RETRIEVAL OF WATER VAPOR FROM RADAR REFLECTIVITY BASED ON BAIYESIAN SCHEME AND ITS ASSIMILATION TEST
-
摘要: 雷达资料是目前为数不多有能力为高分辨率预报模式提供高分辨率信息资料的资料之一。为充分利用该资料所包含的中小尺度信息，文中基于雷达反射率，利用贝叶斯方法反演出大气相对湿度；将质控后的资料引入3Dvar系统进行同化分析，为高分辨率模式提供初值场。以台风“妮妲”登陆为例，通过一维反演及三维变分系统分析，有效地订正了实况有回波而模式预报无回波区域的大气湿度趋于合理，增加背景场的湿度，减小模拟回波比观测偏强的区域的大气湿度；同化大气湿度后模式在前6小时报出的台风外围回波分布、演变更合理，改进了降水雨带的分布与强度。1个月的批量试验反映1D+3Dvar同化雷达资料后，大气对流层中低层（850~400 hPa）增湿明显，其增湿影响程度可延续12小时以上。其逐时降水预报在前12小时的TS均比控制试验高，而大于5 mm以上降水预报偏差则与控制试验的大略一致或更接近1。Abstract: Doppler radar data are among a few that could provide convective scale information for high-resolution numerical weather prediction model (NWP). To make use of these data for initial condition of NWP, in this paper, a one-dimensional (Baiyesian) retrieval of relative humidity followed by a three-dimensional variational data assimilation (1D+3Dvar)technique is adopted. Quality control is conducted before the retrieval data enters 3Dvar system. Take a case of landing typhoon as an example, the one-dimensional Baiyesian retrieval effectively revises the humidity by moistening those areas where the model is unable to reproduce reflectivity and drying those areas where the intensity of simulating reflectivity is greater than that of reflectivity with observation. An experiment with assimilation of humidity indicates that it could successfully predict the distribution and evolution of reflectivity on the periphery of typhoon, and subsequently improve the skill of predicting distribution and intensity of six-hour accumulative precipitation. With the 1D + 3Dvar scheme, one month of experiment indicates that the troposphere from middle to lower level (850~400 hPa) humidity is significantly moistened, and the degree of its impact can be extended for more than 12 hours. TS(threat score) for hourly precipitation forecast is higher in the first 12 hours, and the forecast bias for precipitation greater than 5 mm/h is closer to 1 or remains unchanged compared with the control experiment.
-
Key words:
- Radar reflectivity /
- retrieval of humidity /
- Baiyesian method /
- assimilation experiment /
-
图 3 2016年8月1日15时3 km高度的回波与相对湿度
a.实况回波；b.模式预报的回波(单位：dBz）；c.背景场相对湿度(%)；d.反演同化前后相对湿度增量（单位：%）。
图 7 2016年5月1—31日期间有无同化试验的00时分析场，3小时、12小时相对湿度预报场平均差异廓线图
CTRL为控制试验，BYE为同化试验。单位：%。紫色：分析场；蓝色：3小时预报；红色：12小时预报。计算范围为105~119 °E，21~26 °N。纵坐标为气压(hPa)，横坐标为相对湿度偏差（%）。
图 8 2016年5月1—31日期间00时起报的1~12小时之平均逐小时降水大于1 mm、大于5 mm的TS
CTRL为控制试验，BYE为同化试验，纵坐标为TS，横坐标为预报的时间。
图 9 2016年5月1—31日期间00时起报的1~12小时之平均逐小时降水大于5 mm的预报偏差
CTRL为控制试验，BYE为同化试验。图中纵坐标为预报偏差值，横坐标为预报的时间。
图 10 2016年5月1—31日期间00时起报的1~12小时之平均逐小时2 m温度RMSE
CTRL为控制试验，BYE为同化试验。纵坐标为温度RMSE（单位：°），横坐标为预报的时间（单位：小时）。
表 1 试验设计
试验 日期、起报时间 目的 台风个例试验 2016年8月1日15时从15时起报12小时 初值场1D+3Dvar后的水汽增量及其对短临降水影响分析 批量试验 2016年5月1—31日共31天从00时起报12小时 初值场有无1D+3Dvar对逐时降水TS、2 m温度等影响比较分析 
下载: 导出CSV
-
[1] ROGER DALEY. Atmosphereic data analysis[M]. Cambridge: Cambridge University Press, 1991: 1-457. [2] 薛纪善.气象卫星资料同化的科学问题与前景[J].气象学报, 2009, 67(6):903-911. [3] ZOU X. Tangent linear and adjoint of "on-off" processes and their feasibility for use in 4-dimensional variational data assimilation[J]. Tellus, 1997, 49(1): 3-31. [4] XIAO Q, SUN J. Multiple-radar data assimilation and short-range quantitative precipitation forecasting of a squall line observed during IHOP_2002[J]. Mon Wea Rev, 2007, 135(10): 3381-3404. [5] WANG H, SUN J, ZHANG X, et al. Radar data assimilation with WRF 4D-Var. Part Ⅰ: System development and preliminary testing[J]. Mon Wea Rev, 2013, 141(7): 2224-2244. [6] STEPHAN K, KLINK S, SCHRAFF C. Assimilation of radar-derived rain rates into the convective-scale model COSMO-DE at DWD[J]. Q J R Meteorol Soc, 2008, 134(634): 1315-1326. [7] WATTRELOT E, CAUMONT O, MAHFOUF J F. Operational implementation of the 1D+3D-Var assimilation method of radar reflectivity data in the AROME model[J]. Mon Wea Rev, 2014, 142(5): 1852-1873. [8] BREWSTER K. Application of a Bratserth analysis scheme including Doppler radar data[C]//15th Conference on Weather Analysis and Forecasting. Amer Meteor Soc, Norfolk, VA, 1996: 92-95. [9] WIKLE C K, BERLINER L M.A Bayesian tutorial for data assimilation[J]. Physica. Section D: Nonlinear Phenomena, 2007, 230(1-2): 1-16. [10] CAUMONT O, DUCROCQ V, WATTRELOT. é, et al. 1D+3DVar assimilation of radar reflectivity data: A proof of concept[J]. Tellus, 2010, 62(2): 173-187. [11] MARTIN R, DAHLBOM M. Assimilation of multinational radar reflectivity data in a mesoscalemodel: a proof of concept[J]. Appl Meteor Climatol[J]. 2017, 56(6): 1739-1751. [12] 张诚忠, 薛纪善, 张林, 等.基于雷达反射率反演降水率的一维+三维变分同化研究及试验[J].气象学报, 2012, 70(5): 1128-1136. [13] 范水勇, 王洪利, 陈敏, 等.雷达反射率资料的三维变分同化研究[J].气象学报, 2013, 71(3):527-537. [14] 陈子通, 黄燕燕, 万齐林, 等.快速更新循环同化预报系统的汛期试验与分析[J].热带气象学报, 2010, 26(1): 49-54. [15] 薛纪善, 陈德辉.数值预报系统GRAPES的科学设计与应用[M].北京:科学出版社, 2008: 383. [16] 邹晓蕾.资料同化和应用[M].北京:气象出版社, 2009:120. [17] SMITH P L, MYERS C G, ORVILLE H D. Radar reflectivity factor calculation in numerical cloud models using bulk parameterization of precipitation[J]. J Appl Meteor, 1975, 14(6): 1156-1165. -
点击查看大图
图(10) / 表(1)
计量
- 文章访问数: 737
- HTML全文浏览量: 32
- PDF下载量: 279
- 被引次数: 0
粤公网安备 4401069904700003号