The U.S. Department of Energy Office of Legacy Management (LM) collects groundwater samples at a number of the 90 sites we manage. Some samples are taken solely to help us better understand the groundwater system, while many are required to demonstrate compliance with applicable environmental regulations. To meet required groundwater compliance goals, we perform sampling activities at least once every 5 years and, at many sites, annually or even semiannually. To ensure that human health and the environment are protected, we follow strict protocols for groundwater sampling (how samples are collected) and laboratory analysis, which can be found in various American Society for Testing and Materials (ASTM) procedures—or which adhere to U.S. Environmental Protection Agency (EPA) methods. Adherence to these methods is important because we use groundwater sample results to interpret trends over time, and consistency in sampling and analytical approaches is paramount.

Because contaminant concentrations are important success indicators of groundwater cleanup, in spring 2012, LM scientists were trying to understand why groundwater contaminant concentrations were increasing at a Shiprock, New Mexico, Disposal Site monitoring well. As a first step to discovery, the scientists decided to do something fairly simple. They inserted into the well a probe that measures specific conductance (SC)—water’s ability to conduct an electrical current (a measure of the concentration of chemicals dissolved in the water). As more solids are dissolved in the water (increasing salinity), the water conducts more electrical current, which is measured as higher SC.1

Rather than taking a measurement from only one point in the well (a typical sampling approach), the scientists lowered the probe slowly into the water column, recording SC and temperature at every half-foot interval. Over a span of just 12 feet, SC increased by over 60 percent, from about 12,000 microsiemens per centimeter (µS/cm)—in the uppermost part of the water column—to more than 20,000 µS/cm near the bottom of the well.2 Later sampling indicated the same pattern of increasing concentration with depth for site contaminants (uranium and sulfate). These findings marked the beginning of the Variation Project, an Applied Studies and Technology (AS&T) project aimed at understanding Variation in Groundwater Aquifers.

At LM sites with groundwater contamination, most compliance decisions are based partly, or largely, on groundwater sampling data. These data are used during site studies to propose a groundwater cleanup method and to estimate how quickly cleanup may occur. At many sites, samples are collected routinely and the results are submitted to stakeholders and regulators and used as the basis for interpretations regarding cleanup progress. Since 2014, groundwater samples have been collected from nearly 1,300 wells at about 40 LM sites. As part of routine reporting, the results are plotted on a map or included in a table. While the samples’ geographic locations are well known, the Variation Project is looking at differences in results due to the depth at which a sample is collected.

One might think that there is just one number representing the amount of dissolved chemicals in a well (i.e., that one number would apply no matter where the sample is collected), but at some LM site wells, that is not true. Concentrations do differ depending on how shallow or deep the sample is collected.

Since the early 2012 work, SC profiles have been taken at 400 wells at 15 LM sites in the western U.S. Most of the wells profiled (about 70 percent) had low variation, a finding that is encouraging with respect to interpretations of historical trends. Nonetheless, every site has at least one well with an SC profile variation high enough to warrant further examination. In some cases, as in the example above, the variation in the vertical SC profile measured in a single afternoon can explain the variation in historical sampling results. These results suggest that understanding the vertical changes in contaminant concentrations is important at some LM sites.

In these cases, changes to our groundwater sampling methods may be warranted, even though our approach (low-flow sampling) adheres to current EPA protocols.

As part of LM’s goal to be proactive and better understand groundwater aquifers, the next phase of this study will assess whether the same vertical variation in SC applies to other site contaminants. This was found in some wells, but it is too early to draw conclusions. These research results highlight the fact that, when interpreting groundwater data, it is important to fully account for the third dimension. In other words, depth matters!

_____________
1 Pure water, H2O, without any dissolved chemicals, is a very poor conductor of electricity. Water’s ability to conduct electricity is due to the amount of positively and negatively charged particles (ions) from chemicals dissolved in the water.

2 A siemens (symbolized S) is the Standard International (SI) unit of electric conductance; a microsiemens (1 μS) is equal to one-millionth of a siemens. The conductivity of drinking water generally ranges between 50 and 500 microsiemens per centimeter (μS/cm), whereas that for sea water is about 50,000 μS/cm.