Impact of droughts on the C-cycle in European vegetation:A probabilistic risk analysis using six vegetation models

Identifiant Hal : hal-01136286We analyse how climate change may alter risks posed by droughts to carbon fluxes in European ecosystems.The approach follows a recently proposed framework for risk analysis based on probability theory. In this approach, risk is quantified as the product of hazard probability and ecosystem vulnerability. The probability of a drought hazard is calculated here from the Standardized Precipitation– Evapotranspiration Index (SPEI). Vulnerability is calculated from the response to drought simulated by processbasedvegetation models. We use six different models: three for generic vegetation (JSBACH, LPJmL, ORCHIDEE) and three for specific ecosystems (Scots pine forests: BASFOR; winter wheat fields: EPIC; grasslands: PASIM). The periods 1971–2000 and 2071–2100 are compared. Climate data are based on gridded observations and on output from the regional climate model REMO using the SRES A1B scenario. The risk analysis is carried out for 18 000 grid cells of 0.25 0.25 across Europe. For each grid cell, drought vulnerability and risk are quantified for five seasonal variables: net primary and ecosystem productivity (NPP, NEP), heterotrophic respiration(Rh), soil water content and evapotranspiration. In this analysis, climate change leads to increased droughtrisks for net primary productivity in the Mediterranean area: five of the models estimate that risk will exceed 15 %. The risks increase mainly because of greater drought probability; ecosystem vulnerability will increase to a lesser extent. Because NPP will be affected more than Rh, future carbon sequestration (NEP) will also be at risk predominantly in southern Europe, with risks exceeding 0.25 gCm2 d1 according to most models, amounting to reductions in carbon sequestration of 20 to 80 %