Explainable Sinkhole Susceptibility Mapping Using Machine-Learning-Based SHAP: Quantifying and Comparing the Effects of Contributing Factors in Konya, Türkiye
Top Cited Papers
Open Access
- 13 March 2025
- journal article
- research article
- Published by MDPI AG in Applied Sciences
- Vol. 15 (6) , 3139
Abstract
Sinkholes, naturally occurring formations in karst regions, represent a significant environmental hazard, threatening infrastructure, agricultural lands, and human safety. In recent years, machine learning (ML) techniques have been extensively employed for sinkhole susceptibility mapping (SSM). However, the lack of explainability inherent in these methods remains a critical issue for decision-makers. In this study, sinkhole susceptibility in the Konya Closed Basin was mapped using an interpretable machine learning model based on SHapley Additive exPlanations (SHAP). The Random Forest (RF), eXtreme Gradient Boosting (XGBoost), and Light Gradient Boosting Machine (LightGBM) algorithms were employed, and the interpretability of the model results was enhanced through SHAP analysis. Among the compared models, the RF model demonstrated the highest performance, achieving an accuracy of 95.5% and an AUC score of 98.8%, and was consequently selected for the development of the final susceptibility map. SHAP analyses revealed that factors such as proximity to fault lines, mean annual precipitation, and bicarbonate concentration difference are the most significant variables influencing sinkhole formation. Additionally, specific threshold values were quantified, and the critical effects of these contributing factors were analyzed in detail. This study underscores the importance of employing eXplainable Artificial Intelligence (XAI) techniques in natural hazard modeling, using SSM as an example, thereby providing decision-makers with a more reliable and comparable risk assessment.Keywords
Funding Information
- the Konya Provincial Directorate of Turkish Disaster and Emergency Management Presidency (AFAD) within the framework of the “Determination of Sinkhole Locations Project” (2020K-138637)
This publication has 103 references indexed in Scilit:
- Karst Hydrogeology, Geomorphology and Caves2022
- An Introduction to Statistical Learning2021
- Sinkhole susceptibility mapping: A comparison between Bayes‐based machine learning algorithmsLand Degradation & Development, 2019
- Sinkholes Susceptibility Assessment in Urban Environment Using Heuristic, Statistical and Artificial Neural Network (ANN) Models in Evaporite Karst System: A Case Study from Lesina Marina (Southern Italy)2014
- The Elements of Statistical Learning2009
- Karst Hydrogeology and Geomorphology2007
- Karst unit mapping using geographic information system technology, Mower County, Minnesota, USAEnvironmental Geology, 2002
- Ensemble Methods in Machine Learning2000
- Quantitative analysis of land surface topographyEarth Surface Processes and Landforms, 1987
- Regression Diagnostics1980