USING SPECIFIC CONDUCTANCE TELEMETRY TO MONITOR GROUNDWATER CHLORIDE IN REAL TIME

In the Twin Cities Metropolitan Area (TCMA), there are several thousand wells for pumping groundwater and monitoring for contaminants. It is known that chloride concentrations are increasing in these unconsolidated and bedrock aquifers due to winter road deicing and that chloride concentrations can vary by as much as 40% over the course of a single year. Most wells are currently sampled annually at best and it is not feasible to employ the large number of geotechnicians needed to physically gather and test samples from a sufficient fraction of sites to resolve those subannual variations. We advocate for the deployment of a sentinel network of in-situ specific conductance sensors across the TCMA that can autonomously provide measurements on short timescales.

Specific conductance is easily measured and will be proportional to chloride concentration though the exact relationship will vary between aquifers based on their unique combination of dissolved solids. Taking into consideration the work done to characterize groundwater flow directions in the TCMA as well as subsurface features such as buried bedrock valleys, monitoring sites can be chosen to maximize the insight into lateral movement of chloride within individual aquifers. In addition, utilizing clusters of wells in the same area that sample different aquifers will help constrain the vertical component of chloride movement.

A pilot installation of sensors in two wells on the University of Minnesota, Minneapolis campus demonstrates the promise of this approach. Multiple times per hour, these sensors upload data to a public website including barometric pressure, temperature, specific conductance, and water depth in one well in quaternary deposits well and one in underlying bedrock. Using 50 years of data gathered by three state agencies, we have established a robust empirical relationship between chloride concentration and specific conductance for the quaternary unit which can allow resource managers to rapidly see variations in well conditions at a range of timescales and establish links between natural and anthropogenic processes at the surface and their time-delayed expression in the two aquifers.