LAGRANGIAN-BASED BACKTRACKING OF OIL SPILL DYNAMICS FROM SAR IMAGES : APPLICATION TO MONTARA CASE

摘要

Within the framework of INDESO project (InfrastructurernDevelopment Space Oceanography), we address the issuernof oilspill and aim at developing an operational SARbasedrnsystem for monitoring this issue in Indonesian watersrnfrom space. In this work, we focus on the backtrackingrnof an oilspill detected from SAR observations. As arncase-study, we consider one large oil spill event that happenedrnin Indonesian waters in 2009, referred to as thernMontara oilspill. On 21 August 2009, the Montara WellheadrnPlatform had an uncontrolled release of hydrocarbonsrnfrom one of the platform wells. It was estimatedrnthat 400 barrels (or approximately 64 tonnes) of crudernoil were being lost per day. The uncontrolled releaserncontinued until 3 November 2009 and response operationsrncontinued until 3 December 2009. In this work,rnwe develop a Langragian analysis and associated numericalrninversion tools with a view to further analyzing thernoil spread due to the Montara Wellhead Platform. Ourrnmodel relies on a 2D Lagrangian transport model developedrnby CLS (Collecte Localisation Satellite). Our modelrninvolves four main parameters : the weights of windrelatedrnand current-related advection, the origin and thernduration of the oil leakage. Given SAR oilspill detections,rnwe propose a numerical inversion of the parametersrnof the Lagrangian model, so that the simulated drift matchrnthe SAR observations of the oil spill. We demonstrate thernrelevance of the proposed model and numerical schemernfor the Montara oilspill and further discuss their operationalrninterest for the space-based oilspill backtrackingrnand forecasting.