1 st National Salinity Engineering Conference 9–12 November 2004 Perth, Western Australia The extent and potential area of salt-affected land in Western Australia estimated using remote sensing and digital terrain models D.J. McFarlane Department of Environment, 3 Plain St, East Perth 6004 Email: don.mcfarlane@environment.wa.gov.au R.J. George Western Australian Department of Agriculture, Department of Agriculture, PO Box 1231, Bunbury, 6231 P.A. Caccetta CSIRO Floreat, Private Bag 5, PO Wembley 6913 Abstract: A number of ways of mapping, monitoring and predicting salt-affected land have been used, all of which have strengths and weaknesses. In WA, methods that have been used include ABS surveys, extrapolating from detailed catchment mapping, groundwater levels coupled with soil-landform maps (NLWRA methodology) and satellite remote sensing and digital terrain models (“Land Monitor” – this paper). The ABS method may underestimate the extent of salinity whereas catchment mapping and the NLWRA method overestimates its extent and hazard. The Land Monitor method estimates that about 960,000 hectares were affected by secondary salinity in 1996 and this area had increased by about 14,000 ha per annum since 1989. The area with a salinity hazard (ie may develop salinity in future depending upon controlling factors) is estimated to be up to 5.4 million hectares of total land (between 2.8 and 4.4 million hectares of agricultural land). The results indicate that there may be more time to develop innovative plant and engineering solutions than was previously thought. The Land Monitor method has advantages in that it shows where salinity is within catchments and hazard areas are identified more accurately than in any other method. In addition, the method allows confidence limits to be placed on the estimates. Keywords: Salinity extent, salinity prediction, remote sensing, satellite, digital terrain model. 1. INTRODUCTION The current and predicted future extent of salinity is useful to help decide how much resources agencies and landowners should commit to saving assets that may be at risk, and when they should commit these resources. Some landholders have stated that they would have taken the problem more seriously had they known what affect it may have on their land and water assets. However salinity is often insidious with small changes occurring each year as saline groundwater is slowly drawn into the root zone of plants with larger changes occurring after wet years. Likewise, government needs to know which public assets may be at risk as there is usually a long lead time before vegetative solutions can lower groundwater levels and engineering solutions need to be budgeted well in advance. Ferdowsian et al. [1996] reviewed the methods that had been used in Western Australia to estimate the extent of salinity and its likely long term impact at equilibrium. Since that time two other methods have been further developed – the National Land and Water Resource Audit (NLWRA) method and Land Monitor. This paper briefly reviews each method before providing the first consolidated results from the Land Monitor method. 2. REVIEW OF PREVIOUS METHODS 2.1 Australian Bureau of Statistics (ABS) Four surveys were carried out of West Australian farmers in the 1974 – 1993 period asking whether land that had been previously productive had become saline (a definition of secondary salinity). When plotted, these data show a straight line increase with 529,000 ha (or 3.1% of cleared land)