A Nowcasting Technique for Heavy Rainfall Areas Using Deep Network for Salient Object Detection

With the same precipitation duration, the higher the precipitation intensity, the greater the likelihood of inducing urban waterlogging, flash flooding, mudslides, landslides, and other secondary disasters. In the context of global warming, sub-hourly extreme precipitation events are increasing much faster than those at longer time scales, necessitating advanced monitoring and nowcasting technologies for sub-hourly heavy rainfall. This study selects 30 heavy precipitation events in Chongqing from 2013 to 2021 to train the deep network U2-Net for significant object detection, with weather radar reflectivity mosaics serving as inputs. The network designates heavy rainfall areas as salient objects and autonomously learns the nonlinear relationship between the spatial distribution of reflectivity at a given time and the subsequent 30-minute heavy rainfall areas, and provides forecasting results of the heavy rainfall region. The sample labels are divided into three categories based on thresholds of 10 mm, 20 mm, and 30 mm, obtained from radar-raingauge quantitative precipitation estimates. The model inputs are radar reflectivity mosaics at altitudes of 3 km, 4.5 km, and 7 km. After training and validation, three forecasting models corresponding to the three heavy rainfall thresholds were developed. Testing on an independent dataset revealed that, with a neighborhood radius of 5 km, the models achieved hit rates of 0.66, 0.73, and 0.72, false alarm rates of 0.06, 0.32, and 0.57, and critical success indices of 0.64, 0.54, and 0.37 for the 10 mm, 20 mm, and 30 mm thresholds, respectively. Higher probabilities in the forecast maps indicated more reliable forecast of heavy rainfall areas. In conclusion, the heavy rainfall forecasting models can effectively predict 30-minute heavy rainfall areas by extracting multi-scale features of three-dimensional radar reflectivity. These models help to supplement rainfall monitoring and forecasting in areas with sparse rain gauge networks and provide a reference for nowcasting technologies that require multi-source data feature extraction.