Extracting gully features and their dynamics from high spatial resolution imagery using object based image analysis

Gully erosion is responsible for a substantial amount of soil loss and is generally considered an indicator of desertification. Hence, mapping these features provides essential information needed on sediment production, identification of vulnerable areas for gully formation, land degradation and environmental effects. This paper investigates the use of object-oriented image analysis to extract gully features from imagery, using a combination of topographic, spectral, shape/geometric and contextual information obtained from Ikonos-2 and GeoEye-1 data. A rule-set was developed and tested for a semi-arid to sub-humid region in Morocco. The accuracy of the feature extraction based on percentage of gully system boundary (GSB) was assessed for three sub-watersheds (SW1, SW2, and SW3). Changes in the GSB, in three different sub-watersheds, ranged from moderate (11% in SW2 and 21% in SW1) to a very high increase (91% in SW3). The percentage of GSB indicated negligible overestimations between the reference area and OOA area in SW1 (4%) and negligible underestimations in SW3 (-3%). However, the percentage of GSB in SW2 (24%) was overestimated due to the difference in visual abilities of a human operator digitizing highly complex gully system with fuzzy boundaries. In particular finer edges within the complex gully systems were better identified semi-automatically than was possible by manual digitization, suggesting higher detection accuracy. OOA-based gully mapping is quicker and more objective than traditional methods, and is thus better suited to provide essential information for land managers to support their decision making processes, and for the erosion research community.