Monika Gähler

Burnt area mapping in the European-Mediterranean: SAR backscatter change analysis and synergistic use of optical and SAR data

Optical satellites are commonly the main data source used for burnt area mapping in a rush mode, but data availability can be limited by weather conditions or direct fire effects such as smoke plumes or haze. Thus, our study focused on compensating these shortcomings through the complementary use of SAR data. In order to investigate the potential of TerraSAR-X (ScanSAR and StripMap) data for burnt area mapping, a multi-temporal backscatter coefficient analysis of four different study sites (La Palma, Portugal, Sardinia and Greece) was performed. The backscatter values in the post-disaster images increased significantly in all study sites, whereas VV polarization showed higher difference values than HH polarization. When comparing backscatter values of individual vegetation types (CORINE land cover classification), the absolute backscatter difference of post- and pre-disaster images varied. High growing and dense vegetation showed a larger backscatter change than low growing and sparse vegetation. To provide information about forest fires also during bad weather conditions, a semi-automatic, object-based, fast and transferable algorithm using very high resolution optical as well as SAR data was developed. The optical part of the algorithm is object-based and uses single- or multi-temporal spectral indices. In order to reduce limitations in optical data due to cloud cover, the algorithm also employs change detection techniques (ratio, difference and index calculation) of TerraSAR-X data.

A multi-scale Markov model for unsupervised oil spill detection in TerraSAR-X data

In this paper an automatic near-real time oil spill detection approach using single-polarized high resolution X-Band Synthetic Aperture Radar satellite data is presented. Dark formations on the water surface are classified in a completely unsupervised way using an automatic tile-based thresholding procedure. The derived global threshold value is used for the initialization of a hybrid multi-contextual Markov image model which integrates scale-dependent and spatial contextual information on irregular hierarchical graph structures into the segment-based labeling process of slick-covered and slick-free water surfaces. Experimental investigations performed on TerraSAR-X ScanSAR data acquired during large-scale oil pollutions in the Gulf of Mexico in May 2010 confirm the effectiveness of the proposed method with respect to accuracy and computational effort.