ESTIMATING GROUNDWATER-FLOW VELOCITY FROM CHANGES IN CONTACT RESISTANCE DURING A SALTWATER TRACER EXPERIMENT

The contact resistance of a current electrode is the potential measure d at the surface of the electrode divided by the current strength and is a function of the resistivities and geometry of the formations surr ounding the current electrode. The equation describing the contact res istance of a half sphere electrode shows that the resistivity of the f ormation immediately surrounding the electrode dominates the size of t he contact resistance. Measurements of the contact resistance of a cur rent electrode over time can then be used to estimate changes in the f ormation resistivity close to the electrode over time. The formation r esistivity is directly related to the pore fluid conductivity, which a gain, is related to the concentration of a conducting solute. Existing solute flow theory can be used to relate the change in concentration with time of a solute to the groundwater flow velocity, hence the meas urements of contact resistance can be used to estimate a groundwater f low velocity. Hydrogeologists use the rate of dilution of a tracer inj ected into a borehole, monitored by water samples, to calculate the gr oundwater flow velocity. At Haslemoen, Norway, the exponential decay o f the inverse contact resistance during a tracer experiment was used f or estimating the velocity. A sodium-chloride solute slug of 0.5 m(3) was injected into the groundwater at 4 m depth through a well in which there was installed a short iron bar electrode, reaching from the sur face to just below the groundwater table. The well was cased with a pl astic tube so that only the lowermost part of the iron bar electrode w as in contact with the surrounding formation. The inverse contact resi stance of the electrode was monitored every 2 h over approximately 600 h, showing an exponential decay by time. Fitting of an exponential fu nction to the data gave the groundwater velocity parameter equal to ap proximately 0.26 m/day, a number which was in satisfactory agreement w ith velocity estimations using other methods. (C) 1997 Elsevier Scienc e B.V.