A warming climate is altering species distributions and community compositions. To understand and predict changes in species distributions to climate change, we often use species occurrences together with large-scale regional climate data. This can be problematic for several reasons. Species living near the ground experience small-scale spatial variation in temperatures, i.e., microclimate, that are influenced by topography and vegetation and can therefore deviate a lot from regional temperatures. Further, climate often affects species indirectly via species interactions, and such interactions can also change with climate. And last, species may respond slower than climate changes. Ignoring these aspects can complicate our understanding of species-climate relationships.

In this thesis, I examined how microclimate and changes in microclimate due to forest management impact performances, interactions, and distributions of plant species in boreal forest understory communities. First, I quantified the importance of microclimate for species performances and distributions. Specifically, I compared the effects of spring temperatures measured on local and regional scales on the population dynamics of a southern forest herb (I). I also tested how small-scale spatial microclimate variation contributed to the regional co-existence of northern and southern understory plant species (II). Second, I examined the role of species interactions in driving abundance patterns of two moss species with different temperature niches across their Swedish ranges by transplanting them separately and together across a climate gradient (III). Lastly, I investigated how understory plant communities respond to changes in microclimate caused by forest management (IV), and how past microclimates influence current patterns of species occurrence, abundance, and reproduction (II).

I found that local spring temperatures had a significant effect on the population dynamics of the southern forest herb that could not be detected using regional spring temperatures (I). Spatial variation in microclimate explained the regional co-existence of two northern and two southern species, where the northern species were favoured by cold microclimates and the southern species by warm microclimates (II). In the transplant experiment (III), I found that climate-mediated competition can override the direct effects of climate and limit abundances across ranges. Lastly, I found that microclimate changes caused by forest management activities had a large effect on understory communities (IV), and that current abundances of northern and southern species were partly explained by past microclimate (II).

Overall, I demonstrated that, to understand how species (particularly understory plants) respond to climate, we need to replace the standard use of regional climate data with locally measured climate data or down-scaled gridded climate data that account for variation in topography as well as vegetation. To predict how species will respond to climate change, we also need to include species interactions and how these interactions change with a changing climate. Finally, changes in microclimate following changes in forest structure have large effects on understory species. The last finding is important to consider when studying changes in understory communities in a climate context and could be used to mitigate climate effects on forest biodiversity.