The Study On The Effect Of Directly Assimilating Satellite Radiance On Heavy Rain Forecast In Yangtze River Basin

In recent yearsnatural disasters, like heavy rain and hail frequently happened,which makes human life inconvenient and huge losses of social economy. Assimilation of hyperspectral infrared radiance data,which provides high resolution of temperature and humidity profiles, has significantly improved the forecast accuracy in global model.For the regional model, the infrared hyperspectral data assimilation still have many problems to be solved.Nowadays,there are lots of people have tried to assimilate AIRS data into regional model.Howerer, for hyperspectral infrared IASI data generated from Europe, domestic research is limited to the typhoon simulation.In this paper, three-dimensional variational assimilation method is combined with WRF mesoscale model to explore the improving effect in simulated rainstorm by assimilating the IASI data in regional model.Firstly,we use the NMC method for statistical background error covariance in the study area by one month forecast results of June 2014. The event is based on a severe precipitation weather process that occurred in the middle-lower braces of Yangtze River during 2-28 June 2014.we design two experiments.The first test is order to determine specific IASI data assimilation band, by comparing the results of assimilating Conventional Observation, IASI temperature band, IASI humidity band and the all IASI band, to confirm which result is more ideal.The second test assimilate IASI temperature band, AMSUA, MHS, HIRS4, ATOVS data, and compare their results.The results show that:(1)According to background error covariance matrix analysis,the non equilibrium temperature and relative humidity is a strongly local sex, and the stream function and non equilibrium velocity potential are greatly influenced by the boundary layer.(2)The results of assimilation IASI temperature band test show that:the cost function and gradient are more likely to achieve convergence; the initial field is more reasonable; rain fall area and the simulation of rainfall intensity is more closer to reality. So in this case we mainly choose assimilation IASI temperature detection wavelengths.(3)Comparing the result of assimilation IASI temperature detection band test with other schemes, we can get there are more assimilation data points are within the system, and the initial field and the weather situation field simulation are more reasonable, precipitation and TS score results are better than others,the error analysis of certain highly root mean square error is less than other schemes.(4)Comparing the result of assimilation AMSUA, MHS, ATOVS three testing scheme, we found on precipitation field simulation, and the error analysis of assimilation AMSUA and MHS project results are pretty, but the result of assimilation ATOVS (AMSUA+MHS+HIRS4) is worse. The reason maybe: assimilation too much data may lead to larger observational error, cancel each other out positive effect, and lead to worse results.