Relating Geophysical and Hydrologic Properties Using Field-Scale Rock Physics

Understanding how the parameters estimated in a geophysical investigation are related to hydrologic properties of interest is an important part of a hydrogeophysical study. This problem is often tackled using rock physics to determine how the pore-scale properties of a medium, such as mineralogy, fluid content, and grain geometry, affect the geophysical response of a rock or sediment. It is then often assumed that these pore-scale rock physics insights can be used to interpret field-scale data. However, upscaling rock physics information is not straightforward. For example, the data obtained from an individual geophysical measurement in the field represents an average of local variations in pore-scale properties. Preferential sampling in heterogeneous environments can cause the field-scale relationship between geophysical and hydrological parameters to shift away from that determined at the pore scale, particularly in cases where property variations occur on spatial scales between the pore scale and the geophysical measurement scale. In addition to the sampling physics of individual measurements, spatial variations in the resolution of a geophysical survey can also impact the relationship between geophysical and hydrologic properties at the field scale; resolution is impacted by a number of factors, including the parameters defining the design and inversion of a field-scale geophysical survey. As a result, there is a need for methods that integrate variations in pore-scale rock properties with an understanding of geophysical sampling at the field-scale. To address this problem, we are using numerical analog models as tools to build field- scale rock physics relationships. Our approach allows for the flexible analysis of how factors like variations in geologic heterogeneity, changes in survey design, and uncertainty in subsurface properties and processes impact the relationship between geophysical and hydrologic properties. In this work we place an emphasis on describing how increasing non-linearity of geophysical estimation problems affects our ability to predict field-scale rock physics relationships. Specific examples discussed include the use of cross-borehole radar tomography for estimating water content and electrical resistivity tomography for monitoring the migration of a saline plume.