Comparison of Three Plot Selection Methods for Estimating Change in Temporally Variable, Spatially Clustered Populations.
- Bonneville Power Administration, Portland, OR (US). Environment, Fish and Wildlife
Monitoring population numbers is important for assessing trends and meeting various legislative mandates. However, sampling across time introduces a temporal aspect to survey design in addition to the spatial one. For instance, a sample that is initially representative may lose this attribute if there is a shift in numbers and/or spatial distribution in the underlying population that is not reflected in later sampled plots. Plot selection methods that account for this temporal variability will produce the best trend estimates. Consequently, I used simulation to compare bias and relative precision of estimates of population change among stratified and unstratified sampling designs based on permanent, temporary, and partial replacement plots under varying levels of spatial clustering, density, and temporal shifting of populations. Permanent plots produced more precise estimates of change than temporary plots across all factors. Further, permanent plots performed better than partial replacement plots except for high density (5 and 10 individuals per plot) and 25% - 50% shifts in the population. Stratified designs always produced less precise estimates of population change for all three plot selection methods, and often produced biased change estimates and greatly inflated variance estimates under sampling with partial replacement. Hence, stratification that remains fixed across time should be avoided when monitoring populations that are likely to exhibit large changes in numbers and/or spatial distribution during the study period. Key words: bias; change estimation; monitoring; permanent plots; relative precision; sampling with partial replacement; temporary plots.
- Research Organization:
- Bonneville Power Administration, Portland, OR (US)
- Sponsoring Organization:
- US Department of Energy (US)
- OSTI ID:
- 785591
- Report Number(s):
- DOE/BP-25866-10; Contract 1992AI25866; TRN: AH200131%%392
- Resource Relation:
- Other Information: PBD: 1 Jul 2001
- Country of Publication:
- United States
- Language:
- English
Similar Records
Field evaluation of remote wind sensing technologies: Shore-based and buoy mounted LIDAR systems
Seismic Analysis of Spatio-Temporal Fracture Generation During EGS Resource Development