Mahsa, Almira (2022) Pemodelan Zona Rawan Gerakan Tanah Menggunakan Weight Of Evidence Dan Machine Learning Berbasis GIS di Kabupaten Sigi, Provinsi Sulawesi Tengah. Masters thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 03311950010006-Master_Tesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
03311950010006-Master_Tesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2024. Download (9MB) | Request a copy |
Abstract
Gerakan tanah merupakan salah satu bencana yang sering terjadi di kabupaten Sigi, Sulawesi Tengah. Berdasarkan peta zona kerentanan Gerakan tanah Indonesia, Kabupaten Sigi didominasi dengan zona rawan gerakan tanah tinggi dan zona rawan gerakan tanah menengah. Gerakan tanah sendiri merupakan bencana yang komplek untuk dipelajari dikarenakan umumnya bersifat lokal dengan faktor penyebabnya yang berbeda-beda antara satu daerah dengan daerah lainnya. Salah satu upaya mitigasi yang dapat dilakukan adalah pembuatan peta zona gerakan tanah berbasis lokal. Beberapa parameter penyebab gerakan tanah yang dapat dianalisa diantaranya kerapatan indeks vegetasi, tutupan lahan, geologi regional, jenis tanah, elevasi, arah lereng, kemiringan lereng, kecembungan lereng, jarak dari kelurusan, percepatan tanah maksimum (PGA), dan data historis kejadian gerakan tanah sejumlah 638 data pada periode 2017- 2020. Metode yang akan digunakan untuk mendapatkan peta rawan gerakan tanah pada penelitian ini akan berbasis sistem informasi geografis dan dibantu dengan pendekatan satistik bivariate weight of evidence dan machine learning. Dari data kejadian yang telah didapat nantinya akan dibagi menjadi 70:30 untuk keperluan pelatihan. Metode machine learning memberikan hasil yang lebih baik dalam pembuatan peta rawan gerakan tanah di Kabupaten Sigi, Sulawesi Tengah dengan nilai overall accuracy yang didapat yaitu 0.883, kappa sebesar 0.755 dan AUC 0.833036 sedangkan nilai evaluasi prediksi metode weight of evidence dengan pendekatan AUC sebesar 0.818.
=========================================================================================================
Landslide is one of the disasters that often occurs in Sigi regency, Central Sulawesi. Based on the map of Indonesia's landslide susceptibility zone, Sigi Regency is dominated by a high landslide prone zone and a medium landslide prone zone. Landslide itself is a complex disaster to study because it is generally local in nature with different causes from one area to another. One of the mitigation efforts that can be done is making a local-based landslide susceptibility map. Several parameters causing landslide that can be analyzed include vegetation index density, land cover, regional geology, soil type, elevation, aspect, slope , curvatue, distance from lineament, peak ground acceleration (PGA), and historical data of landslide events. a total of 638 data in the 2017- 2020 period. The method that will be used to obtain a map of the landslide hazard in this study will be based on a geographic information system and assisted by a statistical bivariate weight of evidence and machine learning. From the incident data that has been obtained, it will be divided into 70:30 for training purposes. Machine learning give a more accurate final result than the weight of evidence method in making landslide susceptibility map in Sigi Regency, Central Sulawesi with an overall accuracy value of 0.883, kappa of 0.755 and AUC of 0.833036 while the evaluation value of the weight of evidence method with the AUC approach of 0.818
| Item Type: | Thesis (Masters) |
|---|---|
| Uncontrolled Keywords: | Gerakan Tanah, Machine Learning, Penginderaan Jauh, Sistem Informasi Geografis, Sigi, Sulawesi, Weight of Evidence Geographic Information Systems, Landslide, Machine Learning, Remote Sensing, , Sigi, Sulawesi, Weight of Evidence |
| Subjects: | Q Science > QE Geology Q Science > QE Geology > QE599 Landslides. Rockslides |
| Divisions: | Faculty of Civil, Planning, and Geo Engineering (CIVPLAN) > Geomatics Engineering > 29101-(S2) Master Thesis |
| Depositing User: | Almira Mahsa |
| Date Deposited: | 16 Feb 2022 07:10 |
| Last Modified: | 16 Feb 2022 07:10 |
| URI: | http://repository.its.ac.id/id/eprint/94433 |
Actions (login required)
 |
View Item |