Inter-satellite comparisons and evaluations of three ocean color products along the Zhejiang coast, eastern China

摘要

With its broad spatial coverage and fine temporal resolution, ocean color remote sensing data represents an effective tool for monitoring large areas of ocean, and has the potential to provide crucial information in coastal waters where routine monitoring is either lacking or unsatisfactory. The semi-analytical or empirical algorithms that work well in Case 1 waters encounter many problems in offshore areas where the water is often optically complex and presents difficulties for atmospheric correction. Zhejiang is one of the most developed provinces in eastern China, and its adjacent seas have been greatly affected by recent rapid economic development. Various islands and semi-closed bays along the Zhejiang coast promote the formation of muddy tidal flats. Moreover, large quantities of terrestrial substances coming down with the Yangtze River and other local rivers also have a great impact on the coastal waters of the province. MODIS, VIIRS and GOCI are three commonly used ocean color sensors covering the East China Sea. Several ocean color products such as remote-sensing reflectance (R_(rs)) and the concentrations of chlorophyll a (Chl-a) and total suspended matter (TSM) of the above three sensors on the Zhejiang coast have been evaluated. Cloud-free satellite images with synchronous field measurements taken between 2012 and 2015 were used for comparison. It is shown that there is a good correlation between the MODIS and GOCI spectral data, while some outliers were found in the VIIRS images. The low signal-to-noise ratio at short wavelengths in highly turbid waters also reduced the correlation between different sensors. In addition, it was possible to obtain more valid data with GOCI in shallow waters because of the use of an appropriate atmospheric correction algorithm. The standard Chl-a and TSM products of the three satellites were also evaluated, and it was found that the Chl-a and TSM concentrations calculated by the OC3G and Case 2 algorithms, respectively, were more suitable for use in the study area. Moreover, GOCI has been proved to be effective for monitoring the diurnal dynamics in coastal waters, and the concentration of TSM had a good negative correlation with water level. Overall, compared with MODIS and VIIRS, GOCI is more effective for monitoring the fine changes and diurnal dynamics in the seas adjacent to Zhejiang Province.