National Water-Quality Assessment (NAWQA) Project

   Home    Headlines    Featured Studies    Liaison Committee    Informing Decision Makers    Data    Modeling/Software    Publications    Maps

Potential Corrosivity of Untreated Groundwater in the U.S.
    Press Release | Report | Data (LSI : PPGC) | FAQs


Frequently Asked Questions

About the Study

USGS Role Regarding Drinking Water

Human Health Implications

Information for Homeowners with private wells

Additional information on USGS assessments

About the Study

  • Are the findings of this study applicable to the recent drinking water problems experienced in Flint Michigan?
  • No. The source of corrosive water used for the Flint drinking water supply was the Flint River, a surface-water source whereas the USGS report deals with the potential corrosivity of untreated groundwater. In addition, the high levels of lead in drinking water supplied to residents of Flint were related to inadequate treatment of the corrosive source water pumped from the Flint River by the public water utility. The USGS study is only reporting on the potential corrosivity of untreated groundwater, with the implications of having potentially corrosive water directed primarily at homeowners who rely on private wells for their drinking water supply. Unlike public water supplies like Flint, the quality and safety of drinking water from private wells are not regulated by the Federal government or, in most cases, by state laws. Rather, individual homeowners are responsible for testing, treating, and maintaining their private water systems.

    The purpose of this study is to present national maps of the distribution of two indicators of the corrosivity of untreated groundwater across the Nation: the Langelier Saturation Index (LSI) and the Potential to Promote Galvanic Corrosion (PPGC). The PPGC is based on chloride-to-sulfate mass ratio (CSMR). For each indicator, two maps are presented:  one map shows the values of the indicator at individual groundwater sites and, the other map shows characteristic values for each state.  The maps are based on data at about 27,000 groundwater sites obtained from the US Geological Survey’s (USGS) National Water Information System (NWIS).  The combined corrosivity index map combines the LSI and PPGC into one general indicator of potentially corrosive groundwater to summarize study results at the state level.

    Corrosive groundwater, if untreated, can dissolve lead and other metals from pipes and other components present in household plumbing.  About 43.5 million people rely upon groundwater from private wells and springs for their source of drinking water. The corrosivity of untreated groundwater is only one of several factors that may affect the quality of household drinking water at the tap. Nevertheless, it is an essential factor that should be carefully considered in testing, treating, and maintaining the quality and safety of drinking water obtained from a private water system supplied by groundwater.  

    Water quality data from about 27,000 groundwater sites across the Nation were obtained from the U.S. Geological Survey’s National Water Information System. The samples were collected from 1991 to 2015.

    The study assessed corrosivity in samples from private wells, public supply wells, wells of other types, and springs.

    All groundwater samples were collected prior to any treatment or blending that potentially could alter the quality of the source groundwater. USGS personnel collected water samples after the well had been pumped a minimum of three casing volumes and after basic water-quality properties such as temperature, pH, and specific conductance had stabilized from outlets located before any pressure tanks or treatment systems.

    No. Samples were collected to represent the source of the groundwater being used for drinking water, not the quality of water at the tap.

    Water corrosivity is affected by more than just pH, although pH is a major driver of corrosivity in many cases. The corrosivity of water is characterized by factors such as pH, calcium concentration, hardness, alkalinity, dissolved solids, and temperature. There are several indicators of corrosivity. Two indicators are summarized in this report to provide a national characterization of the potential corrosivity of groundwater: the Langelier Saturation Index (LSI) and the chloride-to-sulfate mass ratio (CSMR). The LSI provides an indication of the extent to which calcium carbonate scale might be deposited inside pipes and other components of a distribution system (Langelier, 1936; Larson and others, 1942). In the absence of a protective scale, lead, if present, can dissolve from plumbing materials into the water (Langelier, 1936; Stumm and Morgan, 1981; Hu and others, 2012). The LSI only indicates the tendency for scaling to occur; it is not a measurement of corrosivity (Singely and others, 1984). Nevertheless, the LSI is commonly used as an indicator of the potential corrosivity of water.
    A three-tier classification system based on the chloride-to-sulfate mass ratio (CSMR) developed by Nguyen and others (2010, 2011) is used to assess levels of concern related to galvanic corrosion of lead in water distribution systems. If the source water entering those systems has a relatively elevated CSMR (Gregory 1985; Edwards and Triantafyllidou, 2007; Hu and others, 2012), the potential for galvanic corrosion to occur is elevated, especially in water with low values of alkalinity (Nguyen and others (2011). In this report, untreated groundwater is assessed and the three-tier classification is applied without considering the absence or presence of lead or other metals in the distribution system. Consequently, the three-tier classification system is referred to in this report as the Potential to Promote Galvanic Corrosion (PPGC).
    The LSI and PPGC classifications are combined into a Combined Corrosivity Index to classify the potential for corrosive groundwater at the individual state scale. Confidence intervals for these classifications are provided in the report Appendix.

    Langelier Saturation Index – maps and water-quality data described in the report are available at:
    Chloride-to-sulfate ratio – maps and water-quality data described in the report are available at:

    Classification of states with respect to potential corrosivity and the Potential to Promote Galvanic Corrosion (PPGC) was based on estimates of characteristic values for the states, and those estimates are subject to uncertainty. The uncertainty associated with the estimate of the average LSI was computed using the standard confidence interval at a 90 percent confidence level (Ott and Longnecker, 2001). The uncertainty associated with estimating the proportion of groundwater sites with a given classification was computed using the Clopper-Pearson interval (Clopper and Pearson, 1934) at a 90 percent confidence level. Given a 90 percent confidence interval, six states could be given a classification that is different from the one based on the characteristic value for the state: Kentucky, Missouri, Montana, New Hampshire, North Carolina, and Vermont. Refer to the report Appendix for further discussion on the classification uncertainty.