DEM Generation Watershed Delineation

10 The dark object method was carried out upon Landsat images 2006 which is assigned as referenced data of normalization. Even the image was suffered from striping caused by Scan Line Corrector SLC failure, the radiometric quality remains intact USGS, 2006. The complete process of atmospheric correction can be seen in the figure below. Figure 3.3. The Process Flow of Atmospheric Correction

D. DEM Generation

DEM Digital Elevation Model is one of raster data used in this study which generated from contour data. Contour data was came from topographic map Rupa Bumi Indonesia map, sheet no. 1209-142 and 1209-231, with contour interval 12.5 m. DEM was generated through surfacing process by using non-linear rubber sheeting method. In order to have same in resolution of data, DEM produced have a resolution as same as Atmospheric Correction: dark object method Data Preprocessing Landsat 12265_ 2006 Landsat 12265_ 1991 12265_ 1994 12265_ 1997 12265_ 2001 12265_ 2004 RC Landsat 12265_ 2006 PIS Selection  STD of PIS of reference image σ Rk  STD of PIS of subject image σ Sk,y  Mean of PIS of reference Image R k  Mean of PIS of subject image S k,y Note: PIS = Pseudo-invariant Set STD = Standard Deviation RC = Radiometric Corrected Transformation  Transformation Coefficient slope: m k, y = σ Rk σ Sk,y intercept: b k, y = R k - m k, y . S k,y  Image Regression Model Y y = m k, y . X k + b k, y Reference Image Subject Images X k Landsat 12265_ 1991 12265_ 1994 12265_ 1997 12265_ 2001 12265_ 2004 Normalized ImagesY y 11 LANDSAT data, that is 30 m x 30 m. In further step of analysis in this study, the DEM was converted to a slope map and then used as a one factor for estimate soil erosion. III.6. D ATA PROCESSING There are three kinds of data processing are carried out to prepare the data for the main analysis in the study that is soil erosion analysis, comprise of image classification, watershed delineation, and slope classes and length generation.

A. Watershed Delineation

The watershed delineation tool uses AVSWAT-2000 version 1.0, the extension for ArcView and Spatial Analysis. It is used to perform watershed delineation based on SWAT Soil and Water Assessment Tool model. SWAT is a model developed to predict the impact of land management practices on water, sediment, and agricultural chemical yields in large, complex watershed with varying soils, land use, and management conditions over long period of time Arnold et al., 1998. The delineation process requires a Digital Elevation Model DEM in ArcInfo grid format. The diagram flow below illustrates the steps in the delineation process: Figure 3.4. Delineation process Topographic Data Digital Elevation Model in Erdas format .img Digital Elevation Model In ArcInfo grid format .grid Surfacing Export .img to .grid Process in ERDAS Outputs . shp file: 1. Sub-watersheds area 2. The stream network  Load the DEM  Preprocess the DEM  Specify the minimum sub-watershed area catchments area  Review and edit the stream network points  Delineation Process Process in ArcView AVSWAT Ext. Data Processing Data Preprocessing 12 Figure 3.5 shows the catchments area which generated from Digital Elevation Model DEM. Delineating process obtained three large watersheds, which overlapped with Telaga Warna Nature Reserve, as follows: Ciliwung, Cigundul and Cibeet watersheds. However, the area study is limited on the catchments area part of watershed that has an upper stream region on nature reserve area. Figure 3.5.b. shows that area study has eight catchments that have upper stream region on nature reserve area. Figure 3.5. Delineating catchments area

B. Slope Classes and Length Generation