A New BEAST: Bayesian Software Tools for Ecological Trend Analysis

Abstract

Many software tools available to biologists are either case-specific or require the ability to write computer programs. The learning curve for newer Bayesian approaches is much steeper than traditional approaches, often requiring an investment of several years. This article introduces a new free software package, BEAST (Bayesian Ecological Analysis of Statistical Trends), for Bayesian estimation of population trends, which requires the user to do no programming. A wide range of single-species population dynamics models are supported, including single-stage population models, stage-structured population models, and age-structured population models. In addition to Bayesian parameter estimation, this software allows users to do population projections, and may be used as a teaching tool by agencies and universities. I provide an example of a theta-logistic population model of stock pigeon (Columba oenas) breeding pairs. The degree of density dependence for this population was found to be stronger when compared to maximum likelihood parameter estimates available from the literature. Short-term population projections indicate that the number of breeding pairs is relatively stable, indicating that the population is at (or near) carrying capacity. This result is in agreement with the most recent data available, which indicate the breeding counts have remained relatively constant.