Global Evaluation of the Noah-MP Land Surface Model and Suggestions for Selecting Parameterization Schemes

This study examines the overall performance of the Noah with multiparameterization (Noah-MP) land surface model in simulating key land-atmosphere variables at a global scale and explores the feasibility of running Noah-MP with regionally different combinations of parameterization schemes. We conducted Noah-MP ensemble simulations and evaluated the annual means and seasonal cycles of the simulated latent heat flux, net radiation (RN), runoff, soil moisture, snow water equivalent, land surface temperature (LST), leaf area index (LAI), and gross primary productivity (GPP) against a wide variety of global products. The results show that the global patterns of the modeled annual means of these variables generally agree with those of the reference data sets. By evaluating the best simulations in the ensemble, we show that Noah-MP performs very well in simulating global LST and RN but produces biases in annual mean LAI and GPP by more than 40% in most herbaceous regions. Overall, the main disagreements between Noah-MP and the reference data sets occurred in the tropical, polar, high-altitude, and hyperarid regions. This study also highlights the potential of land-cover-specific combinations of parameterization schemes to produce optimal modeling results over different land-cover types. In addition, we strongly suggest the use of multi-objective optimization of the key parameterizations and parameters to further improve the Noah-MP's overall performance.