Work

Urban Fog Prediction Based on Aerosol Microphysics and Visibility Parameterization in a High‐Resolution Model

Year: 2026
Type: article
Abstract: Accurate prediction of winter fog over Delhi remains challenging because operational numerical models inadequately represent aerosol–fog interactions. To address this, a modified Thompson–Eidhammer (TE) aerosol‐aware microphysics scheme, coupled with a diagnostic visibility parameterization accounting for extinction by hydrometeors, aerosols, and gases, was implemented within the Weather Research and Forecasting model. The framework includes a dynamic aerosol activation mechanism linking cloud condensation nuclei responds to thermodynamic cooling rates and water vapor tendencies. Using Winter Fog EXperiment observations, the modified configuration (Experimental Run [EXP]) was evaluated against the default TE scheme (Control Run [CTRL]) for a cloud base lowering (CBL) and a radiation (RAD) fog event over Delhi during January 2025. During CBL event, EXP preserved realistic nocturnal cooling and near‐surface stratification, enabling elevated layers near 400 m to reach saturation and descend in a top‐down manner, whereas CTRL incorrectly simulated a bottom‐up fog structure. EXP also reproduced the observed fog lifecycle more accurately, limiting shallow surface fog duration to ∼9 hr, while CTRL overestimated fog duration and onset by ∼15 and ∼6 hr, respectively. For RAD event, EXP more accurately captured the vertical fog structure through its dynamic aerosol–microphysics coupling. Variable fog droplet number concentrations (N d ~ 400–1,200 cm −3 ), compared to the fixed N d (~300 cm −3 ) in CTRL, were crucial for representing fog layering and vertical growth. Consequently, EXP maintained a much smaller nocturnal temperature bias (∼0.5°C) than CTRL (~3–4°C). The results demonstrate importance of aerosol–fog–radiation feedback and realistic aerosol fields for improving fog prediction over polluted urban regions.
Language: English
Open Access status: bronze
OpenAlex ID: W7232544889
FWCI:
Cites: 66
Cited by:
Related to: 10