The East China Sea Oil Spill Information Identification Based On Multi-source Remote Sensing Data

With the development of modern economy, it has greatly increased in the demandof energy, the oil as one of the most important energy has lead to the rapid developmentof the marine transportation. However, in recent years, the more attention about the oilspill at sea has been paid owning of the frequent oil spill accidents. The oil spill at sealead the significant damage to the marine ecological environment, causing thesignificant death of the marine fish, birds, aquatic plants and so on, besides, it also causethe heavy losses for the local tourism industry. For example, the oil spill of the DalianNew Port on Jul.16,2010had led the massive oil leak, basically making the localaquatic farming stop producting, at present some enterprise cannot still be in the normaloperation in the region. The East China Sea is not only an important fishery resourcesproduction area but also an important shipping channel, which has great influence onChina’s economic development, so that it is significant for maintaining the stability ofthe ecological environment in the East China Sea to prevent and mitigate disaster.Remote sensing technology as one of the important methods of Marineenvironment monitoring has been widely applied, the development of the satellitesensors have promoted the sea oil spill monitoring, so that this paper introduce thedevelopment history of the satellite sensors and explain the corresponding applicationfields, especially analysis the MODIS sensor and synthetic aperture radar (SAR) sensorwith extensive application. MODIS sensor which equipped on the EOS series ofsatellites launched by the United States Aeronautics and Space Administration, with thefeature of multi-band, multi-resolution, we can enhance the oil spill information in thepicture by the multi-band synthetic. SAR is a microwave radar sensor with highresolution, so we can use it for the oil spill information extraction in the small area. Inaddition, the changes of the marine oil spill evaporation, diffusion, drift and so on, havegreat influence by the terrain, runoff, temperature, water temperature, wind speed, winddirection, water masses, circulation, tides, tidal current, waves and other ocean factors.This paper introduce the ocean background environment for the East China Sea(BohaiSea, Yellow Sea and East China Sea), provide the basis for the diagnosis of oil spillinformation and the diffusion of the oil spill area. This paper uses three remote sensing sources (i.e., MODIS, Radarsat-2andENVISAT-1remote sensing data) to extract the oil spill information at sea, we analysisthe Dalian "7.16" marine oil spill by MODIS and Radarsat-2data, whose methods bothhave advantages and disadvantages, MODIS data with36spectral bands can bemulti-channel color composite, thereby enhance the extraction of the oil spillinformation, besides the cycle of accessing the MODIS remote sensing data is relativelyshort, we can get four scene picture one day in the same area. However, The resolutionsof MODIS data have three kinds of250m,500m and1000m, which is relatively lowerthan Radarsat-2data, moreover, the weather has great influence on MODIS data. Afterthe pre-processing of MODIS remote sensing data, take measure of band math toremove a little cloud and extract oil spill information.Radarsat-2data is a single-bandradar images with high resolution, so the detailed information can be extracted, besidesit has little influenced by the weather of cloud and fog. However,Radarsat-2data with single-band information can’t enhance information with true colorcomposite, It is difficult to identify the oil slick and sea water with the lower wind speed(<3m/s)，It is also difficult to exclude the interference of sea water with the higherwind speed（＞8m/s）, Radarsat-2remote sensing data, after the pre-processing ofgeometric correction, filtering and so on, take the method of texture analysis to extractoil spill information.The East China Sea is not only an important fishery resources production area butalso an important shipping channel, there is a great probability of ship oil spill risk. Thispaper analysis the28scene ENVISAT-1remote sensing images in the East Seasbetween in Jan.2004and in Sep.2010, By the analysis the ship oil spill information,the result is that the suspected ship oil spill information exists in6scene ENVISAT-1remote sensing images, There is ship oil spill information in two scene images on16th,May.2005and on13th, Sep.2010by texture filtering and the gray value analysis.The innovation of this paper: Extracting the oil spill information by MODISremote sensing data, take measure of band math to enhance the accuracy of oil spillinformation according to the Multi-band feature of MODIS data, in addition, this articlealso make use of Radarsat and ENVISAT remote sensing data, so that realize to analysisthe sea oil spill by multi-source remote sensing data, It can achieve the function of eachother for extracting oil spill information.