Radar proximity prediction method based on heavy rainfall identification and numerical atmospheric mode driving

Abstract

The invention relates to a radar proximity prediction method based on heavy rainfall identification and numerical atmospheric mode driving. The method comprises the following steps: 1, identifying convection nuclear grid points based on phase state partition; 2, identifying the convective nuclear grid points again by adopting the gradients of the radar reflectivity in the horizontal direction andthe vertical direction or the gradients of the radar reflectivity in the horizontal direction and the radial direction; 3, searching for convection nuclear grid points based on a three-dimensional region growing method until all the grid points are searched for; 4, determining all grid point sets as convection regions; 5, continuously monitoring convection nuclear grid points, superposing wind field information, and determining a rainfall falling area; step 6, implementing rainfall proximity prediction. According to the method, the recognition speed and accuracy of the severe convection area are improved, the approach prediction precision of heavy rainfall is improved, and reliable technical support is provided for sudden rainstorm and flood disaster prevention.

Description

technical field [0001] The invention relates to a radar nowcasting method based on heavy precipitation identification and numerical atmospheric model drive, which belongs to the field of radar nowcasting and is mainly used for meteorological and water conservancy departments to carry out work such as rainstorm and flood forecasting and early warning. Background technique [0002] Heavy precipitation is often sudden, short-duration, and large-scale, mainly formed by strong convective weather, and is a serious meteorological disaster. In general, the accumulated rainfall in 1 hour exceeds 20 mm, which is considered heavy precipitation. Heavy precipitation can cause disasters such as mountain torrents, mudslides, and landslides, and it can also cause urban waterlogging, seriously threatening the safety of people's lives and property. Enhancing the identification ability of heavy rainfall and improving the accuracy of nowcasting of heavy rainfall has a very important role and s...

AI Technical Summary

Problems solved by technology

However, practical applications have shown that due to the difficulty in identifying heavy precipitation, the ability of weather radar to monitor heavy precipitation is weaker than that of general precipitation. At present, radar nowcasting is mainly based on empirical or statistical extrapolation methods, such as cross-correlation algorithms, which lack The stability of the physical basis and calculation relationship is not high, and the weather conditions that form heavy precipitation are varied and have strong uncertainties. If the monitoring and identification of heavy precipitation is not in place, the results of nowcasting will not be ideal

Images