Recovery of injected freshwater to differentiate fracture flow in a low-permeability brackish aquifer

Summary A low-permeability weathered siltstone–sandstone aquifer containing brackish water was investigated to measure recoverability of injected freshwater with the aim of determining the significance of secondary porosity in contributing to groundwater flow and transport. Examination of the core, borehole geophysics, Radon-222, electromagnetic flowmeter (EMF) profiles and step-drawdown pumping tests did not identify whether fractures contribute to groundwater flow. A number of injection and recovery tests lasting from 3 days to 3 months using potable water showed a large degree of mixing with native groundwater. Withdrawal greater than 12–17% of the injected volume resulted in recovered water containing more native groundwater than injected water. A finite element solute transport model was set up to reproduce the observed salinity in recovered water. Without the inclusion of discrete fractures in the model it was not possible to get a fit between the observed and modelled salinity of recovered water within a realistic range of dispersivity values. The model was subsequently verified by using data from long-term injection and recovery trials. This evaluation of mixing conclusively demonstrated that the aquifer behaved as a fractured rock aquifer and not as an aquifer with primary porosity alone. Therefore, aquifer storage and recovery can be a very useful hydrogeological method to identify the occurrence of fracture flow in aquifers where there is a measurable concentration difference between the injected water and ambient groundwater.