Evaluation of estimation quality of a general paradigm for indexing animal abundance when observations are counts

摘要: article Relativeabundance indices are widely applied to monitor wildlife populations. A general indexing paradigm was developed for structuring data collection and validly conducting analyses. This approach is applicable many observation metrics, with observations made at stations through the area of interest repeated over several days. The variance formula index derived using a linear mixed model, statistical tests confidence intervals constructed assuming Gaussian-distributed observations. However, methods, like intrusions track plots or camera traps, involve counts zeroes, producing Poisson- To fill this inferential gap between Gaussian analytical assumptions Poisson-distributed we evaluated, via broad Monte Carlo simulation study, estimation interval cover- agewhenGaussianstatisticalinferenceisappliedtodatageneratedfromaPoissondistribution.Themixedeffects model performed well in estimating variances when simulated Poisson were range found field studies(88-96% coverage). Es- timation improved by increasing number Confidence coverage rates verywell (even fewobservation days) whenday-to-dayvariability small,while effectiveestimation re-