The boreal forest is of particular interest to climate change research due to its large circumpolar distribution and accumulated soil carbon pool. Carbon uptake in this ecosystem is nitrogen (N)-limited, therefore factors affecting carbon or nitrogen dynamics in the boreal forest can have consequences for global climate. We used a 2-year field experiment to investigate the response of biotic nitrogen fixation by cyanobacteria associated with boreal forest bryophytes, in a factorial experiment combining simulated climate change with habitat fragmentation treatments. We simulated climate change conditions using open-top greenhouse chambers in the field, which increased mean and maximum temperatures, and created a precipitation gradient from ambient levels in the center to extreme drought conditions at the periphery of the chamber. The dry patches near the chamber walls exhibited almost no N-fixation, despite having similar densities of cyanobacteria (predominantly Stigonema sp.) as other patches. Rates of N-fixation were best explained by a model containing moisture, fragmentation, cyanobacteria density and time; warming was not a significant variable affecting N-fixation. There was no significant interaction between warming and fragmentation. These results suggest that cyanobacteria responded physiologically to drought by reducing N-fixation activity long before any changes in density. Ecosystem processes, such as N-fixation, can respond in the short term to environmental change much more rapidly than changes in the underlying community structure. Such rapid physiological responses may occur faster than demographic insurance effects of biodiversity.
Keywords: Boreal forest; Climate-change; Cyanobacteria; Habitat fragmentation; Nitrogen fixation; Pleurozium schreberi.