1 International Journal of Geoinformatics, Vol. 1, No. 1, March 2005 Mohammad Firuz Ramli 1 , Mohd Kamil Yusoff 1 , Saari Mustapha 2 and Choo Siet Mei 1 1 Faculty of Environmental Studies, University Putra Malaysia, 43400 Serdang, Selangor, Malaysia Tel: + 603-89466753 Fax: + 603-89438109 E-mail: firuzramli@yahoo.co.uk 2 Faculty of Engineering, University Putra Malaysia, 43400 UPM Serdang, Selangor, Malaysia Abstract The Revised Universal Soil Loss Equation (RUSLE) was used to predict the soil erosion hazard for Langkawi Island, Malaysia. The RUSLE was applied within the Geographical Resources Analysis Support System (GRASS), the open source Geographic Information System (GIS). Secondary raster data affecting erosion namely: rainfall erosivity; soil erodibility; cover management; conservation practices; and slope length and steepness were used to generate soil erosion map. The results predicted that about 58% of the Langkawi are subjected to very low to moderate erosion potential and another 42% to the high to extreme erosion potential. Although further studies are needed, it is clearly demonstrated the effectiveness of open source GIS in generating quantitative information on soil erosion studies. Internation Journal of Geoinformatics, Vol.1, No. 1, March 2005 Published by AAG, P.O. Box 44, Klong Luang, Pathumthani 12120, Thailand. Erosion Potential Assessment using Open Source Geographical Resources Analysis Support System (GRASS) for Langkawi Island, Malaysia Technical Letter 1. Introduction Rapid development in Langkawi, Malaysia leads to rapid transformation of vast areas of originally forested area to agricultural and urban areas. If the activities are not properly controlled, these will lead to soil erosion which will eventually result in land degradation and disaster. Thus, it is essential to identify the potential area of soil erosion on a map and subsequently develop a tool for a planning of soil conservation. A Geo- graphic Information System (GIS) can be used to identify areas that potentially vulnerable to soil erosion, and then use or apply the information to estimate losses at various locations (Kertész, 1993; Mellerowicz et al., 1994; Cox and Madra- mootoo, 1998; Shi et al., 2004). Within a raster based Geographical Resources Analysis Sup- port System (GRASS), the Revised Universal Soil Loss Equation (RUSLE) model may be applied for the erosion potential prediction (Shi et al., 2004). GRASS is an open source raster/ vector GIS combined with integrated image processing and data visualisation subsystems. It includes more than 350 modules for manage- ment, processing, analysis and visualization of geo-referenced data (Neteler and Mitasova, 2002). Since it is open source software then it can be downloaded freely from the website, thus it will help researchers with budget constraint. Furthermore, it runs on the Linux platform which is also an open source software. This is another option for those who intend to use GIS software for erosion study. RUSLE is an erosion prediction and con- servation planning tool based on the Universal 017R_FIR 01/03/2005, 18:05 1