Backfill Impacts on Moisture Measurements in Fractured Rock

As the scope of hillslope-hydrology investigations extend deeper, there will likely be an increase in the use of backfill to facilitate sensor installations, particularly in fractured rock. Because of the disparity in hydrologic properties of backfill and the native rock being monitored, discrepancies in measured values are imminent. In this study, we assessed the impact of different types of backfill that can be used to provide hydraulic continuity between sensors and the "measured" environment. During a period of 4 yr, the hydrologic response to seasonal wetting and drying was monitored with identical time domain reflectometry (TDR) sensors embedded in native rock, fracture infill, augured native rock, and silica powder. We found that while all backfills responded to wetting and drying events, there were differences in the response to individual rainfall events and in the amount of moisture measured. Our observations show that the use of any type of backfill for monitoring fractured rock hydrology will result in distorted measurements; however, our analysis suggests that simple calibrations between native rock and backfill measurements allow moisture content changes in the native rock to be quantified using backfill measurements. Backfill that is finer textured than native rock provides the best estimate of water content during long-term, uninterrupted drying events when the backfill measurements are calibrated to native rock conditions. Irrespective of calibration, backfill materials coarser than the native rock provide the best detection of the timing and duration of the hydrologic response to precipitation.