Study For Retrieving And Estimating Precipitable Water Vapor Of Xinjiang Based On RS And GIS

As an important meteorological parameter, atmosphere precipitable water is the most active ingredient in the hydrologic cycle. Regional atmosphere precipitable water is one of important issues in the meteorology and climatology, Xinjiang is located in the northwest arid and semi-arid area, with an average annual rainfall is only 155mm. The water resource is a bottleneck of social economic development and ecological environment improvement in Xinjiang.Xinjiang features an immense area and a sparse population here, ground meteorological stations are scarce. Remote sensing is the only fast and effective method to detect the temporal, spatial distribution and variation characteristics of atmosphere precipitable water. By use of satellite remote sensing data and the foundation GPS to retrieve atmosphere precipitable water is widely used as a new technology in recent years, the advantages are continuous, real-time, high resolution, which can't be resolved by Radiosonde.Based on MODIS 1B data, combinated with foundation GPS atmospheric detection technology, this paper studied the spatial distribution of precipitable water in Xinjiang.This paper studied the spatial distribution of precipitable water combinated foundation GPS atmospheric detection technology and MODIS data in Xinjiang. The conclusion mainly divided into the following several parts:Firstly, discussed the means of retrieving APW using MODIS 1B data, we procured the AWP of two-channel ratio, two-channel ratio weighted and 18,19 channel ratio. Compared the result of three methods to the AWP of GPS over Urumqi. The results shows that the APW of two-channel ratio weighted method has higher dependability, the mean absolute error was 2.23mm, the correlation determination (R2)was 0.939, the average error was11.17%. Secondly, compared the AWP of the ground-based GPS to AWP of the Radiosonde (SONDE) data. It showed that the level of significance is 0.01, correlation coefficient is 0.911, and standard error is 2.66mm. And then, APW retrieved by MODIS was corrected based on relationship between AWP of GPS and those of MODIS. The correlation between AWP of GPS and those of MODIS was further increase after corrected, the mean absolute error was 1.53mm, the correlation determination (R2)was 0.940, the average error was 7.66%. According to the shape file of Xinjiang, the AWP image retrieved by two-channel ratio weighted method was divided into each area. Each area was corrected by Atmospheric pressure of each area. Those areas were spliced together as a whole after corrected.Thirdly,based on different shape files, the ARCGIS software was used to estimate the AWP. The mean values of AWP was calculated during July-September 2007 in each area. There was the highest AWP in KuiTun city, which is 23.79mm, the second is 23.51 in Shihezi, the third is 22.97 in Karamay, and the lowest is 9.68mm in Kezilesu Kirgiz Autonomous Prefecture.