Introduction

The Johnson Creek basin is located in northwestern Oregon on the eastern side of the Portland metropolitan area in Multnomah and Clackamas Counties. Johnson Creek is about 24 mi long and has a surface-water drainage basin that covers an area of 54 mi². The surface-water drainage basin is defined by the surface-water flow divide—a ridge in the land surface from which surface water moves away in both directions perpendicular to the ridge line. Johnson Creek flows generally westward from agricultural areas near Boring, through residential and industrial areas in the cities of Gresham and Portland, and enters the Willamette River in the city of Milwaukie (pl. 1, fig. 1).

Record rainfall in water years (WY) 1996 and 1997 resulted in two types of flooding in the Johnson Creek basin. Intense rainfall in February and November 1996 led to flooding of Johnson Creek. The cumulative effect of 2 years of high annual precipitation led to groundwater flooding, manifested in high streamflow of Crystal Springs Creek—an entirely groundwater fed tributary to Johnson Creek; and an elevated water table in the vicinity of Holgate Lake. Although the flooding of Johnson Creek lasted on a time scale of several days, the groundwater flooding persisted for months to years. These flooding events revealed the interaction of the groundwater and surface-water systems, and the need for better understanding of the hydrology of the Johnson Creek basin.

In addition to better understand specific flooding events, information was needed about spatial and temporal trends in streamflow, and the interaction of the groundwater system and streams in the basin at low flow for the protection of water quality. The U.S. Geological Survey (USGS) began a study in 1997 with the city of Portland. In 2000, the cities of Gresham and Milwaukie, and Clackamas County’s Water Environment Services joined the cooperative study, followed in 2002 by Multnomah County.

The Johnson Creek basin has a temperate marine climate in which winters typically are wet and summers are dry. About two-thirds of the annual precipitation is from November through March, and less than 10 percent from July through September. Most precipitation falls as rain. Although snow falls several times each year, accumulation is uncommon and usually melts rapidly during subsequent rain storms. Annual precipitation in the Johnson Creek basin increases to the southeast from about 40 in. at the mouth of the creek to more than 60 in. in the upland area of the basin. Average annual precipitation (WY 1911–2006) in Portland is about 37 in. (Oregon Climate Service, 2007).

Hydrologic Setting

The Johnson Creek basin is within the southern end of the Portland basin, a large topographic and structural depression that includes most of the Portland metropolitan area in northwestern Oregon and that extends across the Columbia River into southwestern Washington. The Johnson Creek basin is bounded on the south by the Clackamas River and Kellogg Creek basins. A subtle surface-water divide separates the Johnson Creek basin from the Sandy River basin to the northeast and the basins of the Columbia River and Fairview Creek to the north.

The hydrogeologic setting of much of the Portland metropolitan area, including the Johnson Creek basin, is summarized in Snyder (2008, p. 8-9). The study area can be divided into three areas where geology and soils have important effects on groundwater discharge and streamflow:

1. A group of volcanic buttes that are the western extent of Cascade volcanoes arise on the south side of the basin. The highest elevation is about 1,100 ft (Trimble, 1963). The largest tributary to Johnson Creek in the area is Kelley Creek. The greatest component of soils is the Cascade silt loam group (U.S. Department of Agriculture Natural Resources Conservation Service, 2007). The steep slopes and silty soils contribute to low infiltration and rapid runoff.
2. Consolidated sands and gravels underlie the Gresham area and the lowlands on the southern side of the creek east of I-205, which is Troutdale Formation as identified in McFarland and Morgan (1996, p. 9-11) and the “middle sedimentary unit” as identified by Conlon and others (2005, p. 15). Soils in the area are Powell and Cascade silt loams. The runoff characteristics of these soil types are similar to those in the area of the buttes, although the terrain is less steeply sloped.
3. The northern side of the basin west of Gresham and the southern side of the basin west of I-205 is a distinct contrast to the more steeply sloped, less-permeable areas to the south and east. Except for isolated volcanic buttes, the landscape is relatively flat and has terrace deposits that are identified as the “unconsolidated sedimentary aquifer” (McFarland and Morgan, 1996, p. 9-10) and the upper sedimentary unit (Conlon and others, 2005, p. 9). This area of the basin was shaped by Missoula Floods after the last ice age, leaving a coarse mix of fluvial deposits (Waitt, 1985). Although surface drainage comes from the buttes on the northern area of the Johnson Creek basin, distinct stream channels from these features generally do not extend to Johnson Creek. The topographic setting and the highly permeable deposits explains the minimal stream-channel development in this area of the Johnson Creek basin (Hogenson and Foxworthy, 1965, p. 11).

Except for Crystal Springs Creek and other spring inflows in the lower area of the Johnson Creek basin, no known perennial tributaries enter Johnson Creek from the north. Soils in the area are Multnomah and Latourelle loams, and some soils in areas near the creek midbasin are Wapato silt loam. In the lower area of the basin, Johnson Creek cuts through the terrace deposits as it flows toward the Willamette River. Near the confluence with Crystal Springs Creek, Johnson Creek enters an area of a former alluvial channel of the Clackamas River, the Willamette River, or both (Hogenson and Foxworthy, 1965, p. 10-11, 28). The creek turns south in the lower 1 mi because of a bedrock outcrop near the confluence with the Willamette River.

Precipitation over the basin follows several pathways to groundwater and surface-water systems. In the eastern basin, delivery of stormwater to Johnson Creek is expedited by drains in agricultural areas, runoff from roads, ditches, and other hard surfaces, resulting in a rapid rise in response to precipitation. In the lower basin, some runoff is directed to the subsurface through drywells and to the sewer system that discharges outside the basin, resulting in lessened contribution to peak flows from the urban area. Summer flows of Johnson Creek are low through much of the basin because of low precipitation and, in places, minimal groundwater discharge. In some areas, regional groundwater flow paths lead away from Johnson Creek and toward nearby large rivers. In the lower basin, summer flow of Johnson Creek is sustained by groundwater discharge.

About three-fourths of the annual flow of Johnson Creek is from November through March, and less than 10 percent of the annual flow is in July through September. High-flow events typically begin with intense rainfall over 1 to 2 days during which prestorm conditions may include saturated soils, snow, or both.

Land Use

Land use varies in the Johnson Creek basin, and generally follows a gradient from forested, agricultural, and rural-residential areas in the headwaters to urban and industrial areas in the lower basin. Land uses on the southern side of the basin include forest, nurseries, and light to dense residential. The northern side of the basin within the cities of Gresham and Portland is densely populated. Tanner and Lee (2004, p. 4, 6-7) discuss general land-cover categories for areas upstream of several monitoring sites on Johnson Creek.

The Johnson Creek basin, parts of which are in urban, suburban, and agricultural settings, has been affected by human occupation. Channel simplification, installation of storm drains, and paving over surfaces have affected the hydrologic character of the stream. In recent years, development has resulted in conversion of agricultural and light-residential areas to densely urban land uses, particularly in the southern and eastern areas of the Johnson Creek basin. The stream channel of the lower 11 mi of Johnson Creek was altered in a project of the Works Progress Administration in the 1930s. The primary purpose was flood control, and consisted of limiting the stream to a single, trapezoidal, masonry lined channel. Data collection in the basin did not begin until after the completion of this project. Although a systematic analysis of land use changes through time is beyond the scope of this report, the general grid of streets, drains, and ditches probably already was established by the 1940s.

Management of storm runoff in the basin varies considerably. As such, the fate of precipitation and subsequent runoff and/or groundwater recharge can differ, even within the same land-use category. For example, some agricultural areas are drained by a network of ditches and subsurface tile drains that facilitate rapid delivery of precipitation to the stream, whereas runoff from other areas may be managed by onsite infiltration through swales and ponds. Likewise, in urban areas, runoff can be routed out of the basin entirely through a combined sewer system (where sanitary sewage and stormwater runoff are combined in a single pipe), directed to the stream through storm-drain networks, or directed to onsite stormwater drainage systems that are designed to allow for the infiltration of stormwater. Stormwater drainage systems designed to divert stormwater runoff into the subsurface include underground injection control (UIC) systems (for example, stormwater injection systems, sumps, and drywells) and vegetated swales, pervious pavement, disconnected downspouts, and other diversion methods that are designed to allow for the infiltration of stormwater.

Previous Investigations

Previous investigations by the USGS in the Johnson Creek basin fall in three general categories: groundwater and geologic studies, surface-water studies including hydraulic and hydrologic modeling, and water-quality investigations. Although some results of these studies do not relate directly to the analyses in this report, most results contain relevant observations on the hydrology of the basin. For example, chemical characteristics reported in a water-quality study may offer new relevance in understanding the chemical signature of groundwater discharge to the stream.

Previous groundwater studies focused on the Portland basin, in which the Johnson Creek basin is located, and were the foundation for a regional groundwater flow model developed to identify the groundwater flow system and to provide a tool for groundwater managers to use in evaluating development scenarios. These studies and others are listed in Snyder (2008, p. 6, 33-37), a study estimating water-table elevation and unsaturated zone thickness throughout much of the Portland metropolitan area, including the Johnson Creek basin.

Johnson Creek has been included in several surface-water studies, ranging in scope from the Portland metropolitan area to statewide. Laenen (1980) provided equations to estimate peak flow based on basin characteristics of the Portland urban area. Moffatt and others (1990) and Wellman and others (1993) provided statistical summaries of daily and peak streamflow data for streams throughout Oregon. Lee (1995) used dye-tracer methods to determine time of travel and dispersion characteristics of Willamette basin streams. Laenen and Risley (1997) constructed precipitation-runoff and streamflow-routing models of streams in the Willamette basin, and Lee and Risley (2002) estimated base flow of streams in the Willamette basin.

Water-quality studies include analyses and interpretation of whole-water, filtered, and bed sediment data. Edwards (1992), Edwards and Curtiss (1993), and Edwards (1994) provided analyses and interpretation, specific to the Johnson Creek basin, of physical characteristics (pH, dissolved oxygen, alkalinity, and specific conductance), nutrient, trace element, manmade organic compound, and sediment data during high- and low-flow periods. Johnson Creek was sampled in a Willamette River basin study for trace elements, organic compounds, and suspended sediment (Anderson and others, 1996). McCarthy and Gale (1999) used semipermeable membrane devices to assess the distribution of organic compound in a Columbia basin study. Finally, Tanner and Lee (2004) reported on organochlorine pesticides in whole-water samples from several locations on Johnson and Kelley Creeks, including changes in the relation of suspended sediment to pesticides over time.

Purpose and Scope

This report provides a quantitative description of the groundwater and surface-water hydrology of the Johnson Creek basin, including information about the interaction of the groundwater and surface-water systems, groundwater discharge to springs, surface manifestations of the water table, and groundwater influence on streamflow and stream temperature. This report also includes findings on spatial and temporal trends in annual, low, and high streamflow.

The results build upon previous investigations and long-term data. The previous investigations consist of a USGS regional groundwater study on the Portland metropolitan area and hydrologic and statistical studies of streamflow of Johnson Creek. The description of the groundwater system in this report is more general in nature. Conclusions of these investigations were derived from hydrogeologic mapping, well tests, groundwater level data, streamflow, and water-quality data. New data included measurements of groundwater level, streamflow, specific conductance, and stream temperature. Historic groundwater, streamflow, and climate data provided decade-scale context to the dynamics of the hydrologic system. New data collection and analyses provided explanation of specific hydrologic phenomena, particularly related to the interactions of the groundwater and surface-water systems.