Research on Sea Surface temperature Reconstruction from long-term MODIS data

Abstract

Sea surface temperature (SST) is one of the important dynamic parameters to characterize the physical characteristics of seawater, and it is also an important basic auxiliary parameter for the quantitative observation of other ocean parameters. In view of the lack of remote sensing data mainly caused by cloud cover over the ocean, based on the long-term MODIS Aqua L3 8-day mean SST products from January 2003 to March 2015, the missing data in the East China Sea are reconstructed by Data Interpolating Empirical Orthogonal Function (DINEOF) method, and the reconstruction results are evaluated by error analysis before and after cross-validation data reconstruction. It shows that the image reconstructed by DINEOF method can reflect the temporal and spatial variation characteristics of sea surface temperature in the study area, and the root mean square error of reconstruction is 0.4272. DINEOF method can reconstruct large area missing data without any prior information and has high accuracy.