Status and understanding of groundwater quality in the North San Francisco Bay groundwater basins, 2004: California GAMA Priority Basin Project

Abstract

Groundwater quality in the approximately 1,000-square-mile (2,590-square-kilometer) North San Francisco Bay study unit was investigated as part of the Priority Basin Project of the Groundwater Ambient Monitoring and Assessment (GAMA) Program. The study unit is located in northern California in Marin, Napa, and Sonoma Counties. The GAMA Priority Basin Project is being conducted by the California State Water Resources Control Board in collaboration with the U.S. Geological Survey (USGS) and the Lawrence Livermore National Laboratory. The GAMA North San Francisco Bay study was designed to provide a spatially unbiased assessment of untreated groundwater quality in the primary aquifer systems. The assessment is based on water-quality and ancillary data collected by the USGS from 89 wells in 2004 and water-quality data from the California Department of Public Health (CDPH) database. The primary aquifer systems (hereinafter referred to as primary aquifers) were defined by the depth interval of the wells listed in the CDPH database for the North San Francisco Bay study unit. The quality of groundwater in shallower or deeper water-bearing zones may differ from that in the primary aquifers; shallower groundwater may be more vulnerable to surficial contamination. The first component of this study, the status of the current quality of the groundwater resource, was assessed by using data from samples analyzed for volatile organic compounds (VOC), pesticides, and naturally occurring inorganic constituents, such as major ions and trace elements. This status assessment is intended to characterize the quality of groundwater resources within the primary aquifers of the North San Francisco Bay study unit, not the treated drinking water delivered to consumers by water purveyors. Relative-concentrations (sample concentration divided by the health- or aesthetic-based benchmark concentration) were used for evaluating groundwater quality for those constituents that have Federal and (or) California benchmarks. A relative-concentration greater than (>) 1.0 indicates a concentration above a benchmark, and less than or equal to (≤) 1.0 indicates a concentration equal to or below a benchmark. Relative-concentrations of organic and special interest constituents were classified as “high” (relative-concentration > 1.0), “moderate” (0.1 < relative-concentration ≤ 1.0), or “low” (relative-concentration ≤ 0.1). Inorganic constituent relative-concentrations were classified as “high” (relative-concentration > 1.0), “moderate” (0.5 < relative-concentration ≤ 1.0), or “low” (relative-concentration ≤ 0.5). Aquifer-scale proportion was used as a metric for evaluating regional-scale groundwater quality. High aquifer-scale proportion is defined as the percentage of the primary aquifers that have a relative-concentration greater than 1.0; proportion is calculated on an areal rather than a volumetric basis. Moderate and low aquifer-scale proportions were defined as the percentage of the primary aquifers that have moderate and low relative-concentrations, respectively. Two statistical approaches—grid-based and spatially-weighted—were used to evaluate aquifer-scale proportion for individual constituents and classes of constituents. Grid-based and spatially-weighted estimates were comparable in the North San Francisco Bay study unit (90-percent confidence intervals). For inorganic constituents with human-health benchmarks, relative-concentrations were high in 14.0 percent of the primary aquifers, moderate in 35.8 percent, and low in 50.2 percent. The high aquifer-scale proportion of inorganic constituents primarily reflected high aquifer-scale proportions of arsenic (10.0 percent), boron (4.1 percent), and lead (1.6 percent). In contrast, relative-concentrations of organic constituents (one or more) were high in 1.4 percent, moderate in 4.9 percent, and low in 93.7 percent (not detected in 64.8 percent) of the primary aquifers. The high aquifer-scale proportion of organic constituents primarily reflected high aquifer-scale proportions of PCE (1.3 percent), TCE (0.1 percent), and 1,1-dichloroethene (0.1 percent). The inorganic constituents with secondary maximum contaminant levels (SMCL), manganese and iron, had relative-concentrations that were high in 40.8 percent and 24.4 percent of the primary aquifers, respectively. Of the 255 organic and special-interest constituents analyzed for, 26 constituents were detected. Two organic constituents were frequently detected (in 10 percent or more of samples), the trihalomethane chloroform and the herbicide simazine, but both were detected at low relative-concentrations. The second component of this study, the understanding assessment, identified the natural and human factors that affect groundwater quality by evaluating land use, physical characteristics of the wells, geochemical conditions of the aquifer, and water temperature. Results from these evaluations were used to explain the occurrence and distribution of constituents in the study unit. The understanding assessment indicated that a majority of the wells that contained nitrate also had an urban or agricultural land-use classification, had a modern or mixed age classification, and had depths to their top perforations <100 ft (30 m). Geochemical data are consistent with partial denitrification of nitrate in some reducing groundwaters in the terminal and deeper parts of the flow system. High and moderate relative-concentrations of arsenic may be attributed to reductive dissolution of manganese or iron oxides, or to desorption or inhibition of arsenic sorption under alkaline conditions. Arsenic concentrations increased with increasing depth and groundwater age in the North San Francisco Bay study unit. High to moderate relative-concentrations of boron were primarily associated with hydrothermal activity or high-salinity waters in the Napa Sonoma lowlands. Simazine was detected in groundwater classified as modern and mixed age more often than in groundwater classified as pre-modern age, while chloroform was detected most often in groundwater classified as mixed age. Simazine and chloroform also were observed in wells that had surrounding land use classified as agricultural or land use classified as urban, and top of perforation depths less than 100 ft (30 m). Together, the occurrence of chloroform and simazine in shallow wells with modern or mixed groundwater located in urban or agricultural areas suggests that these constituents result from anthropogenic activities during the last 50 years. Tritium, helium-isotope, and carbon-14 data were used to classify the predominant age of groundwater samples into three categories: modern (water that has entered the aquifer in the last 50 years), pre-modern (water that entered the aquifer more than 50 years to tens of thousands of years ago), and mixed (mixtures of modern- and pre-modern-age waters). Arsenic, iron, and total dissolved solids (TDS) concentrations were significantly greater in groundwater having pre-modern-age classification than modern, suggesting that these constituents accumulate with groundwater residence time.