Candidate detection function models used to estimate population densities of NPSA landbirds (2018)
Judge SW, Camp RJ, Vaivai V, Hart PJ. 2021. Candidate detection function models used to estimate population densities of NPSA landbirds (2018). Pages 6. In Population density, distribution, and trends of landbirds in the National Park of American Samoa, Taʻū and Tutuila Units (2011–2018)
Following recommendations by Buckland et al. (2001), the half normal was paired with cosine and Hermite polynomial adjustments, and the hazard-rate was paired with cosine and simple polynomial adjustments. Model precision was improved by incorporating sampling covariates in the multiple covariate distance sampling (MCDS) engine of DISTANCE (Buckland et al. 2015). Covariates included cloud cover, rain, wind, gust, observer, time of detection, canopy cover, canopy height, and unit. All covariates were treated as a factor, except time of detection, which was treated as both a factor and a continuous covariate. Assessing time of detection as a continuous covariate helped to determine if the detection rate varied during the morning. Each detectability model in the candidate set was fit to data pooled across units for each species, and the model selected was that with the lowest 2nd-order Akaike’s Information Criterion corrected for small sample sizes (AICc) (Buckland et al. 2015). Data were truncated at a distance where the detection probability (using a preliminary detection function model) was about 0.1.
- Type
- Published Report Section
- Authors
- Judge, Seth; Camp, Rick; Vaivai, Visa; Hart, Patrick
- Units
- NPSA , PACN
- Keywords
- American Samoa, distribution, forest birds, landbirds, monitoring, population density, trends