Effects of spatio-temporal weather conditions in autumn and winter on body mass and behaviour of the high Arctic Svalbard reindeer (Rangifer tarandus platyrhynchus)

Abstract

There exists large consensus within the research community that climatic changes have a world-wide impact on ecosystems, which is further expected to be most prominent in the Arctic. In these strongly seasonal environments, one key aspect is that global warming affects the duration of seasons. The focus on climate research has principally been directed towards the earlier coming of spring, meanwhile autumn is perhaps a neglected season. A natural consequence of warmer autumn seasons, is a delay in the onset of winter snow cover. This may in turn cause beneficial effects for herbivores, due to a prolonged period with good foraging conditions. In contrast, the increase in rain-on-snow events due to warmer winters may restrict access to forage as the rain refreezes on the ground or inside the snow pack. This may lead to contrasting effects on body mass, which responds continuously to changes in environmental conditions and influence both survival and recruitment parameters in herbivores.

In this study, GPS-data of marked adult Svalbard reindeer (Rangifer tarandus platyrhynchus), were combined with spatio-temporally explicit maps of air temperature, snow density and snow depth to increase our mechanistic understanding of climate change effects in Svalbard reindeer. I followed 35 individual female reindeer, which resulted in 96 individual reindeer years through 7 winter seasons between 2009 and 2016. The aim was to investigate the impact of these weather conditions on movement behaviour and April body mass, which constitutes the main driver of the population dynamics.

The main findings demonstrated that April body mass decreases with increasing amount of snow on reindeer GPS-track in the preceding October, after accounting for weather conditions during the rest of the winter. The predicted mass difference was about 7 kg between the lowest and highest recorded amount of October snow. In this case, the results would strongly emphasize the importance of autumn conditions on Svalbard reindeer population dynamics. Furthermore, the amount of October snow outperformed the temperature estimates as predictors, which indicates that the previously reported positive effect of warm Octobers on April body mass likely is caused by a shorter winter season. In contrast to the prediction and previously reported negative effects of rain-on-snow, both snow density on the reindeer GPS-track and the annual variation in rain-on-snow during winter had no significant negative effect on April body mass. Moreover, snow depth outperformed both snow density and rain-on-snow as predictors, which is emphasizing the importance of snow depth as a hitherto unexplored mechanism behind variation in April body mass in Svalbard reindeer. In addition, I explored behavioral responses, which demonstrated increased movement of the reindeer in warm Octobers with less snow. In winter, increasing snow density appeared as probable causal explanation for increased movement and surprisingly, reduced time spent on ridges. Overall, this study provides important contributions towards increasing the mechanistic understanding of the environmental factors driving the population dynamics in Svalbard reindeer in a rapidly warming Arctic.