Arsenic in Minnesota groundwater—Occurrence and relation to hydrogeologic and geochemical factors

Science Center Objects

Geologic-sourced arsenic is common in Minnesota groundwater. Drinking-water managers, well owners, and well contractors need to know where and why high arsenic in groundwater is likely to occur in wells in order to take measures to protect public health. The USGS is assessing the spatial distribution of high arsenic groundwater in Minnesota, and identifying factors affecting arsenic mobilization.

Project Objectives:

The overall study objective is to better delineate and help explain groundwater arsenic (As) concentrations in Minnesota over space and time.

Specific objectives are:

  • Evaluate methods used by well drillers to collect groundwater samples for arsenic analysis
  • Evaluate temporal arsenic concentration variability over one year in newly constructed wells
  • Assess the relation between arsenic concentration and hydrogeologic, geochemical, and well-construction factors using statistical models

We have made progress on project objectives, as published in three journal articles.

We evaluated methods used by well drillers to collect groundwater samples for arsenic analysis. We found that water sampling protocols—filtration, timing, sample point—influence variability in arsenic concentrations in new drinking water wells. Variability in measured arsenic concentration at a well was reduced when samples were (1) filtered, (2) collected from household plumbing instead of from the drill rig pump, or (3) collected several months after well construction (instead of within 4 weeks of well installation).  Particulates and fine aquifer sediments entrained in groundwater samples, or other artifacts of drilling disturbance, can cause undesirable variability in geochemical measurements, including arsenic measurements. Establishing regulatory protocols requiring sample filtration and/or collection from household plumbing could improve the reliability of information provided to well owners and to secondary data users.

We developed a boosted regression tree (BRT) statistical model to assess the relation between arsenic concentration and hydrogeologic, geochemical, and well-construction factors. We found that variables describing aquifer properties and materials, position on the hydrologic landscape, and soil geochemistry were among the most influential for predicting the probability of elevated As. We also found that certain well construction attributes were influential in predicting As hazard. Smaller distances between the top of the well screen and overlying aquitard (proximity) and shorter well screen lengths were each associated with higher probabilities of elevated As. Influential predictor variables, which are either mapped across the region or are well construction attributes, are proxies in the model for measurable physical or geochemical causes of elevated As (e.g., redox condition, till or aquifer sediment chemistry, and water chemistry), which are not mapped across the region. Our results show a new, novel, and important finding from an As probability model: Controllable well construction choices (not just location or depth) influence As concentrations in drinking water from wells.

We evaluated the effects of geochemical changes after well installation and operation on arsenic concentrations. We examined changes in arsenic and other geochemical constituents over one year in water from 250 new domestic water wells. Our aquifers are late Quaternary-age glacial aquifers or fractured crystalline bedrock aquifers. During the study, arsenic concentrations increased in wells in glacial aquifers, and redox conditions changed toward more reducing. In bedrock aquifer wells, there was no significant change in arsenic concentrations, and conditions became more oxic. The arsenic concentration variability we identified has important implications for water treatment, and for programs that require testing of new wells, such as in Minnesota and New Jersey. Information on how and why concentrations of arsenic vary at new wells provides context on what constitutes a representative sample in situations where testing is required or desired. Measurement and mechanistic characterization of human-induced geochemical changes associated with drilling, installing, pumping, and sampling of drinking water wells can improve guidance to well owners and policy makers on when to sample wells.

The research is funded by the State of Minnesota Clean Water Fund through the Minnesota Department of Health and the USGS Cooperative Matching Fund. The work was also supported by the National Science Foundation Graduate Research Fellowship Program and an internship provided through the Graduate Research Internship Program (GRIP)

Well characteristics

This illustration compares the construction characteristics of two water wells. Note that the distance from the top of the well screen to the confining unit, or aquitard, is much shorter for the well on the right, as is the length of the screen in the underlying aquifer unit. Placing a well screen farther beneath the confining unit and/or using a longer-length screen, as shown for the well on the left, can decrease the likelihood of elevated arsenic concentrations in domestic well water. 

(Credit: Modified from Figure 1 in Erickson & Barnes, 2005, reprinted with permission.)

Additional Publications

Erickson, Melinda L. and Barnes, Randal J. 2005, Arsenic concentration variability in public water system wells in Minnesota, USA, Applied Geochemistry, Volume 21, Issue 2, February 2006, Pages 305–317, doi:10.1016/j.apgeochem.2005.12.005 

Erickson, Melinda L. and Barnes, Randal J., 2005, Glacial Sediment Causing Regional-Scale Elevated Arsenic in Drinking Water, Groundwater, Volume 43, Issue 6,  pages 796–805, DOI: 10.1111/j.1745-6584.2005.00053.x 

Erickson, Melinda L. and Barnes, Randal J, 2005. Well characteristics influencing arsenic concentrations in groundwater, Water research, Volume 39, Issue 16, Pages 4029-4039, doi:10.1016/j.watres.2005.07.026  and Erickson, Melinda L., 2005, Erratum to: Erickson, ML, Barnes, RJ, October 2005. Well characteristics influencing arsenic concentrations in ground water. Water Res. 39 (16) 4029–4039. Water Research, Volume 39, Issue 20, Pages 5277-5278 

Sarah Nicholas, Melinda L Erickson, Laurel G Woodruff, Alan R Knaeble, Matthew A Marcus, Joshua K Lynch, and Brandy M Toner, 2017. Solid-phase arsenic speciation in aquifer sediments: a micro-X-ray absorption spectroscopy approach for quantifying trace-level speciation. Geochimica et Cosmochimica Acta., 211, pp 228-255, http://dx.doi.org/10.1016/j.gca.2017.05.018