资源环境科技发展态势分析平台

In recent years, water vapor as one of the most important greenhouse gases has played an essential role in many weather and climate processes, and it has been studied by various methods. Moderate Resolution Imaging Spectroradiometer (MODIS) sensor installed on Aqua and Terra satellites is one of the existing remote sensing tools capable of producing Precipitable Water Vapor (PWV) products. In this research, one year of ground-based observation from a network of 38 GPS stations were utilized to evaluate the accuracy of MODIS Near Infrared (Near-IR) PWV over Iran. Initially, GPS PWV estimates at three radiosonde stations (OIFM, OIII and OIMB) were evaluated. Agreement of GPS-derived PWV with respective values obtained from radiosonde in terms of bias and RMSE were 1.5 mm and 3 mm respectively. After evaluation of GPS PWV estimations in the study area, MODIS Near-IR cloud free water vapor products at the location of GPS stations were extracted and evaluated. Times series comparison between MODIS and GPS showed that both methods agree very well with a correlation coefficient of 0.95. The mean bias and RMSE of differences between these two techniques (GPS and MODIS) during one year were obtained -2.2 mm and 3.4 mm respectively. The negative value of bias indicates that MODIS Near-IR in comparison with GPS overestimates PWV in this area. In addition, it was found that the bias values and the height of the stations have a linear relation. Therefore, a linear model as a function of height was obtained to calibrate MODIS Near-IR PWV products. Results showed that the calibration with the help of the presented model increases the efficiency of MODIS Near-IR water vapor data. The Amount of bias reduction after calibration of MODIS data was obtained 2 mm.