Land use effects on greenhouse gas emissions from boreal inland waters

Anthropogenic activities perturb the global carbon and nitrogen cycle with large implications for the earth’s climate. Land use activities deliver excess carbon and nitrogen to aquatic ecosystems. In the boreal biome, this is mainly due to forestry and atmospheric deposition. Yet, impacts of these anthropogenically mediated inputs of carbon and nitrogen on the processing and emissions of greenhouse gases from recipient streams and lakes are largely unknown. Understanding the ecosystem-scale response of aquatic greenhouse gas cycling to land use activities is critical to better predict anthropogenic effects on the global climate system and design more efficient climate change mitigation measures. This thesis assesses the effects of forest clearcutting and nitrate enrichment on greenhouse gas emissions from boreal inland waters. It also advances methods to quantify sources and sinks of these emissions. Short-term clearcut and nitrate enrichment effects were assessed using two whole-ecosystem experiments, carried out over four years in nine headwater catchments in boreal Sweden. In these experiments, I measured or modeled air-water fluxes of carbon dioxide (CO2), methane (CH4) and nitrous oxide (N2O), combining concentration, ebullition and gas-transfer velocity measurements in groundwater, streams and lakes. By using Swedish national monitoring data, I also assessed broad-scale effects of forest clearcutting by relating CO2 concentrations in 439 forest lakes to the areal proportion of catchment forest clearcuts. To improve quantifications of CO2 sources and sinks in lakes, I analyzed time series of oxygen concentrations and water temperature in five lakes on conditions under which whole-lake metabolism estimates can be inferred from oxygen dynamics given the perturbing influence of atmospheric exchange, mixing and internal waves. The experiments revealed that aquatic greenhouse gas emissions did not respond to nitrate addition or forest clearcutting. Importantly, riparian zones likely buffered clearcut-induced increases in groundwater CO2 and CH4 concentrations. Experimental results were confirmed by monitoring data showing no relationship between CO2 patterns across Swedish lakes and clearcut gradients. Yet, conclusions on internal vs. external CO2 controls largely depended on whether spatially or temporally resolved data was used. Partitioning CO2 sources and sinks in lakes using time series of oxygen was greatly challenged by physical transport and mixing processes. Conclusively, ongoing land use activities in the boreal zone are unlikely to have major effect on headwater greenhouse gas emissions. Yet, system- and scale specific effects cannot be excluded. To reveal these effects, there is a large need of improved methods and design of monitoring programs that account for the large spatial and temporal variability in greenhouse gas dynamics and its controls by abiotic and biotic factors