Integration of remotely sensed data and GIS in NRCS model / Nuratifah Zaharudin

• Main Author: Zaharudin, Nuratifah
• Format: Thesis
• Language: English
• Published: 2008
• Subjects: Remote Sensing
• Online Access: http://ir.uitm.edu.my/id/eprint/27276/1/TM_NURATIFAH%20ZAHARUDIN%20AP%2008_5.pdf; http://ir.uitm.edu.my/id/eprint/27276/

Rapid and massive (NRCS) was able to calculate the prediction of runoff. The aim of this study is to introduce the capability of remotely sensed data and Geographic Information System (GIS) for NRCS model in Klang River catchments and to overcome the gap for lack of integrated method between GIS application and hydrological model. The first objective of this study is to prepare the landuse information by following the Hydrological Soils Group (HSG) table within catchments’ boundary from satellite imagery as main source. Second, this study aims to combine landuse information and soil classifications to form HSG table through an identified Curve Number (CN) for each intersection in ArcView software. Finally, this study aims to generate runoff value and compute validation with observed data. This study uses Landsat Thematic Mapper (TM) that is able to oversee big areas and could lessen the cost for obtaining landuse information. The research involves three stages of analysis. First, remote sensing technique was used to extract preliminary data. Second, two phases of GIS was involved which are accumulating and programming phase. The third stage that is involved in the research methodology is the validation stage of using SNT model in GIS. In remote sensing technique, 84% of accuracy assessment was obtained with 0.8 kappa value using data processed in 2001 by Landsat TM images. With that achievement, landuse map that contained waterbody, urban, forest, open area and disturbed/transitional area from Landsat TM that represent Klang River catchment was derived. GIS has successfully developed a simulation of NRCS model and graph. Validation stage was done by comparing the simulation of NRCS model from GIS with Observed Unit Hydrograph (OUH). Through validation stage, the SNT model only has an acceptable value of 6% error from the application of 2001 data compared to 12% error from application of 1991 data. Findings from this study have shown that satellite imagery such as Landsat TM are able to provide essential information for hydrology study. The simulation of NRCS model is suitable for monitoring runoff peak of catchment areas.