| Sea surface temperature (SST) is an important parameter of sea environment, especially for the Yangtze estuary. An accurate retrieval of SST is helpful to study the estuarine hydrodynamic environment and marine phenomenon such as upwelling and temperature fronts. Affected by both land runoff and ocean currents, estuarine water is strongly mixed, which makes it different from ocean water in terms of thermal infrared radiation properties and atmospheric conditions. So the algorithm for ocean water is not tuned for the estuarine water.To address this issue, a new Yangtze Estuarine Sea Surface Temperature (YESST) algorithm was developed in this study. To adapt the estuarine environment, the YESST algorithm was based on Qin split-window algorithm and optimized in the calculation of two key parameters which are atmospheric transmittance and sea surface emissivity. The YESST algorithm was applied to Terra-MODIS L1B data and validated with both in situ water temperature data and the standard MODIS SST product. Analysis showed improved accuracy in terms of bias and RMSE when compared with the standard MODIS SST product. The bias is reduced by 0.23℃ and RMSE is reduced by 0.62℃. Including 624 images, a thirteen-year SST dataset, from year 2000 to 2013, of the Yangtze estuarine waters was retrieved by the YESST algorithm. Not only the SST spatial distribution but also the seasonal and inter-annual SST variations of the research area were revealed according to the dataset. The results show that the SST of the Yangtze estuarine water presents a stepwise change spatially from the upstream to the lower reaches of the estuary, which is dominated by the solar radiation. In details, the SST outside the river mouth is higher than the SST in the inner estuary in winter, and it is reversed in summer. On January 29,2007, for example, the SST of the research area is between 3℃ and 13℃, and on July 29,2007, the SST can be as high as 24℃ to 35℃. The temperature gradient from the upstream to the lower reaches of the estuary is higher in winter than in summer. This indicates that the SST in estuarine area is the result of interaction between the land runoff and sea currents, and the terrestrial water is a heat source in summer and a cold source in winter. In addition, the Three Gorges Reservoir (TGR) began regulating the Yangtze water discharge from 2003. The increased runoff in winter season, which means more cold water is input into the estuary, has a more significant cooling effect on SST outside the river mouth than the SST in the inner estuary according to the continuous daily SST data retrieved. But in summer season, when solar radiation plays a decisive role in the SST of estuarine waters, the decreased land runoff doesn’t affect the SST obviously.A morphology-based edge detection method is used to study the sea surface temperature fronts in Yangtze Estuary. This method is based on mathematical morphology, using multi-dimensionand multi-structure elements, which is strongly anti-noise. With this method, a SST fronts dataset in Yangtze Estuary is got based on the retrieved 14-year SST dataset, including the location and gradient magnitude of the fronts. According to this SST fronts dataset, the following coclusions is made. When it is neap tide in dry season, the locations of 6 different-imaging-time SST fronts in Yangtze Estuary shows relatively regular compared with the locations when it is spring tide in dry season. According to the gradient magnitude of Feburary 15,2001 and Feburary 20,2010, the influence of the cold waters taken by Subei Alongshore Current is obvious, which creating another SST front in west side of the SST front caused by Taiwan Warm Current. |
