Bulletin of Electrical Engineering and Informatics | 2021
Computer Model for Tsunami Vulnerability Using Sentinel 2A and Shuttle Radar Tomography Mission Remote Sensing Imagery Optimized by Machine Learning
Abstract
This study aims to develop a software framework for the identification of tsunami high vulnerability areas using the DEM (Digital Elevation Model), LULC (Land Use Land Cover), and VI (Vegetation Index) indicators as part of the tsunami mitigation. This study was carried out in five stages, namely: (1) preprocessing data that consists of a collection of Sentinel-2 satellite image data and identification of the research area i.e. the area of Kebumen Regency, Central Java Province, Indonesia which covers 8 districts, namely Ayah, Buayan, Puring, Petanahan, Klirong, Buluspesantren, Ambal, and Mirit Districts. The images were corrected geometrically, radiometrically, and atmospherically; (2) Data Analysis and Classification which is the activities of Sentinel-2 image data extraction using the NDVI (Normalized Difference Vegetation Index), NDBI (Normalized Difference Built-up Index), NDWI (Normalized Difference Water Index), MSAVI (Modified Soil Adjusted Vegetation Index), and MNDWI (Modified Normalized Difference Water Index) algorithms; (3) Prediction data that was performed using the NDVI, NDBI, NDWI, MSAVI, and MNDWI algorithms, and extracted from Sentinel-2 images using ML (Machine Learning) of Random Forest (RF), Multivariate Adaptive Regression Spline (MARS), and Classification and Regression Tree (CART); (4) Accuracy testing of prediction results with ML, which was performed statistically using the MSE, ME, RMSE, MAE, MPE, and MAPE equations; and (5) Spatial vulnerability prediction which was performed using Ordinary Kriging spatial interpolation. The results show that in 2021 the area was dominated by vegetation density between (-0.1) to (0.3) with moderate to high vulnerability and risk of LULC tsunami as a result of the decreasing of vegetation area. The prediction results for 2021 show a low canopy density of vegetation and a high degree of land surface slope. Based on the prediction results in 2021, the study area mostly shows the existence of built-up lands with a high tsunami vulnerability risk (> 0.1). Vegetation in the study area had decreased to 67% from the original areas in 2017 with an area of 135 km 2 . Forest vegetation had decreased by 45% from 116 km 2 in 2017. Land use for fisheries had increased to the area of 86 km 2 from 2017 with an area of 24 km 2 .