Assessment of Future Soil Ecosystem Services of a Drained Soil Under Different Climate Change Scenarios

ABSTRACT In order to prepare for the necessary adaptation to climate change, it would be useful to have some idea of how the services provided by soils to human societies are going to evolve in the future. Unfortunately, since actual measurements of soil services remain elusive in general, modelling efforts needed to predict their future evolution are fraught with very large uncertainties. In one particular situation, however, on the Saclay plateau south of Paris (France), it has been possible to measure five soil services: three provisioning services (supply of water to nearby stream, provision of, respectively, water and nitrogen to wheat crop), two regulating services (flood mitigation, and filtration of pollutant), as well as the amount of wheat produced on site. In the present article, we take advantage of available measurements to parametrize a soil–plant–atmosphere model (STICS), which we combine with three future climate change scenarios to investigate the extent to which soil services will change until the end of the century. Simulation results suggest that the soils at the Saclay site are unexpectedly resilient to climate change, with statistically significant changes occurring in four of the six services only for the most extreme scenario, under which no mitigation effort at all is taking place. Even then, significant changes occur only between 2071 and 2100. The increase in temperature and relative CO 2 pressure in the air leads to a shortening of the plant growth cycle by approximately 2 weeks when comparing the most pessimistic scenario to the most optimistic one. This shortening of the plant cycle means that the end of the growth period no longer overlaps with the summer season, which is typically associated with drought events and heat waves. As a result, the impact of climate change on plant growth is not easily detectable. At this stage, the simulations rest on the assumption that current rainfall patterns will remain the same in the future as they are now. If, as is predicted, rainfall events become more intense and less frequent in the future, the daily precipitation averages currently available will no longer be sufficient, and more accurate, hourly data will be needed to predict the evolution of soil services accurately.