Geomechanics and Engineering

Techno-Press (테크노프레스)
권호 표지
DOI QR코드

DOI QR Code

Prediction of karst sinkhole collapse using a decision-tree (DT) classifier

  • Boo Hyun Nam (Department of Civil Engineering, College of Engineering, Kyung Hee University) ;
  • Kyungwon Park (Department of Civil Engineering, College of Engineering, Kyung Hee University) ;
  • Yong Je Kim (Department of Civil and Environmental Engineering, Lamar University)
  • Received : 2023.10.18
  • Accepted : 2024.02.11
  • Published : 2024.03.10
⟨ Previous Next ⟩

Abstract

Sinkhole subsidence and collapse is a common geohazard often formed in karst areas such as the state of Florida, United States of America. To predict the sinkhole occurrence, we need to understand the formation mechanism of sinkhole and its karst hydrogeology. For this purpose, investigating the factors affecting sinkholes is an essential and important step. The main objectives of the presenting study are (1) the development of a machine learning (ML)-based model, namely C5.0 decision tree (C5.0 DT), for the prediction of sinkhole susceptibility, which accounts for sinkhole/subsidence inventory and sinkhole contributing factors (e.g., geological/hydrogeological) and (2) the construction of a regional-scale sinkhole susceptibility map. The study area is east central Florida (ECF) where a cover-collapse type is commonly reported. The C5.0 DT algorithm was used to account for twelve (12) identified hydrogeological factors. In this study, a total of 1,113 sinkholes in ECF were identified and the dataset was then randomly divided into 70% and 30% subsets for training and testing, respectively. The performance of the sinkhole susceptibility model was evaluated using a receiver operating characteristic (ROC) curve, particularly the area under the curve (AUC). The C5.0 model showed a high prediction accuracy of 83.52%. It is concluded that a decision tree is a promising tool and classifier for spatial prediction of karst sinkholes and subsidence in the ECF area.

Keywords

Acknowledgement

The authors thank the Florida Geological Survey (FGS) for sharing the data (sinkhole survey and geological data).

References

  1. Beck, B.F. and Sinclair, W.C. (1986), Sinkholes in Florida: An Introduction, Florida Sinkhole Research Institute, Orlando, FL.
  2. Ciotoli, G., Di Loreto, E., Finoia, M.G., Liperi, L., Meloni, F., Nisio, S. and Sericola, A. (2016), "Sinkhole susceptibility, Lazio Region, central Italy", J. Maps., 12(2), 287-294. https://doi.org/10.1080/17445647.2015.1014939.
  3. Ding, H., Wu, Q., Zhao, D., Mu, W. and Yu, S. (2019), "Risk assessment of karst collapse using an integrated fuzzy analytic hierarchy process and grey relational analysis model", Geomech. Eng., 18(5), 515-525. https://doi.org/10.12989/gae.2019.18.5.515.
  4. FDEP (2022), Florida Subsidence Incident Reports, FDEP, Florida Geological Survey.
  5. Florida Office of Insurance Regulation (2010), Report on Review of the 2010 Sinkhole Data Call, Florida Office of Insurance Regulation (FOIR).
  6. Galve, J.P., Gutierrez, F., Remondo, J., Bonachea, J., Lucha, P. and Cendrero, A. (2009), "Evaluating and comparing methods of sinkhole susceptibility mapping in the Ebro Valley evaporite karst (NE Spain)", Geomorphology, 111(3), 160-172. https://doi.org/10.1016/j.geomorph.2009.04.017.
  7. Genis, M., Akcin, H., Aydan, O. and Bacak, G. (2018), "Investigation of possible causes of sinkhole incident at the Zonguldak Coal Basin, Turkey", Geomech. Eng., 16(2), 177-185. https://doi.org/10.12989/gae.2018.16.2.177.
  8. Gkioulekas, I. and Papageorgiou, L.G. (2021), "Tree regression models using statistical testing and mixed integer programming", Comput. Ind. Eng., 153, 107059. https://doi.org/10.1016/j.cie.2020.107059.
  9. Guo, Z., Shi, Y., Huang, F., Fan, X. and Huang, J. (2021), "Landslide susceptibility zonation method based on C5.0 decision tree and K-means cluster algorithms to improve the efficiency of risk management", Geosci. Front., 12(6), 101249. https://doi.org/10.1016/j.gsf.2021.101249.
  10. Kim, J.M., Lee, S., Park, J.Y., Kihm, J.H. and Park, S. (2020), "A set of failure variables for analyzing stability of slopes and tunnels", Geomech. Eng., 20(3), 175-189. https://doi.org/10.12989/gae.2020.20.3.175.
  11. Kim, Y.J., Nam, B.H., Jung, Y.H., Liu, X., Choi, S., Kim, D. and Kim, S. (2022), "Probabilistic spatial susceptibility modeling of carbonate karst sinkhole", Eng. Geol., 306, 106728. https://doi.org/10.1016/j.enggeo.2022.106728.
  12. Kim, Y.J., Nam, B.H., Shamet, R., Soliman, M. and Youn, H. (2020), "Development of sinkhole susceptibility map of east central Florida", Nat. Hazard. Review, 21(4), 04020035. doi:10.1061/(ASCE)NH.1527-6996.0000404.
  13. Li, C., Zou, J.F. and Sheng, Y.M. (2020), "Undrained solution for cavity expansion in strength degradation and tresca soils", Geomech. Eng., 21(6), 527-536. https://doi.org/10.12989/gae.2020.21.6.527.
  14. Liu, L.L., Yang, C. and Wang, X.M. (2021), "Landslide susceptibility assessment using feature selection-based machine learning models", Geomech. Eng., 25(1), 1-16. https://doi.org/10.12989/gae.2019.18.5.515.
  15. Nanehkaran, Y.A., Mao, Y., Azarafza, M., Kockar, M.K. and Zhu, H.H. (2021), "Fuzzy-based multiple decision method for landslide susceptibility and hazard assessment: A case study of Tabriz, Iran", Geomech. Eng., 24(5), 407-418. https://doi.org/10.12989/gae.2021.24.5.407.
  16. Naithani, A.K., Jain, P., Singh, L.G. and Rawat, D.S. (2022), "Engineering geological characteristics of the underground surge pool cavern: a case study", Int. J. Geo-Eng., 13(7). https://doi.org/10.1186/s40703-022-00172-9
  17. Ozdemir, A. (2015), "Sinkhole Susceptibility Mapping Using a Frequency Ratio Method and GIS Technology Near Karapinar, Konya-Turkey", Procedia Earth Planetary Sci., 15, 502-506. https://doi.org/10.1016/j.proeps.2015.08.059.
  18. Papadopoulou-Vrynioti, K., Bathrellos, G.D., Skilodimou, H.D., Kaviris, G. and Makropoulos, K. (2013), "Karst collapse susceptibility mapping considering peak ground acceleration in a rapidly growing urban area", Eng. Geol., 158, 77-88. https://doi.org/10.1016/j.enggeo.2013.02.009.
  19. Safavian, S.R. and Landgrebe, D. (1991), "A survey of decision tree classifier methodology", IEEE T. Syst. Man Cy., 21(3), 660-674. https://doi.org/10.1109/21.97458.
  20. Shamet, R., Nam, B.H. and Horhota, D. (2018), "Development of a sinkhole raveling chart based on Cone Penetration Test (CPT) data", Proceedings of the 15th Multidisciplinary Conference on Sinkholes and the Engineering and Environmental Impacts of Karst and the 3rd Appalachian Karst Symposium, Shepherdstown, WV, April 2-6.
  21. Sheng, M., Zhou, J., Chen, X., Teng, Y., Hong, A. and Liu, G. (2022), "Landslide susceptibility prediction based on frequency ratio method and C5.0 decision tree model", Front. Earth Sci., 10. https://doi.org/10.3389/feart.2022.918386.
  22. Soliman, M.H., Shamet, R., Kim, Y.J., Youn, H. and Nam, B.H. (2019), "Numerical Investigation on the Mechanical Behavior of Karst Sinkholes", Environ. Geotech., 8(6), 367-381. https://doi.org/10.1680/jenge.18.00063.
  23. Strzalkowski, P. (2018), "Sinkhole formation hazard assessment", Environ. Earth Sci., 78(1), 9. https://doi.org/10.1007/s12665-018-8002-5.
  24. Subedi, P., Subedi, K., Thapa, B. and Subedi, P. (2019), "Sinkhole susceptibility mapping in Marion County, Florida: Evaluation and comparison between analytical hierarchy process and logistic regression based approaches", Scientific Reports, 9(1), 7140. https://doi.org/10.1038/s41598-019-43705-6.
  25. Taheri, K., Gutierrez, F., Mohseni, H., Raeisi, E. and Taheri, M. (2015), "Sinkhole susceptibility mapping using the analytical hierarchy process (AHP) and magnitude-frequency relationships: A case study in Hamadan province, Iran", Geomorphology, 234, 64-79. https://doi.org/10.1016/j.geomorph.2015.01.005.
  26. Tacim, G., Posluk, E. and Gokceoglu, C. (2023), "Importance of grouting for tunneling in karstic and complex environment (a case study from Turkiye)", Int. J. Geo-Eng., 14(6). https://doi.org/10.1186/s40703-023-00183-0
  27. Tihansky, A.B. (1999), "Sinkholes, west-central Florida", Land subsidence in the United States: US geological survey circular. 1182 121-140.
  28. Weary, D.J. (2015), "The cost of Karst subsidence and sinkhole collapse in the united states compared with other natural hazards", Proceedings of the 14th Multidisciplinary Conference on Sinkholes and the Engineering and Environmental Impacts of Karst, Rochester, MN, October 5-9.
  29. Weary, D.J. and Doctor, D.H. (2014), Karst in the United States: A digital map compilation and database, U.S. Geological Survey, Open-File Report 2014-1156.
  30. Xu, Z., Xian, M., Li, X., Zhou, W., Wang, J., Wang, Y. and Chai, J. (2021), "Risk assessment of water inrush in karst shallow tunnel with stable surface water supply: Case study", Geomech. Eng., 25(6), 495-508. https://doi.org/10.12989/gae.2021.25.6.495.
  31. Xu, Z., Chengping, Z., Bo, M. and Youjun, X. (2020), "Experimental study on the mechanical response and failure behavior of double-arch tunnels with cavities behind the liner", Geomech. Eng., 20(5), 399-410. https://doi.org/10.12989/gae.2020.20.5.399
  32. Yilmaz, I. (2007), "GIS based susceptibility mapping of karst depression in gypsum: A case study from Sivas basin (Turkey)", Eng. Geol., 90(1), 89-103. https://doi.org/10.1016/j.enggeo.2006.12.004.
  33. Yilmaz, I., Marschalko, M. and Bednarik, M. (2013), "An assessment on the use of bivariate, multivariate and soft computing techniques for collapse susceptibility in GIS environ", J. Earth Syst. Sci., 122(2), 371-388. https://doi.org/10.1007/s12040-013-0281-3.
  34. Yuan, Y.C., Li, S.C., Zhang, Q.Q., Li, L.P., Shi, S.S. and Zhou, Z.Q. (2016), "Risk assessment of water inrush in karst tunnels based on a modified grey evaluation model: Sample as Shangjiawan Tunnel", Geomech. Eng., 11(4), 493-513. https://doi.org/10.12989/gae.2016.11.4.493.
  35. Zhou, Z.Q., Li, S.C., Li, L.P., Shi, S.S. and Xu, Z.H. (2015), "An optimal classification method for risk assessment of water inrush in karst tunnels based on the grey system", Geomech. Eng., 8(5), 631-647. https://doi.org/10.12989/gae.2015.8.5.631.