NUMERICAL SIMULATION ANALYSIS OF A HEAVY RAINFALL IN SOUTH CHINA WITH FOUR-DIMENSIONAL ENSEMBLE VARIATIONAL ASSIMILATION
YANG Yu-xuan
ZHANG Li-feng
ZHANG Bing
ZHANG Ming-yang
XIE Bin-peng
Abstract：The heavy rain event that occurred in South China on 9 December 2015 is studied by using Weather Research and Forecasting (WRF) and the Proper Orthogonal Decomposition-based four-dimensional ensemble variational assimilation (POD-4DEnVar) method to assimilate real Doppler radar observations.A WRF-3DVar contrast experiment is designed and the sensitivities of localization radius to the simulated precipitation are discussed.The contrast experiment results show that the radar data assimilation both using 3DVar and POD-4DEnVar method improves the prediction of precipitation compared with the control experiment,and the effect of POD-4DEnVar method is better than that of 3DVar method.The two assimilation methods are aimed at improving the effect of precipitation simulation by changing different initial factors.The 3DVar method indirectly reduces the water vapor conditions required for the storm by adjusting the initial wind field,and the POD-4DEnVar method directly adjusts the humidity field.During the precipitation,the assimilation experiments change the activity of the cold air and the water vapor flux convergence,which improves the effect of precipitation simulation.The sensitivity experiments show that the POD-4DEnVar assimilation method is sensitive to the localization radius,and the influences of assimilation on the wind and humidity field increase with the increase of localization radius.The localization radius of about 200km can achieve the best prediction of precipitation.
Keywords：four-dimensional ensemble variational technique3DVarradar dataassimilationlocalization radius
Publication Date：2018-01-01
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：11( 217-227 )