USGS Home
Contact USGS
Search USGS

U.S. GEOLOGICAL SURVEY
Scientific Investigations Report 2008–5025

Back to Table of Contents

Introduction

Ground water is a significant source of drinking water in Washington State (fig. 1), where more than 60 percent of the population uses ground water for domestic needs (Ron Lane, U.S. Geological Survey, oral commun., August 2007). Nitrate concentrations in ground water are elevated in parts of the State as a result of various land-use practices, including fertilizer application, dairy operations and ranching, and septic-system use (Williamson and others, 1998; Ebbert and others, 2000). Consumption of water with elevated nitrate concentrations is linked to methemoglo­binemia, or “blue baby” disorder, a potentially fatal condi­tion caused by low oxygen levels in the blood of infants (Fewtrell, 2004) and spontaneous abortion among some women (Centers for Disease Control and Prevention, 1996). High nitrate concentrations also have been linked to bladder and ovarian cancer (Weyer and others, 2001) and non-Hodgkin’s lymphoma (Ward and others, 1996).

To help assure the quality of the State’s drinking-water supply, the Washington State Department of Health (WDOH) requires that public-supply water systems regularly measure nitrate concentrations. However, such testing is only required of public-supply systems and, as a result, citizens and public health officials have only limited information about the potential exposure to elevated nitrate concentrations for people whose primary drinking-water sources are private wells. To assist public health officials, WDOH asked the U.S. Geological Survey (USGS) to develop maps that show the relation between elevated nitrate concentrations and the locations and depths of wells. This information could be used to help estimate exposure to nitrate in drinking water in areas without measurements of nitrate concentrations in ground water.

Background

Ground-water vulnerability maps are designed to estimate the potential for contamination of ground water in an area based on anthropogenic and hydrogeologic factors. Multiple definitions exist for the term “ground-water vulnerability” depending on the research objectives. For this study, we use the definition from the National Research Council (1993), which states that ground-water vulnerability to contamination is

“...the tendency or likelihood for contaminants to reach a specified position in the ground-water system after introduction at some location above the uppermost aquifer.”

The National Research Council (1993) refined the definition on the basis of whether the assessment was contaminant specific, defined as “specific vulnerability,” or for any contamination in general, “intrinsic vulnerability.”

Three previous studies, Tesoriero and others (1998), Frans (2000), and Nolan and others (2002), partially estimated ground-water vulner­ability of the State of Washington to nitrate contamination. Tesoriero and others (1998) developed a logistic regression model to estimate the probability of nitrate contamination exceeding 3 mg/L in the Puget lowland. Well depth, surficial geology, and land use (forest, urban, and agriculture) in a 3.2-km radius of a well were identified as the significantly related variables.

In Frans (2000), logistic regression models were developed to estimate the probability of nitrate contamination exceeding 3 and 10 mg/L for Grant, Franklin, and Adams Counties in eastern Washington. In those models, well casing depth, fertilizer application amounts, and the mean soil hydrologic group were identified as the significantly related variables.

Nolan and others (2002) developed a logistic regression model to estimate the probability of nitrate contamination exceeding 4 mg/L in predominantly shallow, recently recharged ground waters of the conterminous United States. The significantly related variables were nitrogen fertilizer loading, percentage of cropland as pasture, natural log of human population density, percentage of well-drained soils, depth to seasonally high water table, and presence or absence of unconsolidated sand and gravel aquifers. Although this model was applied to the conterminous United States, few data used to calibrate the model were from the State of Washington (Nolan and others, 2002).

Purpose and Scope

This report presents the results of an analysis relating anthropogenic and natural factors to the occurrence of elevated nitrate concentrations in ground water of Washington State. The probability of elevated nitrate concentrations at a particular location was estimated using a logistic regression model. The logistic regression model was entered into a geographic information system (GIS) to produce maps that display the probability of elevated nitrate concentrations throughout Washington State. In this report, nitrate concentration refers to the concentration of nitrite plus nitrate measured as nitrogen, and the term elevated nitrate concentrations refers to those concentrations that exceed 2 mg/L (as N).

Description of Study Area

The State of Washington (fig. 1) has a very diverse climate. The north-south trending Cascade Mountains divide the State into a humid western half and a semiarid eastern half. On average, the annual precipitation is about 70 in. in western Washington and 20 in. in eastern Washington. Locally, annual precipitation ranges from about 20 to 200 in. in western Washington and about 7 to 40 in. in eastern Washington.

The surficial geology of Washington also is diverse. Bedrock is exposed in the Cascade and Olympic Mountains in northeastern Washington, and the Columbia River basalt flows cover the southeastern part of the State. Unconsolidated alluvial materials fill in the bottoms of many bedrock valleys and overlie much of Columbia River basalts, and unconsolidated glacial materials cover most of the Puget lowland region.

A map of hydrogeomorphic regions was developed specifically for this study, because a highly detailed map of regional aquifers for Washington was not available (fig. 2). According to Rupert (2003),

“...hydrogeomorphic regions are similar in concept to regional aquifers, but are distinguished from regional aquifers in that hydrogeomorphic regions are delineated on the basis of general geographic locations of geologic materials and not on actual aquifer locations.”

The hydrogeomorphic regions were delineated based on the surficial geologic map of Washington (Washington Department of Natural Resources, 2005). The hydrogeomorphic regions that were delineated include the glacial material of the Puget lowland, alluvial fill in western and eastern Washington, the Columbia basalt flows, the alluvial material of the Willamette lowland, and the bedrock of northern Washington, the Cascades, and Olympics. In eastern Washington, the alluvial fill hydrogeomorphic region lies on top of the basalt and bedrock units. Therefore, wells in the eastern Washington alluvial fill region were stratified by depth to determine if the wells were deep enough to penetrate the underlying regions. Where possible, well depths were compared to sediment thickness maps (Jones and others, 2006, and Susan Loper, Franklin Conservation District, written commun., 2007) to determine if the wells penetrated the alluvial fill. If such sediment thickness maps were unavailable, wells with depths less than 150 ft were assumed to be in the alluvial fill region. Wells with depths greater than 150 ft were assumed to penetrate into the underlying regions.

Most of the population in Washington resides in the Puget lowland region. Eastern Washington is generally sparsely populated with the exception of a few cities such as Spokane.

Back to Table of Contents