Abstract Temperature influences the rate of most biological processes. Nonlinearities in the thermal reaction norms of such processes complicate intuitive predictions of how ectothermic organisms respond to naturally fluctuating temperatures, and by extension, to climate warming. Additionally, organisms developing close to the ground experience a highly variable microclimate landscape that often is poorly captured by coarse standard climate data. Using a butterfly population in central Sweden as a model, we quantified the consequences of small‐scale temperature variation on phenology, emergence synchrony and number of annual reproductive cycles (voltinism). By combining empirical microclimate and thermal performance data, we project development of individual green‐veined white butterflies (Pieris napi) across 110 sites in an exceptionally high‐resolved natural microclimate landscape. We demonstrate that differences among microclimates just meters apart can have large impacts on the rate of development and emergence synchrony of neighbouring butterflies. However, when considering the full development from egg to adult, these temporal differences were reduced in some scenarios, due to negative correlations in development times among life stages. The negative correlations were caused by temperatures at some sites beginning to exceed the optimum for development as the season progressed. Indeed, which sites were optimal for fast development could change across the lifetimes of individual butterflies, that is, ‘fast’ sites could become ‘slow’ sites. Thus, from a thermal point of view, there seem to be no consistently optimal microsites. Importantly, the fast sites were not always the warmest sites. We showed that such unintuitive effects could play an important role in the regulation of phenological synchrony and voltinism in insects, as most sites consistently favoured two generations. The results were generally robust across years and three different egg‐laying dates. Using high‐resolved empirical climate data on organism‐relevant temporal and spatial scales and considering nonlinear responses to temperature, we demonstrated the large and unintuitive population‐level consequences of locally and temporarily high temperatures. We suggest to—whenever possible—incorporate species‐ and life stage‐specific nonlinear responses to temperature when studying the effects of natural microclimate variation and climate change on organisms. Read the free Plain Language Summary for this article on the Journal blog.
Read full abstract