Constraining the response of the hydrological cycle to global change with a very large perturbed physics GCM ensemble

Uncertainties in the hydrological cycle due to land surface parameterizations can be divided into uncertainties arising from the spatial distribution of vegetation and from the model parameter values. To explore these uncertainties we prepared a large (10.000+ experiments) HadAM3P-C model ensemble including the MOSES2 land surface scheme using the climateprediction.net distributed computing system. The global circulation model simulates global climate system responses to prescribed sea surface temperatures (SSTs), atmospheric CO2. This new ensemble will be the first combining all aspects of land surface uncertainties - model parameter perturbations and the spatial distribution of land cover - with atmospheric uncertainties conventionally explored. Themes we will explore are detection of the impact on the hydrological cycle of natural and anthropogenic land use in different regions and quantifying the impact of classifying and aggregating land surface observations into land cover classes. We expect to present the first results of this new large land surface focused ensemble.