Identifying sources of groundwater nitrate contamination in a large alluvial groundwater basin with highly diversified intensive agricultural production

Highlights

• 200 wells were sampled for nitrate across a large basin with intensive agriculture.

• 44% of samples did not meet the USEPA drinking water standard for nitrate.

• Landuse, soil, water level depth, and well specs were considered as factors.

• Tree-fruit and nuts, citrus, dairies/forage, and urban show most elevated nitrate.

• Depth to groundwater and well depth critically control nitrate in well water.

Abstract

Groundwater quality is a concern in alluvial aquifers underlying agricultural areas worldwide. Nitrate from land applied fertilizers or from animal waste can leach to groundwater and contaminate drinking water resources. The San Joaquin Valley, California, is an example of an agricultural landscape with a large diversity of field, vegetable, tree, nut, and citrus crops, but also confined animal feeding operations (CAFOs, here mostly dairies) that generate, store, and land apply large amounts of liquid manure. As in other such regions around the world, the rural population in the San Joaquin Valley relies almost exclusively on shallow domestic wells (≤ 150 m deep), of which many have been affected by nitrate. Variability in crops, soil type, and depth to groundwater contribute to large variability in nitrate occurrence across the underlying aquifer system. The role of these factors in controlling groundwater nitrate contamination levels is examined. Two hundred domestic wells were sampled in two sub-regions of the San Joaquin Valley, Stanislaus and Merced (Stan/Mer) and Tulare and Kings (Tul/Kings) Counties. Forty six percent of well water samples in Tul/Kings and 42% of well water samples in Stan/Mer exceeded the MCL for nitrate (10 mg/L NO₃-N). For statistical analysis of nitrate contamination, 78 crop and landuse types were considered by grouping them into ten categories (CAFO, citrus, deciduous fruits and nuts, field crops, forage, native, pasture, truck crops, urban, and vineyards). Vadose zone thickness, soil type, well construction information, well proximity to dairies, and dominant landuse near the well were considered. In the Stan/Mer area, elevated nitrate levels in domestic wells most strongly correlate with the combination of very shallow (≤ 21 m) water table and the presence of either CAFO derived animal waste applications or deciduous fruit and nut crops (synthetic fertilizer applications). In Tulare County, statistical data indicate that elevated nitrate levels in domestic well water are most strongly associated with citrus orchards when located in areas with a very shallow (≤ 21 m) water table. Kings County had relatively few nitrate MCL exceedances in domestic wells, probably due to the deeper water table in Kings County.

Introduction

Elevated nitrate levels (more than 2 mg/L NO₃-N) in groundwater used as drinking water have been linked to adverse health effects (Mueller and Helsel, 1996). Consumption of water containing elevated levels of nitrate can cause low blood oxygen in infants, a condition known as methemoglobinemia or “blue baby syndrome”. Methemoglobinemia was the impetus behind the United States Environmental Protection Agency (USEPA) maximum contamination level (MCL) of 10 mg/L NO₃-N (Mueller and Helsel, 1996). Nitrate in drinking water has also been linked to cancer through the formation of carcinogenic N-nitroso compounds (Weyer et al., 2001), to spontaneous abortions (Centers for Disease Control and Prevention, 1996), and to non-Hodgkin's lymphoma (Ward et al., 1996).

Nitrate occurs naturally in groundwater. However, septic leakage, nitrogen fertilizers, and animal manure applied to soil can cause elevated levels of nitrate in groundwater (Owens et al., 1992). High groundwater nitrate has been positively correlated with surrounding agricultural landuse (Vowinkel and Tapper, 1995). In the San Joaquin Valley (SJV) (Fig. 1) as much as 88 kg N/ha/year may leach to groundwater in areas where fertilizers are applied (Harter, 2009). Leaching from dairy corrals, ponds, and from fields receiving manure may be as high as 872 kg/ha/year, 807 kg/ha/year and 486 kg/ha/year, respectively (van der Schans et al., 2009). Increasing trends in nitrate levels in SJV groundwater during the 1950s and 1960s and from the 1970s to 1980s correlated with an increase in fertilizer and manure use, and an increase in confined animal feeding operations (CAFOs) in the SJV over the same time period (Dubrovsky et al., 1998).

Approximately two-thirds of the SJV landscape is in agricultural production (Burow et al., 2008a). More than 250 unique crops are grown in the SJV. It is home to three-quarters of California's dairy herd. The annual gross value of agricultural production in the SJV is more than $25 billion (United States Environmental Protection Agency, 2012). Irrigation water is supplied by both surface water and groundwater, while groundwater is the almost exclusive source of drinking water in rural and embedded urban areas such as Stockton, Modesto, Fresno, Tulare, and Bakersfield (Burow et al., 1998b). Total population for the eight counties in the SJV (Fresno, San Joaquin, Kern, Stanislaus, Tulare, Merced, Kings, and Madera) in 2006 was nearly 3.9 million (California Department of Finance, 2006).

Nitrate contamination of shallow groundwater (≤ 150 m deep) in the SJV is well documented. Twenty groundwater study units, distributed throughout the nation, were compared as a part of the U.S. Geological Survey (USGS) National Water Quality Assessment Program (NAWQA). Among the twenty NAWQA study units, the SJV (also referred to as the San Joaquin–Tulare Lake Basin) had nitrate concentrations in groundwater above the national median (Dubrovsky et al., 1998). The 2006 California State Water Resources Control Board (SWRCB) Groundwater Ambient Monitoring and Assessment Program (GAMA) study of 181 domestic wells in Tulare County (including wells located in the foothills outside the SJV) found that 40% of well water samples exceeded the nitrate MCL (California State Water Resources Control Board, 2010). A similar study conducted in Merced County in 2001 on 40 domestic wells found 63% to exceed the MCL for nitrate (Harter and Romesser, 2001).

Previous studies conducted in agricultural areas overlying unconsolidated aquifers determined a significant relationship between crop type or landuse within circular well buffer zones centered on sampled wells and well water nitrate (Burow et al., 1998a, Kolpin, 1996, McLay et al., 2001). However, previous studies, typically including 50 to 100 well sites, have been limited to relatively few crop type and landuse classifications (Burow et al., 1998a, McLay et al., 2001) or overarching categories such as “irrigated agriculture” (Kolpin, 1996). Studies have also shown that nitrate in groundwater can be affected by vadose zone thickness (Burow et al., 1998b) and soil type (Burow et al., 1998a) and that nitrate in well water samples can be affected by well construction characteristics such as well depth (Burow et al., 1998b).

This study expands on previous work using a larger sample size across a wider diversity of agricultural crops and landuses. The goal of this study is to determine how various landuses affect groundwater nitrate and how other factors, such as well depth, may play a role in the amount of nitrate found in well water samples. Specifically, we consider 78 crop and landuse types (grouped into 10 categories), proximity to dairies, vadose zone thickness, soil type, and well construction characteristics.