南宁城市内涝分布特征及其监测预警系统应用分析

Distribution Characteristics of Urban Waterlogging in Nanning and Application of its Monitoring and Early Warning System

利用南宁市2009—2017年城市内涝资料及其自动气象站逐小时降雨资料,对城内104例内涝过程进行了时空特征分析,并对39个内涝点进行内涝等级划分;基于多普勒雷达产品、数值预报产品及城市密集自动雨量观测站数据,采用内涝积水模型及内涝风险等级划分资料,研发出精细化到街区的市内涝监测预警和风险评估系统,实现了城市内涝的实时监测、预警及风险预估。该系统运用于2015年5月4日短时强降雨造成的城市内涝过程,验证结果表明:(1)模拟结果与历史描述基本相符,最大积水深度和位置基本一致,表明该系统具备一定的暴雨内涝动态监测预报能力,对提高城市暴雨内涝灾害的监测预警和风险评估起到了一定作用;(2)由于标本数量过少以及缺乏城市最新实际管网数据等信息,对于个别点预警值与实测值存在偏差,在系统运行过程中,部分参数和致灾阀值需进一步调整,比如可以不断增加标本以及更新城市最新实际管网数据等信息;(3)文中使用的是自动气象观测点实况雨量模拟积水状况,积水过程存在一定滞后性,下一步考虑结合雷达估测降水QPE资料,预警时间将大大提前。

Basing on the urban waterlogging data of Nanning City from 2009 to 2017 and the hourly rainfall data of the automatic weather stations,we analyzed the spatial and temporal characteristics of 104 waterlogging processes,and classified the grade of 39 waterlogging spots.Then on the basis of Doppler radar products,numerical forecast products,and the data from the urban dense automatic rainfall observation stations,we developed a block-based waterlogging monitoring early warning and risk assessment system by using the waterlogging model and waterlogging risk grading data,thus realizing the real-time monitoring,warning and risk estimates of urban waterlogging.The system is applied to the urban waterlogging process caused by short-term heavy rainfall on May.4,2015.The verification results show that:(1)The simulation results are basically consistent with the historical descriptions in the maximum depth and the location of water accumulation,indicating that the system has certain ability of dynamic monitoring and forecasting of heavy rainfall-triggered waterlogging and has played a certain role in improving the monitoring and early warning and risk assessment of urban heavy rainfall-triggered waterlogging disasters.(2)Due to the small number of specimens and the lack of the latest urban actual pipe network data etc.,there is a deviation between early warning and the measured values in some spots.So during the operation of the system,some parameters and hazard threshold value need to be further adjusted,for example,the samples can be continuously increased and the latest actual pipe network data of the city can be updated.(3)In this paper,the automatic meteorological observation points are used to simulate the water accumulation situation of actual rainfall,which leads to the lagging of water accumulation process,so the QPE data of radar-estimated precipitation should be taken into consideration to advance the early warning time.