Determining the Empirical Relationship Between Surface Ground Penetrating Radar (GPR) Reflection Amplitudes and Sub-wavelength "Thin-layer" Fracture Aperture Under No-flow Conditions
Abstract
Current methods of collecting data for modeling groundwater flow in fractured media (e.g., fractured bedrock) involve expensive and invasive procedures that typically yield poorly-constrained results due to highly spatially variable fracture apertures and the resulting channelization. Surface ground penetrating radar (GPR) surveys present an attractive alternative because the full two-dimensional distribution of fracture aperture may be determined. Typical fractures have sub-wavelength apertures (i.e., are considered "thin layers") and the fluid flow through the fractures is governed by the cubic law; therefore, precise aperture estimates are critical. In practice, researchers have observed a qualitative change in reflection amplitudes and presumed this change correlates with fracture aperture at the sub-wavelength scale. Several researchers have attempted to determine the relationship theoretically. One method involves a simplification of the thin-layer problem by reducing the scope of the investigation to the first interface of the layer. This simplification reduces the thin-layer to a boundary, thereby effectively removing the noise generated by the constructive or destructive interference generated by the second interface. The resulting theoretical relationships describe the relationship between the reflected amplitude and thickness of the layer (e.g., fracture aperture). We have created a physical model consisting of 2 large ultra-high molecular weight polyethylene (UHMW-PE) blocks that have electromagnetic properties of real earth materials, separated by thin (~0.1 mm) inserts to create a range of aperture sizes. To change the aperture, we started by adding 1 insert for the first 51 surveys (0-5 mm) followed by 2 inserts for the next 25 surveys (5.2-10 mm), followed by 5 inserts for the next 20 surveys (10.5-20 mm) and lastly 1 additional survey taken at 300 inserts (30 mm). At each aperture increment, a GPR survey using 1000 MHz antennae was run at the center of the block. By directly obtaining the reflected amplitudes in an idealized fracture media, we then tested the various theoretical equations for modeling fracture aperture. Our preliminary results are not in agreement with the model and warrant a new empirical equation.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2007
- Bibcode:
- 2007AGUFMNS31B0393B
- Keywords:
-
- 1835 Hydrogeophysics;
- 1894 Instruments and techniques: modeling