As a result of concerns about salt intrusion into drinking water aquifers, the effects of highway deicing chemicals on shallow aquifers were studied at eight locations in Ohio from 1988 through 2002. The study was done by the U.S. Geological Survey, in cooperation with the Ohio Department of Transportation and the Federal Highway Administration. Sites were selected along major undivided highways where drainage is by open ditches and ground-water flow is approximately perpendicular to the highway. Records of deicer application rates were kept, and apparent movement of deicing chemicals through shallow, unconsolidated aquifers was monitored by means of periodic measurements of specific conductance and concentrations of dissolved sodium, calcium, and chloride. The State routes monitored were the following: State Route (SR) 3 in Ashland County, SR 84 in Ashtabula County, SR 29 in Champaign County, SR 4 in Clark County, SR 2 in Lucas County, SR 104 in Pickaway County, SR 14 in Portage County, and SR 97 in Richland County.

The study began in 1988 with background data collection, extensive literature review, and site selection. This process, including drilling of wells at numerous test sites and the eight selected sites, lasted 3 years. Routine groundwater sampling at 4- to 6-week intervals began in January 1991 and continued through September 1999. A multilevel, passive flow ground-water sampling device was constructed and used. Other conditions monitored on a regular basis included ground-water level (monitored continuously), specific conductance, air and soil temperature, precipitation,chloride concentration in soil samples, and deicing-chemical application times and rates.

Evidence from water analysis, specific-conductance measurements, and surface-geophysical measurements indicates that three of the eight sites (Ashtabula County, Lucas County, and Portage County sites) were affected by direct application of deicing chemicals. Climatic data collected during the study show that cold weather, and therefore deicing-chemical application rates, varied from south to north across the State. As a consequence, only minor traces of dissolved chloride (mean, 24–43 mg/L (milligrams per liter)) above background concentrations (mean, 13–23 mg/L) were determined in ground-water samples from the southernmost sites (approximately 3930' to 40 N latitude—Champaign County, Clark County, and Pickaway County). At the Ashland and Richland County sites (approximately 4030' N latitude), dissolved-chloride concentrations increased above background concentrations only intermittently (mean background concentrations 4–41 mg/L, rising to a mean of 40–56 mg/L in downgradient wells). At the northernmost sites (41 30' to 42 N latitude—Lucas County, Portage County, and Ashtabula County), deicing-chemical application was consistent throughout the winter, and downgradient dissolved-chloride concentrations (mean, 124–345 mg/L) rarely returned to background concentrations (mean, 7–37 mg/L) throughout the study period.

Other factors than application rate that may affect the movement of deicing chemicals through an aquifer were precipitation amounts, the types of subsurface materials, ground-water velocity and gradient, hydraulic conductivity, soil type, land use, and Ohio Department of Transportation deicing priority.