Evaluation of nonlocal and local planetary boundary layer schemes in the WRF model

Abstract

A realistic reproduction of planetary boundary layer (PBL) structure and its evolution is critical to numerical simulation of regional meteorology and air quality. Conversely, insufficient realism in the simulated physical properties often leads to degraded meteorological and air quality prognostic skills. This study employed the Weather Research and Forecasting model (WRF) to evaluate model performance and to quantify meteorological prediction differences produced by four widely used PBL schemes. Evaluated were two nonlocal PBL schemes, YSU and ACM2, and two local PBL schemes, MYJ and Boulac. The model grid comprised four nested domains at horizontal resolutions of 27 km, 9 km, 3 km and 1 km respectively. Simulated surface variables 2 m temperature and 10 m wind at 1 km resolution were compared to measurements collected in Hong Kong. A detailed analysis of land‐atmosphere energy balance explicates heat flux and temperature variability among the PBL schemes. Differences in vertical profiles of horizontal velocity, potential temperature, bulk Richardson number and water vapor mixing ratio were examined. Diagnosed PBL heights, estimated by scheme specific formulations, exhibited the large intrascheme variance. To eliminate formulation dependence in PBL height estimation, lidar measurements and a unified diagnosis were jointly used to reanalyze PBL heights. The diagnosis showed that local PBL schemes produced shallower PBL heights than those of nonlocal PBL schemes. It is reasonable to infer that WRF, coupled with the ACM2 PBL physics option can be a viable producer of meteorological forcing to regional air quality modeling in the Pearl River Delta (PRD) Region.