A Method Of Hj-1b Thermal Infrared Camera Water Surface Temperature Retrieval

Water surface temperature which is widely used in fishery resources, climate changes, environmental changes detection and weather, etc, is one of the important characteristics which indicates the physical characteristic of water body. As the continuous development of aerospace remote sensing cause, satellite remote sensing has been successfully applied to measure the marine environment and has played a significant role in monitoring, showing great potential for application. Many foreign satellites have carried thermal infrared sensors, such as NOAA/AVHRR, TERRA/MODIS, Landsat/TM(ETM+), etc, and have a large number of studies based on theory and application in temperature retrieval. In this aspect our country is alse not to be outdone, after the CBERS thermal infrared camera disabled, we have launched HJ-1A/B satellite on September 6th,2008, of which HJ-1B has carried an infrared camera, made up the blank of getting thermal infrared data using domestic satellite, and can also be applied to detect environmental factors due to reasonable band settings. We use HJ-1B thermal infrared data to retrieve the water surface temperature. Around the research, we have carried out the following three studies.Firstly, due to comparability of spectral response of HJ-1B IRS thermal infrared channel and TM6, we used temperature retrieval algorithm which developed for TM6 data. Then we revised the Jimenez-Munoz and Sobrino single-channel algorithm to meet the IRS8 temperature retrieval, based on HJ-1B IRS8 spectral response characteristics.Secondly, we study the total precipitable water algorithm based on near-infrared differential absorption technology which applied to MOD05 product and dual-channel sea surface temperature retrieval algorithm which can eliminate atmospheric effects and applied to MODIS temperature retrieval. It is difficult to obtain real-time water vapor profiles, we chose the MOD05 product as input of retrieval model, and chose MODIS temperature products as validation data, because pass time of HJ-1B is close to that of TERRA.In the end, we employed IRS8 five data granules which have a coverage of parts of the Bohai Sea and Yellow Sea to retrieve the temperature, and the results suggest that retrieval temperature is generally higher than MODIS products, and the lower temperature the difference would be greater. Among the five data granules there is one data granule on October 2nd,2009 when pass time is close to each other, and result is much perfect, that is the points of relative error is less than 10% account for 97.26%, the points of relative error is less than 5% account for 84.75%, and the RMSD is less than 1. Then explain that the thermal infrared data of HJ-1B also has a high availability compared with that of MODIS.