Satellite observations of oil spills in Bohai Sea

35th International Symposium on Remote Sensing of Environment

地球科学;生态环境科学

Wei, Y.L.^1,2 ; Tang, Z.Y.^3 ; Li, X.F.^2,4

College of Marine Sciences, Shanghai Ocean University, Shanghai 201306, China^1

International Center for Marine Studies, Shanghai Ocean University, Shanghai 201306, China^2

East China Sea Prediction Center, State Oceanic Administration of China, Shanghai 200081, China^3

GST, NOAA/NESDIS, College Park, MD 20740, United States^4

关键词: Gradient edge detections;    Neural network algorithm;    Oil detections;    Optical image;    SAR(synthetic aperture radar);    Satellite observations;    Satellite sensors;    Semi-supervised;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/17/1/012114/pdf
DOI  :  10.1088/1755-1315/17/1/012114

学科分类：环境科学（综合）

来源: IOP

【 摘 要 】

Several oil spills occurred at two oil platforms in Bohai Sea, China on June 4 and 17, 2011. The oil spills were subsequently imaged by different types of satellite sensors including SAR (Synthetic Aperture Radar), Chinese HJ-1-B CCD and NOAA MODIS. In order to detect the oil spills more accurately, images of the former three sensors were used in this study. Oil spills were detected using the semi-supervised Texture-Classifying Neural Network Algorithm (TCNNA) in SAR images and gradient edge detection algorithm in HJ-1-B and MODIS images. The results show that, on June 11, the area of oil slicks is 31 km2and they are observed in the vicinity and to the north of the oilfield in SAR image. The coverage of the oil spill expands dramatically to 244 km2due to the newly released oil after June 11 in SAR image of June 14. The results on June 19 show that under a cloud-free condition, CCD and MODIS images capture the oil spills clearly while TCNNA cannot separate them from the background surface, which implies that the optical images play an important role in oil detection besides SAR images.