USGS Home
Contact USGS
Search USGS

U.S. GEOLOGICAL SURVEY
Scientific Investigations Report 2008–5045

Back to Table of Contents

Methods of Investigation

For this study, the area of interest was extended eastward to include an additional 700 mi² area beyond the Yakima River Basin to the Columbia River increasing the study area from 6,200 to 6,900 mi² (fig. 5). This extension to the Yakima River Basin was included to accommodate the larger Yakima River Basin aquifer system study area being used for the development of numerical ground-water flow models.

The hydrogeologic framework for the basalt and interbed units was compiled using information from multiple data types and data sources. Data types include well-log information from geochemical, geophysical, test hole, piezometer, and drillers’ well-log records; and interpretative information from unit contour and geologic maps compiled from multiple sources. Sources of information include well records and maps from published and unpublished data and investigations. Previously published data and investigations include: Siems and others (1973), Swanson and others (1979), Meyers and Price (1979 and 1981), Tanaka and others (1979), Biggane (1982, 1983), Drost and Whiteman (1986), Drost and others (1989, 1990, 1997), U.S. Department of Energy (1988), Lane (1988), Whiteman and others (1994), Owens (1995), Sinclair (1998), Jones and others (2006), and Vaccaro and Sumioka (2006). Unpublished data were supplied by John Kirk (Washington Department of Ecology, written commun., 2006), Steve Reidel (unpub. report, Battelle, 2006), WaDOE well records web site (http://apps.ecy.wa.gov/welllog/index.asp), and the U.S. Geological Survey unpublished well records (or non-inventoried well records). Well-record information includes data from inventoried wells (well location verified by field personnel visit) and non-inventoried wells (well location not verified, used reported information). The non-inventoried, well records were selected to help refine the hydrogeologic framework in strategic areas where available well records were sparse. Their reported locations generally are accurate within a radius of 0.25 mi of their actual location. Locations of the inventoried wells generally are accurate within a radius of several hundred feet of their actual location.

A simplified surficial-geology map (Jones and others, 2006), based on the original mapping of 12 quadrangle maps available from the Washington State Division of Geology and Earth Resources (DGER), was used in this study to assist in delineating the hydrogeologic framework and extents of the basalt and interbed units. For the study area, the combined CRBG from the simplified map was subdivided and regrouped into three basalt units: Grande Ronde, Wanapum, and Saddle Mountains. No attempt was made to reconcile matching of the surficial geologic units across originally mapped quadrangle boundaries.

Extents and depth to the top of the basalt and interbed units were constructed for each unit based on information from the (1) simplified surficial-geology map, (2) unit interpretations from previously constructed contour maps, and (3) unit interpretations from about 3,000 well records. A given unit was assumed to be continuous within the boundaries of its extent, except where an underlying unit was present at the surface or where the interpreted data indicated the unit was absent.

Information from well records and available unit contour maps were converted to digital data layers and entered into a Geographic Information System (GIS) to facilitate data comparison and analysis and to construct layers representing the extents and tops of units. These unit layers were constructed from the top down using all available data for each individual unit so the layers for each successive unit were constrained by all the overlying layers.

The accuracy of the layers depicting depth to the hydrogeologic units primarily is dependent on the methods of unit delineations. The delineations made from a geochemical analysis of a core sample, geologists’ lithologic log, or geophysical log are relatively accurate, but vary according to the complexity of the geology. The delineations made from drillers’ lithologic logs are the least accurate, particularly in structurally complex areas. About 99 percent of the unit delineations were based on interpretations of available drillers’ logs in the study area and less than 1 percent of the delineations were based on interpretations of geochemical, geologists’, or geophysical logs. An increased sampling and availability of interpreted geochemical, geologists’, and geophysical logs could help to improve and refine the hydrogeologic framework in complex areas and areas of sparse data.

Data layers for each basalt and interbed unit were interpolated to a 30-meter cell size grid to construct a digital hydrogeologic framework for the basin. Data layers for each unit included a digital elevation model (DEM), the simplified surficial geology, previously constructed hydrogeologic unit contour maps (where available), mapped extent of the hydrogeologic unit, and well-log point values of the depth to the top of the individual unit.

In developing the three-dimensional (3D) framework, the original data interpretations were honored as much as possible. Thus, the calculated depth to the top of the unit-cell values and (or) mapped contours for the hydrogeologic units were compared to the original well and (or) mapped contour data then adjusted to more accurately reflect the original interpretations. The areas where the calculated depths are less accurate are in areas with sparse data; where the surficial geology changes abruptly, with structural complexity over short distances; where the well locations are less accurate; and where unit contour intervals were more generalized. However, most discrepancies were reconciled during the 3D framework construction.

Back to Table of Contents