Simple Procedure for Preliminary Estimation of the Permeability of Randomly Fractured Rock Masses

Abstract

A data-driven approach is used to show the significant discrepancy between the coefficients of permeability that are estimated through packer field tests and those back-calculated from the actual discharge pumped out of randomly fractured rock masses (i.e., masses that do not have predominate oriented fracture sets) using either dewatering or pumping well systems. The presented data, some of the richest of its kind, are used to develop a novel correlation between the estimated and back-calculated coefficients of permeability. A procedure is proposed to evaluate the permeability of the rock masses that considers two different but frequently encountered scenarios: (1) lack of any field permeability test data; or (2) only packer test results are available. Numerical examples are presented to illustrate how the suggested procedure could be used. It constitutes an advance in the state of practice because the commonly used procedures to estimate the permeability of randomly fractured rock masses employ the direct use of packer field test results, which, as the data shows, could be orders of magnitudes out. The proposed procedure could be particularly useful for practitioners when faced with cases where field pumping tests are not feasible due to budget, time constraints, or both. It gives them a simple and reliable approach (based on case histories) to produce preliminary designs of dewatering systems in randomly fractured rocks.