Assessing the timing and magnitude of precipitation-induced seepage into tunnels bored through fractured rock

Abstract

Seepage into tunnels bored through fractured rock is a common occurrence that can cause significant problems for the construction process, tunnel longevity, and the regional hydrogeology. Predictions of seepage using analytical solutions are often inaccurate due to the inherent assumptions and volumetric averaging of fractures. A conceptual model is first developed for this research by using the factors shown by previous studies to have control on net infiltration and seepage. An integrated hydrologic model, ParFlow is then used to investigate the control exhibited by these factors that include: climatic forcing; vegetation; soil type and depth; bedrock type; fracture spacing; and tunnel depth on the timing and magnitude of seepage into tunnels. A fracture continuum is generated for bedrock using FRACK, which maps discrete fracture networks to a finite difference grid with heterogeneous, anisotropic permeability fields. Simulations are run using hourly meteorological forcing over a two-year period. Surface and subsurface properties are varied individually to investigate the change in seepage response. Results show that fracture spacing, bedrock type, and overburden are particularly important pieces in obtaining reliable seepage estimates. Higher fracture spacing causes higher total seepage at a more constant rate than a lower spacing that exhibits a much larger range of fluctuation in seepage volumes. More permeable and porous bedrock actually increases lag times and seepage amounts that are observed to be relatively constant over time. Thicker and less conductive soils both increase lag times and reduce seepage magnitude.