We report a collaborative research program that is developing procedures for use of MODIS imagery for monitoring soil erosion risk across the cropping agricultural lands in South Australia. The image-based monitoring aims to enhance the current field-based assessment of land condition conducted by government agencies, and provide objective and comprehensive reporting for State erosion protection targets. Most recently, we have developed a new Land Condition Image Index for quantifying vegetation and soil exposure, used it to document seasonal spatial variation in cover across SA cropping lands, and explored its relationship to cover levels recorded during roadside transects as part of the existing land condition monitoring program.
The paper reviews issues associated with development, introduction and acceptance of the image-based methodology: understanding the main contributors to seasonal and annual variations in land cover, identifying sources of error in empirical relationships used to test image indices, and changing institutional understanding of the nature and potential of remotely-sensed data for land condition assessment and monitoring.