Climate change affects insects in several ways, including phenological shifts that may cause asynchrony between herbivore insects and their host plants. Insect larvae typically have limited movement capacity and are consequently dependent on the microhabitat conditions of their immediate surroundings. Based on intensive field monitoring over two springs and on larger-scale metapopulation-level survey over the same years, we used Bayesian spatial regression modelling to study the effects of weather and microclimatic field conditions on the development and survival of post-diapause larvae of the Glanville fritillary butterfly (Melitaea cinxia) on its northern range edge. Moreover, we assessed whether the observed variation in growth and survival in a spring characterized by exceptionally warm weather early in the season translated into population dynamic effects on the metapopulation scale. While similar weather conditions enhanced larval survival and growth rate in the spring, microclimatic conditions affected survival and growth contrastingly due to the phenological asynchrony between larvae and their host plants in microclimates that supported fastest growth. In the warmest microclimates, larvae reached temperatures over 20°C above ambient leading to increased feeding, which was not supported by the more slowly growing host plants. At the metapopulation level, population growth rate was highest in local populations with heterogeneous microhabitats. We demonstrate how exceptionally warm weather early in the spring caused a phenological asynchrony between butterfly larvae and their host plants. Choice of warmest microhabitats for oviposition is adaptive under predominant conditions, but it may become maladaptive if early spring temperatures rise. Such conditions may lead to larvae breaking diapause earlier without equally advancing host plant growth. Microclimatic variability within and among populations is likely to have a crucial buffering effect against climate change in many insects.

These datasets are a combination of fine-scale field monitoring data and metapopulation-level field survey data. Monitoring data consist of butterfly larval survival, growth, body surface temperature, and activity under variable weather and microclimatic conditions, as well as the growth of larval host plants under similar environmental conditions. For larval survival and growth, as well as host plant growth, the repeated measurements over the study season were averaged to gain a single value of each variable for each larval group and host plant plot. Larval temperature and activity datasets are time series with several measurements covering one day.

Survey data consist of two separate datasets on local population growth rates from one autumn to the following under variable weather and microclimatic conditions. Microclimatic conditions were recorded separately for each larval group in the field and averaged for local populations.

Rows with missing values were removed from all datasets.

Explanation of the variables can be found in the attached ReadMe file describing the dataset.