Reports
Geochemistry and factors affecting ground-water quality at three storm-water-management sites in Maryland
1994, Wilde, F.D.
Report of Investigations 59
Abstract
The effects of infiltration of storm runoff on ground-water chemistry and quality were examined at three suburban storm-water management impoundments at sites in Annapolis, Greenmount, and Prince Frederick, Maryland. Geochemical and hydrologic data were collected from December 1985 though June 1989. The Annapolis and Prince Frederick sites are in the Coastal Plain physiographic province, and the Greenmount site is in the Piedmont physiographic province. This study was a cooperative effort of the U.S. Geological Survey, the Maryland Geological Survey, the Maryland Department of the Environment, and the Governor's Council on Chesapeake Bay initiatives.
The objectives of the study were to determine whether the chemical composition of ground water beneath the impoundments changed as a result of storm-water infiltration, whether ground-water quality was adversely affected, and whether contaminants being sequestered in impoundments could potentially be mobilized to ground water.
Native geologic materials collected from drill cores at each site and bottom materials, collected each September from the Annapolis and Greenmount ponds were analyzed for particle size, selected chemical constituents, and the predominant mineralogy. Aqueous solutions were analyzed for major ions, a large suite of trace elements, and volatile and polyaromatic organic compounds; pH, dissolved oxygen, alkalinity, chloride, and specific conductance were measured in the field. Samples of runoff, impoundment water, unsaturated-zone water, and ground water were collected triannually for extensive chemical analyses and at least monthly for field measurements. The results were compared for two sites with storm-water ponds (Annapolis and Greenmount) and one porous-pavement site with a subsurface impoundment (Prince Frederick).
Either primary or secondary maximum contaminant levels established by the U.S. Environmental Protection Agency (USEPA) for aluminum, cadmium, chloride, chromium, and lead in drinking water were exceeded from time to time in ground-water samples collected beneath and downgradient from the impoundments. In addition, uncharacteristically high concentrations of barium, copper, molybdenum, nickel, strontium, vanadium, and zinc occasionally were reported in ground water beneath impoundments, and median concentrations of barium, cadmium, chloride, copper, nickel, and zinc were elevated in some ground-water samples beneath the study sites. Chromium and lead were rarely detected in ground water. Low concentrations of arsenic were detected sporadically in storm water and ground water. Concentrations of volatile organic compounds were usually below or near detection in storm-water and ground-water samples; small concentrations of polyaromatic organic compounds were detected only in pond-bottom materials.
Pond-bottom materials generally were effective scavengers of trace metals introduced to storm-water impoundments in the runoff. Between 1986 through 1988, concentrations of lead increased from below detection to 28 parts per million (ppm), and zinc concentrations increased from 54 to 344 ppm in bottom materials collected from the Annapolis impoundment. In addition, copper concentrations increased 3.5 to 40 ppm and nickel concentrations increased from 6 to 16 ppm in bottom materials at the Annapolis site. For the same time period at the Greenmount impoundment, concentrations of lead increased from 20 to 90 ppm; zinc concentrations increased from 59 to 469 ppm; and nickel concentrations increased from 35 to 48 ppm (there was a net decrease in the copper concentration). Despite this accumulation of metals in bottom materials, concentrations of these and other metals were considerably elevated in ground water beneath and downgradient from the impoundments. Cadmium concentrations did not increase in bottom materials, although cadmium was a common constituent in storm water and sorbed readily to the bottom materials in laboratory tests. Ground-water samples collected from the control wells at each site had concentrations of cadmium that were below or near detection, whereas concentrations were as high as 27, 26, and 8.4 micrograms/L (micrograms per liter) beneath the impoundments at Annapolis, Greenmount, and Prince Frederick, respectively.
Storm water was the primary source of most contaminants that were found at elevated concentrations in ground-water samples collected beneath the impoundments. Contaminants entered ground water as a result of several variables, including direct storm-water infiltration; impoundment-related modifications of pH and redox that periodically favored metal mobilization from pond-bottom or aquifer materials; and formation of anionic or neutral complexes.
Metal mobility in the impoundments was mitigated by ion exchange, sorption, and mineral precipitation; storm water was aerobic and usually had neutral or higher pH that generally did not favor the presence of soluble species of most metals. Nevertheless, cadmium sorption in the impoundments may have been excluded by competing cations. Moreover, conditions for complexing of cadmium with organic compounds and chloride was favorable in impoundments and could have enhanced cadmium transport to ground water.
Algal photosynthesis modified the pond-water chemistry at Annapolis and Greenmount, increasing pH to 9.0 or greater, whereas algal respiration and rainwater dilution decreased pH to 6.5. Algal mediation of the pond-water pH at Greenmount resulted in a median pH of 9.2. Because aluminum solubility increases exponentially at about pH 9.0, aluminum concentrations in pond water at this site exceeded the U.S. Environmental Protection Agency's drinking-water regulation of 50 micrograms/L in most samples and may have contributed to elevated aluminum concentrations measured in ground water beneath the impoundment. Algal activity was less intense in the Annapolis impoundment, where the median pond-water pH was 7.7 and the median aluminum concentration was 30 micrograms/L; occasional measurements of aluminum concentrations greater than 50 micrograms/L in pond water corresponded with pH near or greater than 9.0. In ground water beneath the Annapolis pond, aluminum concentrations were below 50 micrograms/L in the samples collected, with only one exception.
The solubility of most trace metals increases with pH decrease below neutrality. Therefore, the decrease of pond pH below neutral possibly mobilized iron, copper, nickel, and zinc periodically from bottom materials; alternatively, metals dissolved in storm water could have been transported to ground water because kinetics were unfavorable for sorption to bottom materials.
The pH of ground water tended to keep the metals in solution at each site. Ground-water pH beneath the impoundments was reduced to below background pH: from 5.13 to between 4.18 and 4.94 at Annapolis, from 5.38 to between 4.9 and 5.29 at Greenmount, and from 6.71 to between 4.39 and 4.8 at Prince Frederick. Periodic mobilization of iron from the impoundment to ground water (and the consequent precipitation of iron hydroxides in the aquifer) has been suggested as a cause of reduced pH in beneath-pond ground water at the Annapolis and Greenmount sites. Lithologic composition is the primary control on order-of-magnitude changes in ground-water pH at Prince Frederick.
With the exception of chloride, ground-water contamination was least at the Prince Frederick site, possibly because of low contamination concentrations in storm water entering the Prince Frederick impoundment, or because contaminant mobility was restricted by the well-buffered and stable chemical environment in the impoundment. Dissolution of the rock aggregate in the Prince Frederick impoundment buffered impounded water to a pH of about 8.4, but dissolution also released high concentrations of magnesium and low concentrations of nickel and possibly chromium that affected ground-water chemistry.
Chloride contamination was ubiquitous in ground water receiving storm-runoff infiltrate. The ground water beneath the impoundments was modified to a chloride-dominated solution at each site throughout the period of study (native ground waters were mixed cation-mixed anion, magnesium-nitrate/chloride, and calcium-bicarbonate types at respective study sites). The only major source of chloride to ground water was storm-water infiltration during periods of road salting (road salting occurred no more than five times a year during the study period).
Chloride concentrations were measured at least biweekly and before, during, and after selected storms. The effect of seasonal and storm-specific recharge on concentrations of chloride in ground water beneath the impoundments monitored was a temporary dilution – usually lasting no longer than several days. Chloride concentrations in ground water beneath impoundments increased whenever rainfall was low and evapotranspiration rates were high, resulting in highest concentrations during the summer and early autumn at each site. The magnitude and persistence of the chloride contamination indicated that chloride, although a very soluble constituent was not being flushed readily from the ground-water systems studied. Chloride concentration in ground water beneath impoundments increased whenever rainfall was low and evapotranspiration rates were high.
Probable factors contributing to the persistent chloride domination of the major-ion chemistry of ground water at each study site were (1) low ground-water flow rates relative to storm-water infiltration rates; (2) limited dilution potential because sites were within a maximum of 300 feet from the inferred ground-water divide; and (3) capillary forces.
The unsaturated-zone chloride data suggest that capillary processes cause retardation of chloride transport and can allow chloride buildup, especially in the zone of tension saturation. This could serve as a model for explaining inhibition of transport of other contaminants. With the exception of chloride concentrations, however, the periods of data collection and (or) sample frequency generally were insufficient to determine temporal trends in concentrations for trace metals and other constituents.
