Day or night? Testing thermal imaging technology for estimating ungulate population densities in southern Sweden

Abstract: To be able to make a proper judgement and choose the best management or conservation solution when working with populations of wild ungulates or any other wildlife, it is important to obtain good estimates of population densities. Inventing or improving wild animal monitoring methods is a key to protect what may be endangered or to avoid human-wildlife conflicts. That thought has inspired me to test an advanced, modern equipment to investigate whether upgrading traditional point transect monitoring method by using a thermal imaging device and conducting the observations during the night will show a potential to become more implemented in wildlife monitoring. For my study location I have chosen Öster Malma area in southern Sweden. I decided to focus on ungulates, with fallow deer being the main target. Night observations with thermal imaging binoculars Pulsar Merger LRF XP50 were compared with day observations with day optics binoculars Minox X-range 10x42. Collected data were also compared with data from dung counts conducted through other projects in the same study area. I have specifically tested four hypotheses: more animals in general and more different species will be observed during the night observations than during day observations; fallow deer will be the most spotted species; there will be a correlation between data collected during my observations and the dung counts and; that the density estimates from the night observations will be more accurate than the ones based on day observations. The results confirmed that there were more successful observations during the night observations. Fallow deer occurred to be the most spotted species in the area. There was a correlation between data from dung counts and data from day and night point transects observations and this correlation occurred to be stronger with the day observations than with the night observations. Surprisingly the density estimates, for the average number of clusters, after distance sampling analysis in R was similar for both day and night observations, but overall the estimates are more accurate for the data from night observations. In general, the findings support that implementing night wildlife monitoring with usage of thermal imaging technology can improve monitoring methods and provide more detailed information.