| Popis: |
Bawakaraeng and Lompobattang mountains are located at a southern part of South Sulawesi province and surrounded by the districts with high growth rates and had an important role in supporting that growth. In addition to providing a fertile land, this area is also threatened by disasters, particularly landslides. Landslide disasters occur almost every year especially during the rainy season and landslide induced flash floods or debris flow occurred in the upstream. Hence, the information of landslides susceptibility will greatly assist planners to optimize the future development planning. This study aims to predict landslide susceptibility using a statistical approach to find high accuracy. In this study 8 parameters were usedthe landslide conditioning factors namely, lithology, distance from the road, distance from the river, distance from the fault, land use, curvature, aspect and slope. The scales of data used were above 1: 50,000. The study area was chosen in a mountainous watershed where elevation is above 500 m and landslide occurred in high. All data were converted to raster forms with pixel size of 30 m. This research used ARCGIS 10.0 to process and analyzes spatial data. Besides,Microsoft Exeland SPSS were usedfor statistical data processing. This study divided landslide data as training and validation.The relative operating characteristic curve (ROC) and area under ROC curve (AUC) were usedto validate the performance of logistic regression in predicting future landslides.The best model was selected among twelve trials that were chosen from equal number of landslide and non-landslide pixels. The success rate, determined from the AUC of training data set, was found to be 0.866, which means that model has accuracy of 86.6 % accuracy to predict future landslides. The prediction rate, calculated from the AUC of the validation dataset was found0.855, which means a prediction accuracy of 85.5%. The close similarity of the success rate and prediction rate values showed how the logistic regression model is reliable in predicting future landslide with a good level of accuracy. |
