Fractured Bedrock Hydrogeologic Characterization Using Digital Rock Physics

Abstract

Bedrock groundwater systems in mountains are critical water resources, yet they are poorly understood. In part, this is due to sparse data on complex flowpaths. Mountainous environments are typically characterized by fractured and variably weathered bedrock with complex pore networks. The extent to which flow is partitioned between fractures in the bedrock, and rock matrix remains challenging to assess quantitatively. In this study, we use novel quantitative micro Computed Tomography (CT) to characterize the density, porosity, pore structure, and permeability of fractured argillaceous bedrock core from a forested montane hydrologic monitoring site.

By CT scanning a rock core, digital representations of the sample can be captured, and used to create digital rock physics models. One advantage of rock physics models is the ability to work with intact scanned cores. Most lab equipment for porosity and permeability testing cannot handle rocks larger than a few centimetres. By working with larger rock physics models, we are more likely to capture a representative elementary volume (REV) to be used in our analysis.

Density models can be created by scanning alongside objects of known density. Using these objects for calibration, CT attenuation can be converted to density at each voxel (3D pixel). A porosity model can be created by using an inverse relationship to density for each voxel. Effective medium theory is then used to create a velocity model of the rock. We used a finite difference method simulation to solve the wave equation at each node through the model of the fractured sample and computed wave-speeds. Fluid flow can also be simulated through the CT based models. Fluid flow modeling can quantify water flux partitioning between fractures and the rock matrix.

We compare the digital rock physics models to laboratory measurements of density, velocity, porosity and pore structure. Pore information has been evaluated with helium pycnometry, mercury intrusion porosimetry, and laboratory nuclear magnetic resonance. Fluid flow simulations, porosity, and velocity data are compared to field scale measurements at our intensive monitoring site to improve understanding of fluid pathways at the hillslope and catchment scale.

AAPG Datapages/Search and Discovery Article #90350 © 2019 AAPG Annual Convention and Exhibition, San Antonio, Texas, May 19-22, 2019