Uncertainty Analysis of a Coupled Hydrological-plant Growth Model for Grassland under Elevated CO2

The continuous increase in atmospheric carbon dioxide (CO2) contributes to changes in plant evapotranspiration and terrestrial water budgets in two ways. Firstly, elevated CO2 can result in a water saving effect, since increasing CO2 reduces stomatal opening and therefore decreases transpiration. Secondly, CO2 fertilization increases biomass accumulation and leaf area at plant canopy level, likely increasing transpiration1. Vegetation and hydrological models can be used to investigate the CO2 response and the bidirectional effects outlined above, including their relative contribution to the changes in the water cycle. However, the intrinsic plant-soil interaction and the uncertainty related to model parameterization have rarely been considered.Hence, we coupled a detailed plant growth and soil hydrological model by using the generic model frameworks Plant growth Modelling Framework (PMF)2 and Catchment Modelling Framework (CMF)3. Up to date response mechanisms have been implemented in PMF to simulate the various ways of how plant physiology is influenced by elevated CO2. Both models interact by using the Python computer language. Applying the coupled PMF-CMF model we investigate the effects of elevated CO2 in a number of plant physiological and environmental variables such as biomass, leaf area index and soil moisture using field data of a long-term Free Air Carbon Enrichment (FACE) experiment in Giessen, Germany. In this experiment, various grassland varieties (herbs, legumes, grass) grow under elevated (+20%) and ambient CO2 since 1997.A Monte Carlo based uncertainty analysis (GLUE) is conducted to investigate the coupled PMF-CMF parameter space. The focus will be on the identification of parameters for plant and soil, which are the drivers for the CO2 response of the terrestrial water balance. We will present first results of the simulation of biomass accumulation and transpiration under ambient and elevated CO2 concentrations.