Evaluation and Comparison of Noah and Pleim–Xiu Land Surface Models in MM5 Using GÖTE2001 Data: Spatial and Temporal Variations in Near-Surface Air Temperature

Abstract

In this study, the performance of two advanced land surface models (LSMs; Noah LSM and Pleim?Xiu LSM) coupled with the fifth-generation Pennsylvania State University?National Center for Atmospheric Research Mesoscale Model (MM5), version 3.7.2, in simulating the near-surface air temperature in the greater Göteborg area in Sweden is evaluated and compared using the GÖTE2001 field campaign data. Further, the effects of different planetary boundary layer schemes [Eta and Medium-Range Forecast (MRF) PBLs] for Noah LSM and soil moisture initialization approaches for Pleim?Xiu LSM are investigated. The investigation focuses on the evaluation and comparison of diurnal cycle intensity and maximum and minimum temperatures, as well as the urban heat island during the daytime and nighttime under the clear-sky and cloudy/rainy weather conditions for different experimental schemes. The results indicate that 1) there is an evident difference between Noah LSM and Pleim?Xiu LSM in simulating the near-surface air temperature, especially in the modeled urban heat island; 2) there is no evident difference in the model performance between the Eta PBL and MRF PBL coupled with the Noah LSM; and 3) soil moisture initialization is of crucial importance for model performance in the Pleim?Xiu LSM. In addition, owing to the recent release of MM5, version 3.7.3, some experiments done with version 3.7.2 were repeated to reveal the effects of the modifications in the Noah LSM and Pleim?Xiu LSM. The modification to longwave radiation parameterizations in Noah LSM significantly improves model performance while the adjustment of emissivity, one of the vegetation properties, affects Pleim?Xiu LSM performance to a larger extent. The study suggests that improvements both in Noah LSM physics and in Pleim?Xiu LSM initialization of soil moisture and parameterization of vegetation properties are important.