Pemetaan Potensi Air Tanah Menggunakan Metode Random Forest di Kabupaten Kediri

Nugroho, Wisdom Hidayat Agung (2024) Pemetaan Potensi Air Tanah Menggunakan Metode Random Forest di Kabupaten Kediri. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5016201065-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2026. Download (12MB) | Request a copy

Abstract

Ketersediaan air tanah merupakan salah satu solusi dalam memastikan keberlanjutan sumber daya air termasuk ketersediaan air bersih. Agar proses pemanfaatan air tanah menjadi lebih efektif dan juga untuk mendukung salah satu goals pada Sustainable Development Goals (SDGs) yaitu air bersih dan sanitasi layak, diperlukan pemetaan potensi air tanah sehingga sehingga sumber daya air dapat dikelola secara optimal. Sehubungan dengan itu, penelitian tugas akhir ini bertujuan untuk mengetahui sebaran potensi air tanah di Kabupaten Kediri menggunakan metode machine learning dengan algoritma Random Forest (RF). Penelitian ini menggunakan 18 parameter yang meliputi elevasi, slope, aspect, curvature, drainage density, kerapatan sungai, jarak dari sungai, lineament density, TWI (Topographic Wetness Index), NDVI (Normalized Difference Vegetation Index), tutupan lahan, jenis tanah, jenis batuan, curah hujan dan ditambah dengan kanal 2, kanal 3, kanal 4, serta kanal 8 dari citra satelit Sentinel-2A. Koordinat lokasi sumur bor air tanah digunakan sebagai data training dan testing dengan perbandingan 80:20, 70:30, dan 60:40. Melalui evaluasi performa masing-masing model dengan confusion matrix, diketahui model terbaik pada penelitian ini adalah model rasio 70:30 dengan nilai Akurasi (Acc), Sensitivitas (Sen), Spesifisitas (Spe), Positive Predictive Value (PPV) senilai 0,978, Matthew’s Correlation Coefficient (MCC) dan Cohen’s Kappa (CK) sebesar 0,956, serta Area Under Curve (AUC) senilai 0,994. Pada model ini, parameter elevasi memiliki pengaruh yang paling tinggi terhadap model dengan nilai IDI (Importance Degree Index) senilai 100. Berdasarkan model tersebut, Kecamatan Kras, Landat, Ngadiluwih, Wates, Gurah, Gampengrejo, Pagu, Pare, Pagu, Papar, Plemahan, Kunjang, dan Purwoasri memiliki tingkat potensi air tanah yang tinggi.
=================================================================================================================================
The availability of groundwater is one of the solutions to ensure the sustainability of water resources, including the availability of clean water. To enhance the effectiveness of groundwater utilization and to support one of the Sustainable Development Goals (SDGs), namely, clean water and sanitation, it is necessary to map groundwater potential so that water resources can be managed optimally. This ensures that water resources can be managed optimally. Therefore, this final project aims to determine the distribution of groundwater potential using machine learning methods with the Random Forest (RF) algorithm. This research utilizes 18 parameters, including elevation, slope, aspect, curvature, drainage density, river density, distance from rivers, lineament density, TWI (Topographic Wetness Index), NDVI (Normalized Difference Vegetation Index), land cover, soil type, geology, rainfall, and in addition, band 2, band 3, band 4, and band 8 from Sentinel-2A satellite imagery. The coordinates of groundwater well locations are used as training and testing data with ratios of 80:20, 70:30, and 60:40. Through performance evaluation of each model using the confusion matrix, it is known that the best model in this study is the 70:30 ratio model with Accuracy (Acc), Sensitivity (Sen), Specificity (Spe), Positive Predictive Value (PPV) values of 0.978, Matthew’s Correlation Coefficient (MCC) and Cohen’s Kappa (CK) values of 0.933, and Area Under Curve (AUC) values of 0.993. In this model, drainage density has the highest Importance Degree Index (IDI) value (100). Based on this model, the districts of Kras, Landat, Ngadiluwih, Wates, Gurah, Gampengrejo, Pagu, Pare, Pagu, Papar, Plemahan, Kunjang, and Purwoasri have high groundwater potential levels.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Clean Water and Sanitation, Groundwater, Machine Learning, Random Forest, Air Bersih dan Sanitasi Layak, Air Tanah, Machine Learning, Random Forest
Subjects:	G Geography. Anthropology. Recreation > G Geography (General) > G70.212 ArcGIS. Geographic information systems.
Divisions:	Faculty of Civil, Planning, and Geo Engineering (CIVPLAN) > Geomatics Engineering > 29202-(S1) Undergraduate Thesis
Depositing User:	Wisdom Hidayat Agung Nugroho
Date Deposited:	17 Jul 2024 05:05
Last Modified:	17 Jul 2024 05:05
URI:	http://repository.its.ac.id/id/eprint/108393

Actions (login required)

View Item