Climate Model Evaluation Using a Satellite Simulator for TRMM PR

Abstract:

The interpretation of climate model data requires a solid validation of the simulated processes such as the global atmospheric water and energy cycle utilizing in situ measurements, remote sensing data and reanalysis as a reference. Here we develop a satellite simulator for the Tropical Rainfall Measuring Mission precipitation radar (TRMM PR) utilizing the CFMIP Observation Simulator Package (COSP). This enables the evaluation of the climate model data in the instrument’s parameter space thereby reducing uncertainties on the reference side. Assumptions for microphysical calculations are chosen to be consistent with the climate model microphysical schemes. The simulator was developed and applied within the MiKlip project funded by BMBF (German Federal Ministry of Education and Research) to evaluate decadal climate predictions performed with the MPI-ESM (ECHAM6) developed at the Max Planck Institute for Meteorology. In this study we focus on a comparison between two climate models, namely the MPI-ESM (ECHAM6) and the HadGEM2, in order to evaluate the sensitivity of the satellite simulator results to microphysical properties and sub-grid scale variability. Utilizing joint height-radar reflectivity distributions the simulated radar reflectivites are found to be similar to observations. However, results suggest a slight underestimation of the observed occurrence rates for both models. While MPI-ESM (ECHAM6) shows a clear contribution from both stratiform and convective rainfall to the simulated radar reflectivities, results for HadGEM2 show less contribution from stratiform rainfall. Differences between the two models are traced back to hydrometeor mixing ratio and effective radius distributions. As expected, the results are also shown to be sensitive to the method used to generate sub-grid scale precipitation variability.