BDS-retrieved minute-level atmospheric stability indices for convective event monitoring

Monitoring atmospheric stability is important for forecasting convection. To compensate for the insufficient temporal resolution of traditional techniques, precise precipitable water vapor (PWV) derived from BeiDou Navigation Satellite System (BDS) observations was employed to establish a model for retrieving the minute-level K-Index and Showalter Index for convective events. BDS-derived PWV (BDS-PWV) revealed the bias and root-mean-square error (RMSE) of 0.5 and 1.9 mm, respectively, with respect to the Fifth Generation of the European Centre for Medium-Range Weather Forecasts (ECMWF) Reanalysis (ERA5). Compared with radiosonde data, the corresponding values were − 0.5 and 2.5 mm, respectively. This indicates the feasibility of using BDS-PWV for further modeling. Subsequently, two atmospheric stability indices were retrieved from the BDS-PWV using the traditional least-squares (LSQ), backpropagation neural network (BPNN), and random forest (RF) methods. The evaluation process adopted both the ERA5 reanalysis and satellite-based products from the Moderate Resolution Imaging Spectroradiometer (MODIS). The results indicated that the RF method performed best among the three methods. Taking ERA5 reanalysis as the reference, the average correlation coefficients, RMSEs, and mean absolute percentage errors (MAPEs) are 0.98/0.97, 2.8/1.3 °C, and 3.8 %/5.0 % for the K-Index and Showalter Index, respectively. A similar performance was obtained when compared with MODIS products with average correlation coefficients, RMSEs, and MAPEs of 0.94/0.90, 4.8/1.9 °C, and 6.1 %/7.4 %, respectively. These results demonstrate the feasibility of using the BDS-retrieved minute-level K-Index and Showalter Index for meteorological studies. Finally, the two modeled atmospheric stability indices were applied to three representative convective events that occurred in 2024 over Beijing and its surrounding areas. Temporal analysis of the two indices clearly revealed the onset and sustained processes of the convective events. Moreover, the forecasting results based on the two atmospheric stability indices demonstrated their potential for indicating convective rainfall. The proposed BDS-retrieved minute-level atmospheric stability indices can be useful indicators for monitoring and nowcasting atmospheric convection.