An update to WRF surface layer parameterization over an Indian region

Abstract

Surface layer parameterization schemes in numerical weather prediction models are based on the Monin-Obukhov similarity theory (MOST), which utilizes empirical functions to incorporate the effects of near-surface atmospheric stability. In the present study, an effort has been made to implement and evaluate the performance of recently developed similarity functions under stable stratification in the surface layer parameterization of Weather Research and Forecasting Model version 4.2.2 (WRFv4.2.2). For this purpose, the commonly used revised version of MM5 surface layer module in WRF model is updated using the similarity functions suggested by Srivastava et al. (2020). The model is configured with three nested domains around the flux tower installed at Ranchi (23.412° N, 85.440°E), India. The simulations are carried out for a complete year, and the model simulated near-surface atmospheric variables are compared with the observations. The study reveals that updated similarity functions lead to a noticeable improvement in WRF model performance. In particular, the modified scheme reduced the mean absolute error and root mean square error for 10-m wind speed (2-m temperature) by about 22 % (10 %) and 23 % (8 %), respectively, with improved correlation coefficients during January. The analysis suggests that the new similarity functions could potentially be used in weather forecast model over the Indian region.