溢油海面耀光反射率的遥感估算研究

溢油是海洋环境监测的重要对象。在海洋环境动力作用下，溢油会形成乳化油、油膜等多种污染类型；准确识别不同溢油污染类型并估算溢油量，有助于海洋溢油事件的应急处理与灾损评估。光学遥感技术具备对海面溢油进行识别、分类与定量估算的能力。海面溢油的光学遥感应用研究中，太阳耀光反射差异有利于目标探测的同时，也给其识别分类与定量估算带来不确定性影响。面向溢油海面耀光反射率的遥感估算难题，本研究拟通过小场景仿真实验，反演典型溢油污染类型的等效折射率，并构建参数查找表；促进卫星光学遥感数据中溢油海面粗糙度的遥感反演，发展顾及风速的溢油海面粗糙度统计函数；解决溢油海面耀光反射率计算模型的核心参数缺失问题。针对不同卫星光学遥感数据，尤其是中国海洋一号C星数据，面向典型溢油污染事件（包括中国近海溢油污染），通过溢油海面耀光反射率的优化计算与验证，提高海面溢油量的光学遥感估算精度，促进海洋溢油的光学定量遥感研究。

Remote detection and quantification of oil spills represent one of the key research directions in marine environment monitoring. During the weathering process, spilled oils can form different floating pollution types, such as oil slick and oil emulsions. Optical identification and quantification of various spilled oils play an important role in oil spill monitoring, assessment and the resulting response. Spaceborne and airborne multi/hyper-spectral optical sensors have been used to detect, quantify, and map oil spills. Sunglint reflection of different surfaces (oil-free and oiled surface) helps detection but presents challenges on classification and quantification of various spilled oils. How to accurately calculate the sunglint reflectance of various spilled oils is still an important challenge for optical remote sensing of marine oil spill. Considering the difficulty of remotely inverting of both parameters (refractive indexes and roughness), an outdoor simulation experiment will be designed to collect the sunglint reflectance of various oils with the same roughness. A look-up table (LUT) of equivalent refractive indexes of various oils will be generated from observations. Based on the optical remotely identification of spilled oils, this LUT could be used for estimation of surface roughness of spilled oils under sunglint of different optical remotely imageries. And then, the statistical function between wind speed and surface roughness of spilled oils could be given. This LUT and statistical function can provide a better understanding of the oil-water contrast and the impact of sunglint on oils surface reflectance. If the sunglint reflectance of spilled oils could be accurately calculated, the laboratory-established models between oil thickness or concentration and optical contrast would be applied to estimate spilled oil volume from different optical remotely sensed imageries, especially form Chinese HY-1C covering oil spills in the coastal water of China. Optical remote sensing will play an increasingly important role in assessing marine oil spills.