현장스펙트럼을 이용한 Case-II 해역에서의 SSMM 대기보정

Abstract

Application of the Landsat TM/ETM+ signal over aquatic environment requires the retrieval of water-leaving radiance (Lw) from the total radiance recorded at the top-of-the-atmosphere (TOA) and thus of the water constituents’ of interests. This paper compares the image-based atmospheric corrections such as empirical line method, COST method, and Spectral Shape Matching Method(SSMM) in the Landsat TM/ETM+ imagery, especially from Case-II waters around Korean peninsula. The SSMM which is simple and easy to implement on any satellite imagery relies on the assumption that spectral form of water-leaving radiance or reflectance for typical turbid or clear waters is nearly stable and provide a way to extract the path signal from the total signal recorded at the top of the atmosphere (LTOA). To assess the potential use of SSMM, several field campaigns were conducted in the Seamangeum area located at central western coastal area, KOREA corresponding with the Landsat-7 satellite's schedules. In situ observations were made from the coastal waters and land objects around the Seamangeum using ASD field spectroradiometer, consisting of Chl, Ap, SS, aDOM, F(d) . Firstly, SSMM was applied using the nearly stable spectrum from ocean area and the results were verified with in situ water-leaving spectra. Secondly, the SSMM was applied using spectra obtained from the land area and the results were compared with that from using stable ocean water-leaving radiance spectra. Finally, the performance of SSMM was compared with Empirical Line Method and COST model. The findings revealed that SSMM produced spectra consistent with in situ data, whereas other two methods sometimes yielded significantly high errors in these areas.