Abstract
Strong phenological shifts in response to changes in climatic conditions have been reported for many species, including amphibians, which are expected to breed earlier. Phenological shifts in breeding are observed in a wide number of amphibian populations, but less is known about populations living at high elevations, which are predicted to be more sensitive to climate change than lowland populations. The goal of this study is to assess the main factors determining the timing of breeding in an alpine population of the common toad (Bufo bufo) and to describe the observed shifts in its breeding phenology. We modelled the effect of environmental variables on the start and peak dates of the breeding season using 39 years of individual-based data. In addition, we investigated the effect of the lunar cycle, as well as the individual variation in breeding phenology. Finally, to assess the individual heterogeneity in the timing of breeding, we calculated the repeatability of the timing of arrival at the breeding site. Breeding advanced to earlier dates in the first years of the study but the trend continued only until the mid 1990s, and stabilised afterwards. Overall, toads are now breeding on average around 30 days earlier than at the start of the study period. High temperatures and low snow cover in winter and spring, as well as reduced spring precipitation were all associated with earlier breeding. Additionally, we found evidence of males arriving on average before females at the breeding site but no clear and strong effect of the lunar cycle. We only found weak evidence of among-individual variation in shifts in the breeding phenology, as well as a low repeatability of arrival timing. Our findings show that the observed changes in breeding phenology are strongly associated with the environmental conditions. These results contribute to filling a knowledge gap on the effects of climate change on alpine amphibian populations. Moreover, we show that changes in phenology, especially in the mountains, can be hard to predict as local microclimatic conditions do not necessarily reflect the observed global climatic trends.