Performance of Land Surface Schemes in the WRF Model for Climate Simulations over the MENA-CORDEX Domain

Abstract

Land–atmosphere interactions need to be optimally represented in climate models for the realistic representation of past and future climate. In this work, six different versions of land surface schemes (LSS) are used to simulate the climate over the Middle East–North Africa (MENA) region for the period 2000–2010 with a horizontal resolution of 0.44°, using the Weather Research and Forecasting (WRF) model. The monthly time series output is evaluated against observations for several surface climate variables using statistical metrics (climatology, 5th and 95th percentiles, standard deviation, linear trend) and Taylor diagrams. The resulting biases are presented for the whole MENA domain as well as 7 sub-domains. A ranking procedure objectively retrieves a performance spectrum among the schemes. The LSS that is closest to observations and is, therefore, considered as the best performing is Noah, followed by its augmented version (NoahMP). For these simulations at the relatively coarse horizontal resolution of 50 km, the more elaborate LSSs are not performing very well. These results are useful for the choice of LSS in climate change modelling of the MENA-CORDEX as a whole, as well as its sub-regions.