Resolving the source and mixing proportions of modern leakage to the Memphis aquifer in a municipal well field using geochemical and ³H/³He data, Memphis, Tennessee, USA

The Memphis aquifer in southwestern Tennessee is a confined to semi-confined unconsolidated sand aquifer and is the primary municipal water source for the Memphis metropolitan area. Leakage of modern water from shallow unconfined aquifers through the upper Claiborne confining unit locally degrades water quality in the Memphis aquifer and makes the aquifer more vulnerable to contamination. Major solute chemistry, tritium, and ³H/³He data were used to investigate the source and mixing proportions of modern water in the Memphis aquifer in the Davis well field, Memphis, Tennessee. Water quality in several production wells in the Davis well field has declined over the past 30 years, mainly through increased total dissolved solids, iron, alkalinity, and hardness. Trends in chemical data, tritium, and other hydrogeologic data support a source for the leakage from the Mississippi River Valley Alluvial aquifer. Mixing proportions of alluvial water in production well water obtained by inverse chemical modeling with PHREEQCi range from 7 to 45%. For two of the production wells, MLGW 414 and 432, ³H/³He data yield mixing ratios similar to those obtained from PHREEQCi in three of four cases; the dissimilar ratio is likely due to a poor solution from the PHREEQCi inverse modeling. Modeling of the age distribution obtained from MLGW 414 and 432 using an exponential-piston flow model (EPM) with an inverse solution computer code yielded mixed results. The EPM solution obtained for MLGW 414 converged with difficulty only for a 5-year transit time in the shallow aquifer and is consistent with a source from the Mississippi River Valley Alluvial aquifer; however, the modeled age of the water is greater than that observed. In comparison, the EPM solution for MLGW 432 converged for the 5- and 7-year transit periods in the shallow aquifer and yielded model ages consistent with observed ³H/³He ages; however, the extent of the maximum radii for infiltration source is not consistent with a Mississippi River Valley Alluvial aquifer source. Other potential sources for leakage to MLGW 432 include water from the Fluvial-terrace aquifer migrating along a fault east of the well field or infiltration of water from a lake south of the well field.