桂林地区短时暴雨自动监测预警系统研究

Research on auto-monitoring and early warning system for short-time rainstorm in Guilin Region

通过制作混合扫描面,调整Z-I关系和自动站雨量校准等方法对多普勒天气雷达估测降水进行了优化,使雷达估测降水误差明显减小。运用交叉相关法、降水算法对雷达反射率因子进行计算,得到了未来1～2 h的定量预测降水。按照暴雨预警信号标准,将估测降水与预测降水累加,可及时获得暴雨预警信号信息,自动生成暴雨预警信号等级分布图,对达到预警信号标准的站点及时通过手机发布暴雨预警信号提示信息。结合多层次地理信息系统,形成了集自动站资料查询、雷达降水监测、临近预测、暴雨预警功能为一体的桂林雷达定量降水监测预警平台,为有效监测预警暴雨引发的山体滑坡、洪涝等气象灾害提供了客观定量依据。

By means of mixed scanned sector preparation, Z - I relationship adjustment as well as auto gauge rainfall calibration, the Doppler weather radar quantitative precipitation estimation （QPE） are optimized which can obviously decrease the QPE error. The radar quantitative precipitation forecast （QPF） ahead of 1 - 2 h is calculated by performing cross correlation method to retrieve radar reflectivity. According to the standard of rainstorm prewarning sign （RPWS）, QPE and QPF are accumulated to obtain RPWS information in time, and the distribution picture of RPWS creates automatically, the alarm message is released by mobile telephone as soon as RPWS reaches warning standard. Combining multilayer GIS, the fiat of QPE monitoring and pre-warning is constructed that includes auto station data inquery, radar precipitation monitoring, nowcasting and rainstorm early warning, which can provide objectively quantitative basis to monitor and warn meteorological disaster such as landslide, flood,etc.