 2004-05
- Publisher
- SUNY College of Enrivonmental Science and Forestry
- Units
- FIIS
- Keywords
- Abundance, Density, Distance Sampling, population index, Thesis, White-Tailed Deer