Modification Of Z-R Relations In TRMM PR Rain-profiling Model Over Qinghai Tibet Plateau And Its Application

The high altitude and complex topography lead to sparse distribution of ground observations over Qinghai-Tibet Plateau. In order to gain recognition of the spatial and temporal characteristics of precipitation in the Plateau, different precipitation estimation techniques are introduced to demonstrate their advantages and disadvantages in the first place. Although the accuracy of TRMM PR products is relatively higher compared to other satellite data, its reliability over the Plateau is still unclear.Based on the TRMM PR products and rain gauge data from 2005 a to 2007 a, the instantaneous bias and seasonally average bias are analyzed depending on different rain types and seasons. In the comparison, the results of rain gauges are regarded as the truth to analyze the error of TRMM PR. It showed that there is a good correlation between PR estimated surface rainfall rate and rain gauge rainfall rate in the stratiform rain over the plateau. On the contrary, the correlation in the convective rain events is relatively worse. PR has an overall bias of-13%, with a negative bias of 35% in stratiform rain over the plateau and a positive bias of 42% in convective cases. Then the rain-profiling algorithm of TRMM PR is analyzed to demonstrate the variation of the coefficient A and b in the Z-R relationships, It is known that the PR rainfall rates are calculated from the radar reflectivity factor Z profiles by using a power law: Z=ARb in which the parameters A and b are both functions of the 0°C isotherm and storm top. In this work, new Z-R relationships more suitable for the precipitation of Qinghai-Tibet Plateau are developed by using Z from PR and Rg from rain gauges.On the basis of the study above, the initial coefficients in Z-R relationships on the level of 20 degree Celsius are modified to make the rain-profiling model more suitable for the precipitation over Qinghai Tibet Plateau. Then the rainfall rates derived from the modified model and from PR 2A25 products are compared with corresponding rain gauge data to examine the accuracy of the modified algorithm, the results suggest that the differences between the revised PR surface rainfall rates and rain gauge observations over Qinghai Tibet Plateau are reduced, thus the modified model can achieve better accuracy over Qinghai Tibet Plateau.But the temporal resolution of PR is rather low as to missing some important weather systems over Qinghai Tibet Plateau. In this case, FY-2C VISSR precipitation estimation products are utilized in the next section to improve the temporal resolution of precipitation products estimated by satellites. However, the accuracy of FY-2C VISSR precipitation estimation products are relatively worse on the condition that the results are estimated empirically instead of calculated by a rain-profiling model. So the error analysis is vital, FY-2C VISSR precipitation estimation products are checked by comparing with the revised PR data with higher accuracy to analyze the bias and correlation between them. Additionally, the results of FY-2C VISSR precipitation estimation products are adjusted by using the way of linear regression to make them closer to the results of revised TRMM PR estimation on the whole. At last, the way of modification is examined by importing independent samples to demonstrate the rationality of the correction procedure.