Grasslands in Europe are characterized by a wide range of management intensities and vegetation compositions, tuned to highly variable environmental conditions across Europe. In the future, we expect both climate and land use to change, with uncertain consequences for the provisioning of ecosystem services to society. Thus, understanding how grasslands respond to multiple global change drivers, in particular to drought and management, is one of the key interests in ecosystem science. We therefore used rainout shelters to understand the response of Swiss grasslands to drought (as expected in 2070) as well as economic assessments to evaluate respective consequences for farmers. Moreover, we measured net ecosystem greenhouse gas (CO2, CH4 and N2O) and water vapor (H2Ov) fluxes of different grasslands using eddy-covariance techniques (gas analyzers: CO2, H2Ov; laser systems: CH4, N2O) to quantify the effects of grassland management practices on biospheric-atmospheric fluxes.
Results/Conclusions
Simulating severe summer droughts in grasslands with different management intensities during eight years generally resulted in decreased yields and increased weed pressure (although site-specific), but also revealed a very high resilience of these grasslands after drought. Once the vegetation received precipitation, yields returned to ambient conditions, without time-lags. Increased root growth in the top soil even during the drought and preferential water uptake from upper soil layers (determined using stable water isotopes) facilitated this fast regeneration growth. However, the economic evaluation of these yield effects was surprising: despite highest yield losses at the most extensive site, it was economically the least risky one for the famer, due to agri-environmental payments.
Restoring a permanent grassland after 10 years of intensive management (incl. ploughing, harrowing, sowing, fertilizing) revealed large N2O losses, small (around zero) CH4 fluxes and only slowly increasing CO2 uptake rates until full sward establishment. Thus, N2O (48%) and CO2 losses (44%) clearly controlled the annual GHG budget, turning this grassland in this particular year into a GHG source. Expanding this work to 15 European sites showed very similar patterns, all grasslands were N2O sources, small CO2 sinks (but see above), and showed close to zero CH4 fluxes (except one intensively grazed site). Fertilization events triggered increased N2O losses, lasting for 3 to 6 days after each event, with highest losses at soil temperatures between 12 and 18°C and water-filled pore space between 60 and 80%. Thus, comprehensive, interdisciplinary assessments of global change impacts on ecosystem services are urgently needed to obtain the full picture for real life conditions.