GEOCHEMICAL PERSPECTIVES ON KARST AQUIFER STRUCTURE IN THE TEMPERATE MIDWEST

Delineating karst aquifers is difficult because of their extreme heterogeneity and anisotropy. In this study, we combined regional and temporal geochemistry data from karst springs in southern Indiana to create a geochemical framework for differentiating aquifers in a classic karst landscape. During base flow conditions in 2019 and 2020, 104 perennial springs in nine counties of the Indiana Uplands were visited and sampled.

A cluster analysis sorted these data into five spring groups based on ion chemistry. Two Ca²⁺/HCO₃^2-dominated groups were separated by their specific conductance; interpreted as a difference in residence time for meteoric recharge to springs. Two Ca²⁺/SO₄^2- dominated groups are distinguished by their different Na⁺/Cl^- concentrations, and are interpreted as deeper sourced, mineralized waters. The last group is intermediate between fresh and mineral waters. Relating the spring geochemistry to their spatial distribution, we observed that the freshwater springs with longer residence time concentrate in the Mitchell Plateau physiographic region whereas the shorter residence time freshwater springs and the mineral springs more frequently occur in the Crawford Uplands physiographic region.

We used measurements of δ¹³C to ascertain whether land use or bedrock geology have a first-order control on the chemistry of the springs. No relationship between land cover and measured δ¹³C was found. In contrast, a 2‰ shift in the measured δ¹³C between physiographic regions supports an important role for bedrock geology.

A Durov diagram of the spring data suggests linear mixing between freshwater and mineralized water on a regional scale. To discern the connection between the spring groups and regional aquifer mixing, we used a dataset from a temporal study on karst geochemistry at two of the sampled springs, Bluespring Caverns and Orangeville Rise. Our synthesis reveals a regional differentiation of springs with chemical contributions from mineralized waters sourced to interactions with evaporites and other springs with contributions sourced to basin brines. The value of this study is two-fold: it serves as a scaffold for exploring the geochemical regimes of karst aquifers in the temperate Midwest and is a step forward in attempts to further discretize aquifer mapping in a challenging landscape.