Comparision Between NCEP And NASA Reanalysis Data In Qinghai-Tibet Plateau Area And Activities Of South Asia High And Its Influence On Drought And Flood Anomaly Of China In Summer

Using monthly high resolution seismic reanalysis data MERRA from NASA, NCEP-DOE(NCEP-2) reanalysis data and monthly rainfall data from 160 primary stations in China during 1979~2009 in summer, through the bilinear interpolation, correlation analysis, synthetic analysis, Morlet wavelet analysis and SVD analysis methods, surrounding on the usability of new high resolution seismic reanalysis data MERRA, we compare circulation and system characteristics ( particularly in Qinghai-Tibet Plateau area and South Asia High) between MERRA and NCEP-2 reanalysis data. Further, we discuss influences of 100 hpa geopotential height anomaly of MERRA reanalysis data on summer drought and flood in China. Finally,activity characteristics of 100hPa South Asia High of MERRA reanalysis data and their likely mechanism and relations to drought and flood anomaly in Yangtze River Valley are clearly investigated. The main results are as follows:(1) In East Asia, flow patterns of geopotential height and temperature field between MERRA and NCEP-2 data are basically identical, but there are also differences: the largest differences are located in the Qinghai-Tibet Plateau subject area, and in two data differences of 600hPa are greater than differences of 300hPa and the differences of 100hPa are smallest. The results also show that there is a good consistency in the meridional and zonal circulation of 100hPa between MERRA and NCEP-2 data, but the main differences between them appear at 600hPa and land surface. Especially in complicated topography regions differences of the vertical motion exist and the positions of secondary circulation are also different. In addition, the consistency of distributions and intensity of divergence and vorticity described by MERRA and NCEP-2 data is that, the Qinghai-Tibet Plateau area is lower than its adjacent areas, and 300hPa and 100hPa are higher than 600hPa. In the Qinghai-Tibet Plateau main area and the surrounding areas, the trends of MERRA data both in geopotential height and temperature interannual variabilities are coincident with the trends of NCEP-2 data, which shows that MERRA data at 600hPa and 300hPa is reliable as well as NCEP-2 data. The obvious differences of geopotential height interannual variability trend between MERRA and NCEP-2 data are found at 100hPa. From comprehensive analysis, we can get: MERRA data is a better high-resolution reanalysis data so far, especially in Qinghai-Tibet Plateau area and its surrounding areas.(2) At 100hPa, value of South Asia High (SAH) characteristic line in MERRA is smaller than value in NCEP-2 and its value is 1676dgpm, but the scope of SAH in MERRA data is significantly bigger than the scope in NCEP-2 data. In addition to SAH eastward index, their interannual variability trends of the same SAH characteristic index are also similar, especially in two data, interannual variabilities of the SAH ridge line index curve coincide. Prior to the early 1990s, SAH eastward index, area index and intensity index in NCEP-2 data show positive anomalies; the same indices in MERRA data show negative anomalies; corresponding to the indices in NCEP-2 data, the index value in MERRA data is smaller; after the early 1990s, and vice versa. There is no significant difference between means of SAH eastward and ridge line index in two data. Only variances of SAH intensity index between NCEP-2 and MERRA data have significant differences, and NCEP-2 greater than MERRA. The wavelet variance verification indicates that the same SAH characteristic index of NCEP-2 and MERRA data during 1979 and 2009 in summer are in good agreement: both have the same significant oscillation periods, and their phases are also consistent, but there are some differences between their predominant period: in MERRA data SAH area and intensity index performance quasi-4 years oscillation as the predominant period; in NCEP-2 data SAH area and intensity index performance quasi-4 years and 8 to 9 years oscillation at the same time.(3) By use of SVD method twice, the relationship between summer 100hPa geopotential height field of MERRA data and the corresponding period 500hPa geopotential height field is studied firstly(SVD1), then we discuss influences of Summer 500hPa geopotential height field of MERRA Reanalysis Data on the corresponding period rainfall field in China(SVD2). The results show that there is closely time-space correlation between two fields of SVD1 or SVD2. In SVD1, cumulative covariance contribution of the first two modes is 99.22%, and correlation coefficients of them are above 0.75, which can well reflect the main features of the interaction between the two fields. In SVD2, cumulative covariance contribution of the first two modes is 74.95%, and T-tests are showed no significant difference (α= 0.001). The distribution of the first mode in SVD1 indicates that when summer 100hPa geopotential height is negative(positive) anomalies in SAH region, the corresponding period 500hPa geopotential height in the Subtropical High region will be negative(positive) anomalies. The distribution of the first mode in SVD2 indicates that when summer 500hPa geopotential height in the Subtropical High region is lower (higher) than normal, the precipitation in Yangtze River Valley is less(larger). SVD analysises show that in the first mode of two SVDs 10~32°N, 110~160°E is significant key region at 500hPa, and it reflects main information of strength changes of the Subtropical High. The distribution of the second mode in SVD1 indicates that when summer 100hPa geopotential height in 25~40°N, 40~85°E region is negative(positive) anomalies, and the postion of SAH ridge line is more northward (southward)than that in the normal years, then the corresponding period 500hPa geopotential height in 30~40°N, 110~145°E region will be negative(positive) anomalies, and the postion of the Subtropical High ridge line is more northward (southward)than that in the normal years. The distribution of the second mode in SVD2 indicates that when summer 500hPa geopotential height in 30~40°N, 110~145°E region is higher than normal, and the postion of the Subtropical High ridge line is more northward than that in the normal years, the precipitation in South China, Yunnan, east of southern Yangtse River and Nouth China, central region of Inner Mongolia and northern region of northeast China, is larger, the precipitation in whole Yangtze River Valley is smaller, and vice versa. SVD analysises show that in the second mode of two SVDs 30~40°N, 110~145°E is significant key region at 500hPa, which reflects main information of ridge line changes of the Subtropical High. Due to the conclusion of SVD, we suggest we study and analyse activity characteristics of SAH by using 100hPa MERRA data.(4) The results show that precipitation in Yangtze River Valley was generally less than normal during 1979~1989, from 1990 to 1999 more rainfalls appeared, even frequent floods occured, and after 2000 precipitation had kept decreasing, we can find persistent drought in these days. The drought and flood in Yangtze River Valley from 1979 to 2009 is closely related with atmospheric circulation at 500 hPa and the corresponding period of 100hPa, wind field at 850hPa, and preceeding sea surface temperature anomalies (SSTA); it is found that there are obvious differences of troposphere and stratosphere circulations as well as SSTA between drought and flood years. In flood years, the Subtropical High at 500hPa shifts southward and westward and becomes strengthened, and SAH shifts eastward to east of 100°E and its intensity is stronger than usual, especially at 850hPa, there is a convergent region in Yangtze River Valley formed by strong cold air from the north and airflow of warm and moist from the low-latitude area, meanwhile sea surface temperature of Western Pacific, Northwest Pacific, the South China Sea and the North Indian Ocean exhibits positive anomalies; and vice versa in drought years. The correlation analysis shows that: in summer, Nino3 +4 area (5°S~5°N, 120~170°W) SSTA positive (negative) anomalies will result in the following year intensity of SAH stronger (weaker) and shifting eastward (westward ), and the relationship between 100hPa SAH and the Western Pacific Subtropical High at 500hPa is "Relative Motion", thereby affect precipitation in Yangtze River Valley.