Seasonal variation in the activity pattern of red squirrels and their mammalian predators

Camera traps are a powerful tool to monitor the activity pattern of species over long time periods. Camera data can be used to analyze activity patterns of or temporal niche partitioning among species. Temporal niche partitioning plays an important role for species coexistence and is under constant evolutionary pressure from predator-prey relationships. Our study aimed to investigate temporal shifts in activity patterns of red squirrels and two mammalian predators, red fox and marten (Martes spec.). Previous studies yielded different activity patterns for these species depending on location, season, predator-prey interactions, and food- availability. We used camera traps to monitor the activity of squirrels, red fox and martens throughout a whole year in a mountainous forest in southwestern Germany. We also investigated a possible difference in activity among different color morphs as coloration in mammals is often associated with concealment, for squirrels, however, such an effect could not be demonstrated so far. We found a diurnal activity for squirrels in all seasons with a bimodal activity pattern during spring, trimodal in summer and unimodal in fall and winter. Activity patterns did not differ between black and red color morphs. The activity of the squirrels showed only low to moderate overlaps with fox and marten, but there was still slight variation among seasons. Activity overlaps were higher in spring and summer, which coincides with breeding season. Our results suggest that predators might adjust their activity during squirrel breeding season to increase the chance of preying on their offspring. Squirrels in turn could benefit by minimizing the activity overlap during wintertime when they are highly visible to predators as trees are leaf-free or even snow might enhance the contrast between them and their background. Lastly, our results indicate that temporal niche portioning among red squirrel, red fox and martens might be rather fine scaled.