Temperature is one of the most decisive parameters when it comes to determining characteristics and distributions of life worldwide. For plants, as sessile organisms, it is particularly important to be able to deal with the temperatures they are exposed to at a given location. To understand evolutionary processes in plants, it is therefore crucial to gain knowledge about how plants deal with extreme temperatures and respond to changes in temperature. Such knowledge is especially needed given the ongoing rise in temperature under climate change. However, investigations in natural systems on how plants cope with opposing temperature extremes and the effect of long term heating are scarce. In Iceland, a subarctic island in the North Atlantic, a limited number of plant species grow on geothermally heated soils and non-heated soils alike. This constitutes a fascinating natural laboratory for studying local adaptations to opposing temperature extremes and especially constant warming. In my thesis, I investigated responses to geothermal heating in three grass species: Agrostis stolonifera and Agrostis vinealis, which are among the few vascular plants growing on the most heated soils, and Festuca rubra, which grows on moderately heated soils.

First, I reconstructed phylogenetic relationships among Icelandic populations of A. stolonifera and A. vinealis and accessions across both species distribution ranges (Paper I). For A. stolonifera, but not for A. vinealis, I found a distinct geothermal lineage, which is not the closest relative of non-thermal populations. In a subsequent test of thermal tolerance for the geothermal lineage of A. stolonifera, I found no difference in survival following cold treatment, but geothermal plants survived exposure to higher temperatures. However, geothermal plants overall performed worse at colder conditions, which indicates a trade-off between heat tolerance and performance at colder temperatures (Paper II). Comparing survival ability and flowering phenology of the geothermal and non-thermal lineages of A. stolonifera in an overwintering experiment, I found no differences in survival rates but delayed flowering in geothermal A. stolonifera (Paper III). I additionally compared winter survival ability and phenology among several geothermal and non-thermal populations of F. rubra, as well as between northern and southern Swedish populations. I found no difference between geothermal and non-thermal populations of F. rubra but delayed flowering and higher performance in the northern Swedish population (Paper IV).

The different findings for different species and temperature conditions emphasize the complexities of plant evolutionary responses to elevated temperatures. Whether a species adapts to elevated temperatures seems to depend not only on the level of maximum temperature rise, but also on species’ evolutionary histories, and on winter conditions. The found trade-off between heat tolerance and performance at optimal conditions suggests that adapting to extreme heat may limit viability under cooler conditions. These findings highlight the peculiarities of geothermal ecosystems, their value for studying thermal tolerances and provide a framework for future work on thermal adaptations of geothermal grasses.