Interaction prediction between groundwater and quarry extension using discrete choice models and artificial neural networks

摘要

Groundwater and rock are intensively exploited in the world. When a quarry isdeepened the water table of the exploited geological formation might bereached. A dewatering system is therefore installed so that the quarryactivities can continue, possibly impacting the nearby water catchments. Inorder to recommend an adequate feasibility study before deepening a quarry, wepropose two interaction indices between extractive activity and groundwaterresources based on hazard and vulnerability parameters used in the assessmentof natural hazards. The levels of each index (low, medium, high, very high)correspond to the potential impact of the quarry on the regional hydrogeology. The first index is based on a discrete choice modelling methodology while thesecond is relying on an artificial neural network. It is shown that these twocomplementary approaches (the former being probabilistic while the latter fullydeterministic) are able to predict accurately the level of interaction. Theiruse is finally illustrated by their application on the Boverie quarry and theTridaine gallery located in Belgium. The indices determine the currentinteraction level as well as the one resulting from future quarry extensions.The results highlight the very high interaction level of the quarry with thegallery.