Evaluating time-removal models for estimating availability of boreal birds during point count surveys: Sample size requirements and model complexity

The condor

¹ 

关键词: boreal birds;    population size;    abundance;    detectability;    bias–variance tradeoff;   

DOI  :  10.1650/CONDOR-18-32.1

学科分类：动物科学

来源: Central Ornithology Publication Office

【 摘 要 】

ABSTRACT We used conventional and finite mixture removal models with and without time-varying covariates to evaluate availability given presence for 152 bird species using data from point counts in boreal North America. We found that the choice of model had an impact on the estimability of unknown model parameters and affected the bias and variance of corrected counts. Finite mixture models provided better fit than conventional removal models and better adjusted for count duration. However, reliably estimating parameters and minimizing variance using mixture models required at least 200–1,000 detections. Mixture models with time-varying proportions of infrequent singers were best supported across species, indicating that accounting for date- and time-related heterogeneity is important when combining data across studies over large spatial scales, multiple sampling time frames, or variable survey protocols. Our flexible and continuous time-removal modeling framework can be used to account for such heterogen...

【 授权许可】

CC BY