Abstract

Groundwater quality in the 2,079 square mile Northern San Joaquin Basin (Northern San Joaquin) study unit was investigated from December 2004 through February 2005 as part of the Priority Basin Project of the Groundwater Ambient Monitoring and Assessment (GAMA) Program. The GAMA Priority Basin Project was developed in response to the Groundwater Quality Monitoring Act of 2001 that was passed by the State of California and is being conducted by the California State Water Resources Control Board in collaboration with the U.S. Geological Survey and the Lawrence Livermore National Laboratory.

The Northern San Joaquin study unit was the third study unit to be designed and sampled as part of the Priority Basin Project. Results of the study provide a spatially unbiased assessment of the quality of raw (untreated) groundwater, as well as a statistically consistent basis for comparing water quality throughout California. Samples were collected from 61 wells in parts of Alameda, Amador, Calaveras, Contra Costa, San Joaquin, and Stanislaus Counties; 51 of the wells were selected using a spatially distributed, randomized grid-based approach to provide statistical representation of the study area (grid wells), and 10 of the wells were sampled to increase spatial density and provide additional information for the evaluation of water chemistry in the study unit (understanding/flowpath wells).

The primary aquifer systems (hereinafter, primary aquifers) assessed in this study are defined by the depth intervals of the wells in the California Department of Public Health database for each study unit. The quality of groundwater in shallow or deep water-bearing zones may differ from quality of groundwater in the primary aquifers; shallow groundwater may be more vulnerable to contamination from the surface. Two types of assessments were made: (1) status, assessment of the current quality of the groundwater resource; and (2) understanding, identification of the natural and human factors affecting groundwater quality.

Relative-concentrations (sample concentrations divided by benchmark concentrations) were used for evaluating groundwater quality for those constituents that have Federal or California regulatory or non-regulatory benchmarks for drinking-water quality. Benchmarks used in this study were either health-based (regulatory and non-regulatory) or aesthetic based (non-regulatory). For inorganic constituents, relative-concentrations were classified as high (equal to or greater than 1.0), indicating relative-concentrations greater than benchmarks; moderate (equal to or greater than 0.5, and less than 1.0); or, low (less than 0.5). For organic and special- interest constituents [1,2,3-trichloropropane (1,2,3-TCP), N-nitrosodimethylamine (NDMA), and perchlorate], relative- concentrations were classified as high (equal to or greater than 1.0); moderate (equal to or greater than 0.1 and less than 1.0); or, low (less than 0.1).

Aquifer-scale proportion was used as the primary metric in the status assessment for groundwater quality. High aquifer- scale proportion is defined as the percentage of the primary aquifer with relative-concentrations greater than 1.0; moderate and low aquifer-scale proportions are defined as the percentage of the primary aquifer with moderate and low relative- concentrations, respectively. The methods used to calculate aquifer-scale proportions are based on an equal-area grid; thus, the proportions are areal rather than volumetric. Two statistical approaches—grid-based, which used one value per grid cell, and spatially weighted, which used the full dataset—were used to calculate aquifer-scale proportions for individual constituents and classes of constituents. The spatially weighted estimates of high aquifer-scale proportions were within the 90-percent confidence intervals of the grid-based estimates in all cases. The understanding assessment used statistical correlations between constituent relative-concentrations and values of selected explanatory factors to identify those factors potentially affecting constituent relative-concentrations and occurrence. Individual constituents detected at high relative-concentrations or those detected at low relative-concentrations in substantial proportions of the aquifer (greater than 10 percent) were evaluated statistically in relation to selected explanatory factors. Explanatory factors evaluated in this report were land use, well depth, depth to top-of-perforation, lateral and vertical position within the flow system, groundwater age, and geochemical conditions.

The status assessment for inorganic constituents showed that relative-concentrations (one or more) were high, relative to health-based benchmarks, in 13 percent of the primary aquifer, moderate in 29 percent, and low in 58 percent. High relative-concentrations of inorganic constituents in the primary aquifer reflected high proportions of arsenic (high relative-concentrations in 9.4 percent of the aquifer) and boron (7.6 percent). Inorganic constituents with aesthetic-based benchmarks [non-regulatory secondary maximum contaminant levels (SMCLs)], manganese, iron, total dissolved solids (TDS), chloride, and sulfate, were detected at high relative-concentrations in 34, 11, 5.8, 3.9, and 2 percent of the primary aquifer, respectively. SMCLs are benchmarks given to constituents with technical properties that can make drinking water undesirable with respect to taste, staining, or scaling at high relative-concentrations.

The status assessment for organic constituents showed that relative-concentrations (one or more) were high in 2.7 percent, moderate in 6.9 percent and in 90 percent of the primary aquifer of the Northern San Joaquin study unit. High relative-concentrations of organic constituents in the primary aquifer reflected high relative-concentrations of the discontinued soil fumigant 1,2-dibromo-3-chloropropane (DBCP, 2.7 percent of the primary aquifer). Maximum relative-concentrations were equal to or greater than 0.1 and (or) a detection frequency greater than 10 percent for five organic constituents: chloroform, DBCP, methyl tert-butyl ether (MTBE), simazine, and tetrachloroethylene (PCE).

The understanding assessment for inorganic constituents showed that groundwater age, normalized lateral position, and redox conditions were the most significant explanatory factors related to inorganic constituent relative-concentrations. Groundwater age was shown to be associated with relative-concentrations of arsenic, gross-alpha radioactivity, and total dissolved solid (TDS) relative-concentrations. High and moderate relative-concentrations of arsenic, iron, and manganese primarily were associated with geochemical conditions. Relative-concentrations of arsenic were high in oxygen-rich high-pH waters and in anoxic waters. High relative-concentrations of iron and manganese were most often associated with low-oxygen anoxic waters. Normalized lateral position, a sampled well’s position in the Northern San Joaquin study unit relative to the basin center and basin edge, was shown to be associated with arsenic, nitrate, and TDS relative-concentrations. High and moderate relative-concentrations of arsenic and TDS were found more frequently closer to the valley trough (basin center), where relative-concentrations of nitrate tended to decrease, than in wells near the valley margins (basin edges).

The understanding assessment for organic constituents showed that groundwater age, well depth, and land use within 500 meters of the sampled well were the most significant factors affecting organic constituent relative-concentrations. Trihalomethanes, fumigants, pesticides, and solvents were all shown to have higher relative-concentrations in young groundwater than in old groundwater. Fumigant and pesticide relative-concentrations were related to well perforation depth, with wells with shallow depths to top-of-perforation having higher constituent relative-concentrations than those with deeper depths to top-of-perforation. Detections of trihalomethanes and solvents were positively associated with urban land use and negatively associated with agricultural land use. Fumigant detections were strongly correlated with a specific agricultural land use—orchards and vineyards.

First posted September 28, 2010