Assessing fitness consequences of migratory tactics requires long-term individually based monitoring

Gaillard, Jean-Michel

Middleton et al. (2013; hereafter referred to as Middleton et al.) studied an elk population where migrants winter in wilderness areas and summer 40–60 km away in Yellowstone National Park (USA), all areas not subject to human disturbance. Despite the lack of habitat degradation or physical barriers caused by human activities, the proportion of migrants decreased from 81% in the late 1970s to the early 1980s to 47.5% in recent years. In parallel, the productivity of migratory females decreased over the last 20 years. Compared to resident females, young and lactating females engaging into migration had a lower reproductive output than their nonmigratory counterparts in recent years. The authors concluded that deterioration of summer range habitat induced by climate change and high predation by wolves (Canis lupus) and grizzly bears (Ursus arctos horribilis) in summer range account for the patterns observed. While this case study improves our understanding of ungulate migratory tactics by providing clear evidence for lower pregnancy rates of migrants compared to residents in the absence of human disturbance, it fails to either quantify fitness differences between migrant and residents or identify the contribution of factors involved in such differences. Three major limitations, common to most studies of fitness consequences of migratory tactics in ungulates, likely account for these shortcomings: high heterogeneity in data quality and temporal scale, strong assumptions, and the lack of long-term individually based data.