Satellite remote sensing for detailed landslide inventories using change detection and image fusion

The availability of high spatial and spectral resolution remote sensing systems may be accompanied by changes in techniques for applying the data if appropriate data processing methodologies can be demonstrated. Landslide monitoring, which requires large areas to be surveyed at a detailed level, has previously been unsatisfactory due to its reliance on air photograph interpretation. This study demonstrates the synergistic use of medium resolution, multitemporal Satellite pour l'Observation de la Terre (SPOT) XS, and fine resolution IKONOS images for landslide inventories. The post-classification comparison method of change detection using the Maximum Likelihood classifier with SPOT XS images was able to detect approximately 70% of landslides, the main omissions being those smaller than approximately half a pixel wide. The visual quality of images obtained from Pan-sharpening of IKONOS images was comparable to that obtainable from 1: 10000 scale air photographs, enabling detailed interpretation of landslides and associated environmental features. A methodology combining the two levels of survey is proposed for regional scale landslide monitoring.