Preliminary Calculations of inflow to the D-holes at the Stripa mine

A prediction of inflow to the D-holes at Stripa has been based on data from six boreholes including fracture characteristics, stress, single borehole geophysical logging, cross-hole and reflection radar and seismic, and single hole packer test measurements. Maps of fracture traces on the drift walls have also been made. The interpretation of seismic and radar tomography data has predicted six major fracture zones. We have compared the zones with the single hole hydrologic tests to see if the geophysical zones could account for the hydrologic anomalies. Based on these studies we developed a hydrologic conceptual model where all the conductive elements of the model are confined to fracture zones. We have modeled the zones as planes with a square grid of conductors that form the possible paths for fluid flow. In order to determine which of the grid elements are active and which are inactive, we employ Simulated Annealing. Simulated annealing is an inverse technique which is used to find the pattern of conductances which cause the model to behave in the way the insitu cross-hole tests behaved. First, well tests that were performed insitu are numerically simulated in the model using the network generator, CHANGE plus the fracture flow code, TRINET. Then the Simulated Annealing algorithm is used to modify the model until the behavior of the model matches the observed behavior. This is done by searching for patterns of conductors that respond to the well test in nearly the same way that the real well test was observed to respond. This final configuration of the grid can be considered as an averaged, rastorized version of a possible conductance geometry. To make a prediction with this model, we change the boundary conditions and calculate flow for the case that has not yet been measured