P42 Evaluation of 2m air temperature heating and cooling onset over a desert region in the Arabian Peninsula

Temimi, Marouane, Weston Michael, Naira Chaouch, Jun Zhao, Vineeth Krishnan Valappil, Masdar Institute of Science and technology, United Arab Emirates, and Oliver Branch, Institute of Physics and Meteorology, University of Hohenheim, Germany

Results from operational modelling over the desert of the Arabian Peninsula indicate that the WRF model heats up too early during the morning and cools down too soon during the afternoon. Furthermore, the model struggles to model maximum and minimum temperatures within a suitable error range. Daily maximum temperature has a mean gross error of 1.4 degrees C (positive bias) and daily minimum 1.9 degrees C (negative bias). Through a process of elimination, this appears to be caused in part by the calculation of heat and momentum roughness lengths as used in the land surface model and the surface layer parameterisation. We test available options for adjusting the Czil parameter, a coefficient used to calculate the heat and momentum roughness lengths. We adjust Czil values to constants of 0.1, 0.8 and then dynamically based on vegetation height. Results showed an improvement in modelling maximum temperature when using the variable Czil, but only in combination with other parameters like PBL and landuse/landcover. Improvements in the timing of heating/cooling and the minimum temperature were minimal, where minimum temperature is more dependent on the landuse/lancover. Sensitivity results from landuse/landcover change, PBL schemes and Czil are presented and summarised in the analysis.