Comparison of three different indirect methods to evaluate ungulate population densities

Abstract: Reliable species population estimates are important for population conservation and management. Information about population sizes is needed to introduce certain preservation and regulation approaches. Several different methods, both performed in the field and analytical, can be used to estimate species densities. The aim of this thesis was to evaluate densities of moose (Alces alces), red deer (Cervus elaphus), roe deer (Capreolus capreolus), and fallow deer (Dama dama) in a confined area in Västerbotten, Sweden, via three different non-invasive wildlife census methods: camera trapping, traditional dung pellet group counts, and DNA-based ID-ing of dung samples. Due to analytical problems in the lab, the latter method was excluded from any further analysis. Species densities were estimated via the Random Encounter Model (REM) from recorded pictures of camera traps. Further investigations from camera trap data were made to compare sex-ratio, day- and night-capture events, the influence of snow depth and the use of habitat type among the ungulate species. Density estimates of dung counts considered the defecation rate of each species and the accumulation period of dung. For moose, camera traps yielded a population estimate similar to dung counts. However, estimated densities from the dung count approach were much lower for all other three ungulate species than densities from the camera trapping. Even though the use of camera traps is more time consuming and costly, this method is evaluated as a trustworthy estimator of population sizes. In addition, further information about habitat associations, activity patterns, behaviour, or community structure can be successfully accomplished.