A GIS-based Weights-of-Evidence model for mapping cliff instabilities associated with mine subsidence
Autor: | Ernest Y Baafi, G. M. Brassington, Phillip N Flentje, D. R. Palamara, Hani Zahiri |
---|---|
Rok vydání: | 2006 |
Předmět: |
geography
Geographic information system geography.geographical_feature_category business.industry General Engineering Subsidence Curvature Training (civil) Rockfall Mining engineering Earth and Planetary Sciences (miscellaneous) Cliff General Earth and Planetary Sciences Environmental Chemistry Sedimentary rock Longwall mining business Geology General Environmental Science Water Science and Technology |
Zdroj: | Environmental Geology. 51:377-386 |
ISSN: | 1432-0495 0943-0105 |
DOI: | 10.1007/s00254-006-0333-y |
Popis: | The Weights-of-Evidence (W-of-E) technique was applied, within a geographic information system (GIS), to derive a model of rockfall potential associated with mining-induced subsidence. The purpose of the model was to describe the potential for rockfalls from up to 60 m high steep sandstone gorges and slopes associated with proposed underground longwall operations within the immediate vicinity of a previously mined area. Ten known rock falls associated with the previous mining operation were used as training points. Six evidential themes were considered-slope, cliff height, planform curvature, profile curvature, the distance of the cliff areas from the longwall panels, and the distance of the cliff areas from the river. Two models were created, one based on a mine layout in which longwall panels extend beneath the steep areas of a nearby river, and a second in which the mine layout is modified so that mining does not occur directly beneath or within 50 m of the steep slopes. This is to allow for the comparison of rockfall potential based on different mining configurations. The results demonstrate that the W-of-E method is a suitable tool for mine subsidence impact assessment, and suggest that not mining directly under the Nepean river may decrease rockfall potential, on average, by approximately ten times. Numerous limitations with the results, relating to the availability of appropriate evidential theme data and the accuracy of training points, are discussed. |
Databáze: | OpenAIRE |
Externí odkaz: |