|Trend Evaluation of the Rocky Mountain Bird Observatory's Colorado Breeding Bird Data, 1998-2007
| Amy Davis, Master's Candidate, Department of Statistics
Colorado State University
April 11, 2008
Birds are integral components of their ecosystems, and declines in bird numbers can provide early warnings of problems in ecosystem health. Therefore, it is important to be able to detect even slight declines over time in bird densities. The Rocky Mountain Bird Observatory (RMBO) has been monitoring Colorado's breeding birds since 1998. This paper uses ten years of RMBO monitoring data to investigate the densities of four bird species in four Colorado habitats and to develop techniques to analyze time trends.
Program Distance is used to compute density estimates while accounting for detectability problems inherent in wildlife populations. The densities are then examined by sampling unit, transect, and transformed to fit the assumptions of normality and homoscedasticity for mixed model analysis. The mixed model included the following fixed effects: linear year, habitat, and linear year by habitat interaction, as well as the following random effects: transect nested within habitat, a categorical year component, and a categorical year by habitat interaction. Three different error structures were considered for the model, and the best was selected using AICc. Also, a series of parametric and nonparametric tests were used to evaluate trends. Based on these methods, there is no significant change in densities for the four bird species examined over the 10 years studied.