Determination of the potential erosion risk using remote sensing (RS) and geographic information system (GIS) techniques

Thesis Type: Doctorate

Institution Of The Thesis: Uludağ Üniversitesi, Turkey

Approval Date: 2007

Thesis Language: Turkish




Sediment transport from agricultural and forestry lands to Uluabat Lake (Ramsar Site) by Mustafakemalpaşa River is a serious problem within the basin. Predictive erosion models are useful tools for evaluating soil erosion and to establish soil erosion management plans. RUSLE is the commonly used erosion model for this purpose in Turkey as well as in the World. This research integrates the Revised Universal Soil Loss Equation (RUSLE) with a Geographic Information System (GIS) environment to investigate the spatial distribution of annual soil loss potential in Mustafakemalpaşa River Basin. Problems and capabilities in application to GIS and using Remote Sensing (RS) techniques were also discussed. The RUSLE-R factors were developed from local rainfall annual precipitation data using Modified Fournier Index (MFI), topographic (LS) factors were developed from a DEM, soil (K) data were determined from digitized soil maps and land use/cover (C) data were generated from Landsat-7 ETM images. According to the RUSLE/GIS model, the total soil loss potential due to erosion by water of the Mustafakemalpasa River basin is found as 11.296.061,75 tons year-1 that an average soil loss of 11,18 tons ha-1 year-1. According to sub basins it is found 5.656.609,72 tons year-1 for Emet sub basin (average 11,41 tons ha-1 year-1), 5.278.342,50 tons year-1 for Orhaneli sub basin (average 11,26 tons ha-1 year-1) and 361.109,53 tons year-1 (average 7,89 tons ha-1 year-1) for Mkp sub basin. Moreover RUSLE produces only local erosion amount values and cannot be used to estimate the sediment yield for a watershed. For this purpose SDR (sediment delivery ratio) equations were used and compared with the sediment monitoring reports of the Döllük stream gauging station, settled in Mustafakemalpaşa River, within 41 years collected data. While, predicted sediment amount and yield as a function of the sediment delivery ratio and soil loss amount of RUSLE were 1.640.942,7 tons year-1 and 170,1 tons year-1 km-2, these were measured as a 1.082.010 tons year-1 and 127,59 tons year-1 km-2 at the Döllük stream gauging station respectively. Comparative results showed that RUSLE integrated with GIS in the study area found to be effective in generating potential soil erosion risk rates. Furthermore, with the basin wide measured data (such as R factor: from pluviograph reading, C factor: from measured cropping and management values) GIS is powerful, accurately and easy tool mainly economic, time and manpower gaining in modeling soil erosion.