Physical Processes Affecting Radiation Fog Based on WRF Simulations and Validation

Abstract

The goal of this work is to assess the sensitivity of the Weather Research and Forecasting (WRF) model to various microphysical and land surface model parameterizations, as well as their effectiveness in capturing a radiation fog event that occurred on 15–16 February 2000 over Delhi in the Indo-Gangetic Plain (IGP). Fog forecasting using state-of-the-art mesoscale models is challenging due to limitations in understanding of atmosphere-land surface feedbacks and fog microphysics. Fog can have adverse effects because of low visibility on transport and aviation but affects positively agriculture and forestry by absorbing fog water. In India, the IGP is particularly susceptible to fog during the winter months of December, January, and February (DJF). To reach the goal in this work, preliminary investigation is carried out with five model experiments centered at Delhi for testing model sensitivity to nesting and gravitational settling of fog. The non-nested domain fares better at fog forecasting as compared to the nested domain. Accounting for the gravitational settling of fog droplets in the model further enhances model performance. Thereafter, to assess the model sensitivity to parameterization schemes, 40 model suite combinations with five microphysical (MP) schemes, two planetary boundary layer (PBL) schemes, and four land surface model (LSM) schemes are compared. The validations for fog formation, development, and dissipation are performed using observations collected over Safdarjung airport in Delhi. We conclude that the Thompson MP scheme, along with the Mellor-Yamada Nakanishi-Niino (MYNN 2.5) PBL and Rapid Update Cycle (RUC) LSM scheme, is able to capture the radiation fog event better than other schemes, and it is critical for evaluating the life cycle of radiation fog.