Comparison of hydraulic and chemical methods for determining hydraulic conductivity and leakage rates in argillaceous aquitards

• Published in: Journal of hydrology (Amsterdam) 2016-01, Vol.532, p.102-121
• Main Authors: Batlle-Aguilar, Jordi, Cook, Peter G., Harrington, Glenn A.
• Format: Article
• Language: English
• Subjects: Aquifers; Clay aquitards; Computational fluid dynamics; Estimates; Fluid flow; Fluid flux; Ground-water flow; Hydraulics; Inter-aquifer leakage; Leakage; Permeability; Porewater velocity; Regional; Scale of application
• ISSN: 0022-1694; 1879-2707
• DOI: 10.1016/j.jhydrol.2015.11.035

•Hydraulic methods are used to estimate hydraulic conductivity.•Chemical methods are used to estimate leakage rate or porewater velocity.•KZ increases up to three orders of magnitude with horizontal scales of up to 1000m.•Each tracer used in aquitard porewater has a lower limit of resolution of velocity.•Methods testing only a portion of the aquitard thickness may overestimate KZ. The capacity of argillaceous aquitards to effectively separate fresh water aquifers from lower quality waters has been widely studied for several decades using both hydraulic and chemical methods. Hydraulic methods, namely laboratory, field and regional groundwater flow modelling, are used to estimate hydraulic conductivity (K), whereas chemical methods, namely aquitard porewater and aquifer groundwater chemistry, are used to estimate either leakage rates or porewater velocity (V). We reviewed a total of 40 formations where either K or V (or both) have been estimated. Typical vertical hydraulic conductivity (KZ) values in argillaceous materials estimated using hydraulic methods are 10−12–10−9ms−1. Usual V values, estimated using chemical methods, ranges between 0.01 and 1mmy−1, although the range is as wide as 10−4–103mmy−1, when inferred from hydraulic KZ measurements. Based on a Péclet number of 1, we calculated the lower limits of porewater velocity that can be reliably identified for different tracers, in most cases ranging from 100mmy−1 for temperature. Despite the limited number of sites where both methods were applied, comparison between hydraulic and chemical-derived values showed a reasonable correlation, although the range of KZ and V estimated using hydraulic methods is larger than that obtained using chemical methods. Methods applied to the whole aquitard thickness (some field hydraulic methods, regional groundwater flow modelling, aquitard porewater and aquifer chemistry) consistently result in lower KZ values, most probably indicating the presence of very low KZ layers within the aquitard, likely to be missed while using field hydraulic methods that test only a section of the entire thickness. KZ was observed to decrease with depth, presumably due to an increase in consolidation resulting in loss of porosity and smaller aperture of fractures and joints. Multi-disciplinary studies involving the use of hydraulic and chemical methods at different scales, combined with geophysical techniques to locate fractures at local and regional scale