Hydrogeophysical characterization of transport processes in fractured rock by combining push‐pull and single‐hole ground penetrating radar experiments

Abstract

International audienceThe in situ characterization of transport processes in fractured media is particularly challenging dueto the considerable spatial uncertainty on tracer pathways and dominant controlling processes, suchas dispersion, channeling, trapping, matrix diffusion, ambient and density driven flows. Weattempted to reduce this uncertainty by coupling push-pull tracer experiments with single-holeground penetrating radar (GPR) time-lapse imaging. The experiments involved different injectionfractures, chaser volumes and resting times, and were performed at the fractured rock research siteof Ploemeur in France (H+ network, hplus.ore.fr/en). For the GPR acquisitions we used both fixedand moving antenna setups in a borehole that was isolated with a flexible liner. During the fixedantennaexperiment, time-varying GPR reflections allowed us to track the spatial and temporaldynamics of the tracer during the push-pull experiment. During the moving antenna experiments,we clearly imaged the dominant fractures in which tracer transport took place, fractures in whichthe tracer was trapped for longer time periods and the spatial extent of the tracer distribution (up to8 meters) at different times. This demonstrated the existence of strongly channelized flow in thefirst few meters and radial flow at greater distances. By varying the resting time of a givenexperiment, we identified regions affected by density-driven and ambient flow. These experimentsopen up new perspectives for coupled hydrogeophysical inversion aimed at understanding transportphenomena in fractured rock formations

Combining periodic hydraulic tests and surface tilt measurements to explore in situ fracture hydromechanics

“…A leakage should produce a pressure drop in the first fracture, as well as an important change in mechanical stress distribution. The presence of a well‐connected fracture network at our site, involving permeable subvertical drains crossing the subhorizontal fractures of B1, was demonstrated by many previous studies [ Dorn et al , , ; Klepikova et al , ; Le Borgne et al , ; Shakas et al , ]. Another effect could be a sharp decrease in fracture compliance with distance from the well.…”

“…Such leakage effect could dramatically decrease fracture deformation beyond the intersection. A way to test this hypothesis on the field would be to combine the harmonic hydromechanical test described in this paper, with geophysical methods which are sensitive to subsurface fluid paths like GPR [e.g., Shakas et al , ] or self‐potential data, already used in the periodic test framework by Maineult et al [] and by Soueid Ahmed et al []. Furthermore, we were able to evidence the distinct mechanical responses of the three tested fractures at depth.…”

Neutrally buoyant tracers in hydrogeophysics: Field demonstration in fractured rock

“…Our results suggest that wethanalt, a mixture of saline water and ethanol, is a suitable tracer for conducting geophysical monitoring using electrical or electromagnetic methods, when density effects are undesirable. Tracer test experiments conducted in push‐pull and push‐wait‐pull configurations, in conjunction with single‐hole GPR monitoring, confirm that wethanalt provides a strong GPR signal and does not exhibit the density‐driven downward flow observed in our past experiments with dense saline tracers [ Shakas et al , ]. Therefore, wethanalt may significantly improve our ability to monitor flow and transport processes in situ with hydrogeophysical methods, without the complications of density‐driven flow and instabilities.…”

“…We then take the difference of each section and the reference (taken before the initiation of the push‐pull experiment), apply a time gain, and finally use a Kirchhoff migration algorithm with a constant velocity model of . This results in migrated GPR difference sections where the presence of the conductive tracer manifests itself as alternating (green‐orange) stripes, whose horizontal extent is caused by the finite size of the source wavelet [ Shakas et al , ]. For visualization purposes, we suppress any reflections that are below the estimated noise level of our GPR data set (computed as 15% of the maximum amplitude).…”

Time-lapse ground-penetrating radar full-waveform inversion to detect tracer plumes: A numerical study

“…Time-lapse geophysical imaging methods are used in environmental science and engineering to monitor tracer migration (Singha and Gorelick, 2005), contaminant leaks (Binley et al, 1997), tide-driven salinity changes (Slater and Sandberg, 2000), hydrogeophysical characterization of transport processes (Shakas et al, 2016) and other applications. Time-lapse changes can be quantitatively characterized due to the changing geophysical properties of the groundwater and soil states.…”