The surficial sand and gravel aquifer is susceptible to effects from land-use in the Upper Mississippi River Basin study unit of the National Water-Quality Assessment (NAWQA) Program. The purpose of this report is to describe the ground-water quality and the assessment of how different land-uses affect the shallow ground-water quality in the surficial sand and gravel aquifer. Ground-water quality was compared in three different land-use areas; an urban residential/commercial area on the edge of the Anoka Sand Plain in a portion of the Twin Cities metropolitan area (urban study), an intensive agricultural area in the Anoka Sand Plain (agricultural study), and a forested area in the Bemidji-Bagley Sand Plain (forested study). Ground water was sampled and analyzed for about 200 constituents, including physical parameters, major ions, selected trace elements, nutrients, dissolved organic carbon, selected pesticides, selected volatile organic compounds (VOCs), and tritium. The urban study wells were sampled during June and July 1996. The agricultural study wells were sampled during May and September 1998. The forested study wells were sampled during June 1998.

The depth to water below the land surface generally was less than 20 ft in all three land-use studies. The median pH value in the urban study was 7.2, with medians in the agricultural and forested studies at 7.4 and 7.5, respectively. The median specific conductance was significantly greater in the urban study than in the agricultural and forested studies (914, 553, and 487 µS/cm respectively). The median dissolved oxygen concentration in the urban study (0.9 mg/L) was significantly less than the median in the agricultural or forested studies (5.3 and 2.3 mg/L respectively). Alkalinities in the agricultural study, with a median of 178 mg/L as CaCO₃, were significantly less than medians in the urban or forested studies (261 and 246 mg/L as CaCO₃, respectively).

The water composition in the surficial aquifer in all three land-use studies is dominated by calcium, magnesium, and bicarbonate. Sulfate and chloride concentrations in water samples were significantly greater in the urban study than in the agricultural or forested studies, and concentrations in the forested study were least. Most of the water samples in all three land-use studies were very hard (greater than 180 mg/L as CaCO₃).

Nitrate-nitrogen concentrations were greatest in the agricultural study, in which 38 percent of water samples exceeded the U.S. Environmental Protection Agency’s maximum contaminant level (MCL) for nitrate-nitrogen. Nitrate-nitrogen was greater than the MCL in 3 percent of urban study samples. None of the forested study samples exceeded the MCL for nitrate-nitrogen. Concentrations of phosphorus generally were less than 0.05 mg/ L, with no significant differences between the land-uses.

A total of 19 pesticides were detected in water samples from one or more land-use study wells, with 11 pesticides detected in the urban study, 14 detected in the agricultural study, and 4 detected in the forested study. Atrazine, deethylatrazine, and simazine were the only pesticides detected in all three land-use studies. A significantly greater percentage of pesticide detections were present in water samples from the agricultural study than from the urban or forested studies (86.2, 56.7, and 46.7 percent, respectively). Prometon was the most frequently detected pesticide in the urban study. Atrazine and deethylatrazine were the most frequently detected pesticides in the agricultural and forested studies.

Twenty-one VOCs were detected in water samples from one or more land-use study wells, with 19 detected in the urban study, 7 detected in the agricultural study, and none detected in the forested study. Chloromethane, trichloromethane, methylbenzene, trichlorofluoromethane, and benzene were all detected in both the urban and agricultural studies. A significantly greater percentage of VOC detections were present in water samples from the urban study than from the agricultural study (90 and 50 percent, respectively). Carbon disulfide was the most frequently detected VOC in the urban study. The compound 1,2,3,4-tetramethyl benzene was the most frequently detected VOC in the agricultural study.

Tritium concentrations indicate that the water in the surficial sand and gravel aquifer has been recharged since 1953. Median tritium concentrations ranged from 11.6 to 12.8 tritium units. No significant difference in tritium concentrations was present between the three land-use studies.

Comparisons of previous land-use studies in Minnesota with the three NAWQA land-use studies generally indicated the same patterns. Ground-water quality in surficial sand and gravel aquifers is affected by land-use practices. Ground water in urban studies has greater specific conductances, alkalinities, chloride, sodium, sulfate, and dissolved solid concentrations than agricultural or forested/undeveloped studies. Nitrate-nitrogen was detected in greater concentrations in agricultural studies than in urban studies, with concentrations in the forested/undeveloped studies less than in the agricultural or the urban studies. Agricultural studies have the greatest detection rates, numbers, and total concentrations of pesticides. Pesticide detection rates and total pesticide concentrations in the urban studies were less than in the agricultural studies, with the most frequently detected pesticides (prometon and dicamba) different than those in the agricultural studies (atrazine and deethylatrazine). A greater number of VOCs were detected in urban studies and at greater concentrations than in agricultural studies. Few pesticides or VOCs were detected in forested/undeveloped studies.