Application of an adaptive neuro-fuzzy inference system to ground subsidence hazard mapping

  • Authors:

  • Inhye Park;Jaewon Choi;Moung Jin Lee;Saro Lee

  • Affiliations:

  • Department of Geoinformatics, University of Seoul, Siripdae-gil 13, Dongdaemun-gu, Seoul 130-743, Republic of Korea;Geospatial Analysis & Evaluation Center, National Disaster Management Institute;Department of Earth System Sciences, Yonsei University, 134 Shinchon-dong, Seoul, Republic of Korea and Korea Environment Institute, 290 Jinheungno, Eunpyeong-Gu, Seoul 122-706, Republic of Korea;Geological Mapping Department, Korea Institute of Geoscience & Mineral Resources (KIGAM), 92 Gwahang-no, Yuseong-gu, Daejeon 305-350, Republic of Korea

  • Venue:
  • Computers & Geosciences
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

We constructed hazard maps of ground subsidence around abandoned underground coal mines (AUCMs) in Samcheok City, Korea, using an adaptive neuro-fuzzy inference system (ANFIS) and a geographical information system (GIS). To evaluate the factors related to ground subsidence, a spatial database was constructed from topographic, geologic, mine tunnel, land use, and ground subsidence maps. An attribute database was also constructed from field investigations and reports on existing ground subsidence areas at the study site. Five major factors causing ground subsidence were extracted: (1) depth of drift; (2) distance from drift; (3) slope gradient; (4) geology; and (5) land use. The adaptive ANFIS model with different types of membership functions (MFs) was then applied for ground subsidence hazard mapping in the study area. Two ground subsidence hazard maps were prepared using the different MFs. Finally, the resulting ground subsidence hazard maps were validated using the ground subsidence test data which were not used for training the ANFIS. The validation results showed 95.12% accuracy using the generalized bell-shaped MF model and 94.94% accuracy using the Sigmoidal2 MF model. These accuracy results show that an ANFIS can be an effective tool in ground subsidence hazard mapping. Analysis of ground subsidence with the ANFIS model suggests that quantitative analysis of ground subsidence near AUCMs is possible.