CMA-CW/dust Product Upgrade Achieves Operational Status

At 00:00 Beijing Time on December 22, 2025, the China Meteorological Administration (CMA) officially put its regional chemical weather numerical prediction system, CMA-CW v1.0, into operation. Its forecast products are showcased internationally through the WMO SDS-WAS Asian Center webpage.

Figure1 CMA-CW v1.0 Dust Forecast Results

CMA-CWv1.0 is based on the online coupling of regional weather operational forecasting model, GRAPES_Meso5.1 (Global/Regional Assimilation and Prediction System - Mesoscale 5.1 version), and chemical model, CUACE (Chinese Unified Atmospheric Chemistry Environment). By utilizing real-time simulated chemical species, the system establishes an aerosol–cloud–radiation interaction mechanism that enables two-way feedback between chemical reactions and weather processes. Through quantifying the interactions among anthropogenic emissions, radiation, and clouds, the system not only improves forecasts of fog–haze and sand and dust storms but also enhances the prediction skill for conventional weather elements such as precipitation and temperature within the numerical weather prediction framework. Moreover, the model incorporates a heterogeneous chemical mechanism involving nine reaction pathways, reducing uncertainties in the forecasting of secondary inorganic aerosols. It also updates the widely used CB05CL gas-phase chemical mechanism, leading to improved predictions for aerosols, ozone, and other pollutants. An atmospheric composition assimilation module has been specifically developed for this system, significantly mitigating uncertainties associated with initial chemical fields. In addition, the implementation of time-step separation for meteorological and chemical integration greatly enhances computational efficiency, ensuring stable operational performance.

Figure 2 CMA-CW v1.0 System Framework

For SDS forecasting, CMA-CW v1.0 provides 0–168-hour forecasts with refined spatial coverage across Asia, providing forecast products such as dust concentration, dust flux, and cumulative dry and wet deposition. The system demonstrates strong performance in predicting dust events affecting China, effectively addressing key deficiencies in CMA's previous operational model (CMA-CUACE), such as excessive rapid movement, poor stability, and underestimation of severe dust storm intensity.