USGS - science for a changing world

Oregon Water Science Center

green dotHome green dotData green dotStudies green dotPublications/Media green dotInformation

Geomorphology and Flood-Plain Vegetation of the Sprague and Lower Sycan Rivers, Klamath Basin, Oregon

By Jim E. O’Connor, Patricia F. McDowell, Pollyanna Lind, Christine G. Rasmussen, and Mackenzie K. Keith

In cooperation with the University of Oregon and the U.S. Fish and Wildlife Service


Study Area, Methods, and Data Sources

Lower Sycan River

Lower Sycan River. (Photograph by Jim O'Connor, USGS)

Analyses were based on integrated consideration of geologic and geomorphic setting, historical observations, and current conditions and processes. The approach was multifaceted, relying on mapping and stratigraphic analysis of flood-plain deposits and flanking features; evaluation of historical records and observations; systematic mapping and analysis of flood-plain and channel characteristics (including morphologic and vegetation conditions) from historical surveys, maps, and aerial photographs; and a 2006 survey of depositional features left by early 2006 high flows. This section provides a synopsis of the study area and analysis framework, including general methods and source data. More complete descriptions of the specific methods and analyses are provided in subsequent sections describing the geomorphic mapping and valley segment delineation, physical channel and flood-plain processes, and flood-plain and riparian vegetation.

The Study Area

This study focused on the lower, alluvial segments of the Sprague River system, including the lower portions of the Sycan River, North Fork Sprague River, South Fork Sprague River, and the entire 136.1 km of the Sprague River between the confluence of the North and South Forks and the Sprague River confluence with the Williamson River at Chiloquin (fig. 1). Analysis of the Sycan River focused on the lowermost 11.3 km downstream of its emergence from the Sycan River Canyon near Coyote Bucket. Analyses primarily concentrated on the channel and flood plain within an area that was mapped and defined as the “geomorphic flood plain” of the main-stem Sprague River and the lower alluvial portions of the three major tributaries. As described in more detail in the “Geomorphic History” section, the geomorphic flood plain (plate) was mapped on the basis of topography and existing soils information (Cahoon, 1985), supported by field and stratigraphic observations.

For many analyses, results and inferences were stratified by 13 valley segments subdivided from the geomorphic flood plains on the basis of valley form and major tributary junctions (table 2; fig. 1). These segments included nine encompassing the 136.1 km of main-stem Sprague River, two for the lower Sycan River, and one segment for each of portions of the South Fork Sprague and North Fork Sprague Rivers within the study area. Segment lengths (as measured along flood-plain centerline) ranged from 3.6 km to 18.8 km.

Geomorphic Mapping

Geomorphic mapping establishes the basic context for understanding modern channel conditions by defining major elements of the late Cenozoic geologic history shaping the geomorphology of the study area and outlining the geomorphic flood plain, which is the domain for assessing channel change and flood-plain vegetation conditions. Additionally, the process of defining and mapping valley-bottom landforms substantially aids in identifying key flood-plain and channel processes (for example, Coffman and others, 2011, Simenstad and others, 2011). The mapping domain broadly corresponds with the extent of LiDAR topography acquired in November 2004 (Watershed Sciences, Inc., 2005) and includes the main-stem Sycan, Sprague, and North Fork Rivers within the broad alluvial valleys. The mapping encompasses the main flood plains and contiguous alluvial and colluvial landforms (plate). The geomorphic mapping was based on aerial photographs, the 2004 LiDAR, U.S. Geological Survey 7.5-minute topographic maps, existing soil mapping (Cahoon, 1985), reconnaissance field observations, and stratigraphic sections, primarily along bank exposures but supplemented by augering. Linework was digitized at scales ranging from 1:5,000 to 1:10,000, using the 2004 LiDAR as a base.

Stratigraphic Studies

In conjunction with the geomorphic mapping, stratigraphic studies revealed conditions and processes prior to 20th century historical records. Stratigraphic sections measured at cut banks along the main-stem Sprague River and all major tributaries (appendix A) enabled documentation of depositional environments and changes in channel and flood-plain materials, mainly for the last 10,000 years. In addition, augering transects provided information on the distribution of specific deposits, such as those from the Sycan flood (Lind, 2009), as well as documentation of the elevation and age of paleochannel positions for the last 3,500 years. Stratigraphic measurements and descriptions at streamside exposures and from the augering transects included characterization of major depositional units, their sedimentary structures and textures, nature of contacts between units, pedogenic alteration and bioturbation, and interpretation of depositional environment. Deposit chronology was by radiocarbon dating of detrital and in situ organic materials (appendix B). In addition, three samples of volcanic materials were submitted for identification (appendix C). At some sites, analysis of Cesium-137 content allowed determination of deposits emplaced after widespread above-ground nuclear testing in the 1950s (appendix D).

Historical Observations and Photographs

Early written accounts of channel and flood-plain conditions are sparse because the Sprague River basin was not near major exploration and transportation routes. Aside from the General Land Office (GLO) surveys, the earliest accounts documenting channel and flood-plain conditions are from the early 1900s. These include a land classification and timber survey encompassing the Klamath Quadrangle (Leiberg, 1900; includes the study area except the area west of FK 91.8), a preliminary report on water resources for the Klamath Indian Reservation (Henshaw, 1912), and field notes of botanist W.E. Lawrence summarizing observations in July 1922 and July 1934 (W.E. Lawrence Field Journals, Oregon State University Herbarium, Corvallis, Oregon).

Early USGS records of streamflow measurement and related activities also provide information on channel and flow conditions, as well as documenting the effects of diversions and log-driving activities on the Sprague River. In particular, station notes and photographs for USGS streamflow measurement sites on the South Fork Sprague River near Bly, North Fork Sprague River near Bly, Fivemile Creek near Bly, the Sprague River near Beatty, Sycan River near Beatty, and Sprague River near Chiloquin provide evidence of channel conditions dating back to 1904, with extensive accounts for the period between 1912 and 1930 (USGS Oregon Water Science Center, Portland, Oregon, streamflow measurement station records).

Also recorded in some of the streamflow measurement station records (USGS Oregon Water Science Center, Portland, Oregon) as well as in 1939–1943 correspondence archived from Klamath Agency files (Larry Dunsmoor, Klamath Indian Nation, written commun., 2006) is evidence for decreased beaver, otter, and muskrat populations. Army Corp of Engineers and related Klamath County memoranda from the 1960s through the 1990s document levee construction and damage (Klamath County, 2008, written commun.).

19th Century Cadastral Surveys

The first systematic surveys in the Sprague River basin were the cadastral (public record) surveys by the U.S. General Land Office (GLO; now succeeded by the U.S. Bureau of Land Management). The Sprague River study area was surveyed in several different contracts spanning 1865–1892 (table 3). Most of the main-stem Sprague River was surveyed between 1866 and 1872. The Braymill and Coyote Bucket segments, and a portion of the Chiloquin Canyon segment, were surveyed for the first time in 1892.

The GLO surveys entailed laying out townships of 36 sections, with each section being 1-mile square. The exterior boundaries of the townships and interior section lines were established by surveying along lines running north-south and east-west. Where these lines cross rivers, the location (relative to monumented section corners) and width of the river were recorded in the accompanying GLO field notes. Although the resulting township maps show continuous channels, the river was surveyed only at the section and township boundary lines; with the course between lines estimated by the cartographer. Later for some townships, riverbank lines were surveyed (termed “meandering”). This was done for parts of three townships within the Sprague River study area between 1871 and 1872 (table 3). As described in detail later, these meander surveys provide more complete information on the channel path between section lines, but since these surveys typically followed the bank edges rather than the water’s edge, there is ambiguity as to whether these meander surveys closely delimit the wetted channel. Consequently, width analyses based on the GLO surveys are chiefly based on the widths determined from the boundary and section line surveys as recorded in the GLO field notes, although the meander surveys were digitized from the maps and used for some channel planform analyses.

Aerial Photographs

The earliest aerial photographs available for the Sprague River system were acquired in 1940–1942 (henceforth described as the 1940 photographs). These photographs, plus aerial photographs from 1968 and 2000, served as the primary basis to map river and flood-plain conditions for different times during the 20th century (table 4). For the 1940 and 1968 aerial photographs, contact print photographs were scanned, rectified, georeferenced, and mosaicked following the methods of Hughes and others (2006), to create digital imagery with resolution of 1-m pixels. The 2000 imagery was available as digital orthophotograph quadrangles from the U.S. Geological Survey. The 1940 aerial photographs cover most but not all of the study area, extending to FK 95.2 on the South Fork Sprague, NFFK 5.8 on the North Fork, and SYFK 13.6 on the Sycan River. Consequently, for some analyses, parts of these upstream areas are not included. Secondary photo sources included 1975 USGS orthoimagery and 2005 digital orthophotographs from the U.S. Dept. of Agriculture (table 4).

Table 4. Primary map and aerial photograph data sources.

LiDAR Topography A LiDAR data set covering most of the flood plain of the Sprague River, North Fork, South Fork and lower Sycan River was used to extract data on elevations and for morphologic interpretation. This high-resolution LiDAR was also the basis for the geomorphic mapping. The LiDAR data were collected in November 2004, and they represent normal fall low-flow conditions (Watershed Sciences, Inc., 2005). The absolute vertical accuracy of the LiDAR point data is 0.052 m deviation root mean square error (RMSE), 0.051-m standard deviation, 0.032-m median (50th percentile) absolute deviation, and 0.1057-m 95th percentile absolute deviation. We worked with a gridded elevation data set with 1-m spacing developed from the point data.

Survey of Effects of 2006 High Flows

High flows in winter and spring of water year (WY) 2006 (1 October 2005 through 30 September 2006) provided an opportunity to investigate depositional patterns associated with geomorphically effective flows. Between 23 June and 15 July 2006, we measured deposit thickness, texture, and elevation at 71 locations along the main-stem Sprague River, from FK 93.2 just downstream of the confluence of the North and South Forks of the Sprague River, to FK 17.2 at the upstream end of the Braymill segment. These measurements allowed documentation of the distribution and texture of flood deposits, as well as their distribution and characteristics relative to landform.

Mapping and GIS Analysis Framework

From the rectified and mosaicked aerial photograph sets, we mapped geomorphic features, vegetation classes, and cultural (built) features within the geomorphic flood plain (table 5). All features for most of the study area were mapped from the 1940, 1968, and 2000 aerial photographs. The 1940 aerial photographs, however, did not completely cover the upstream parts of the South Fork Sprague River (upstream of FK 95.2), North Fork Sprague River (upstream of NFFK 5.8), and Sycan River (upstream of SYFK 13.6). Supplemental maps of channel centerline position were obtained from 1975 USGS imagery forming the basis of a 1977 soil survey (Cahoon, 1985), the 2004 LiDAR, and 2005 USGS digital orthophotographs (table 4). On-screen digitizing of these flood-plain features was at scale of 1:2,000, although finer resolution inspection and field checking were required for identification and confirmation of some mapping classes and boundaries.

We mapped the water surface and any exposed bars, as visible on each aerial photograph. The Sprague River channel has well defined, steep banks, and there is no evident active channel extending beyond the water and bar surfaces in most locations. Exposed bars are rare and add only a small area to the overall channel area. Therefore in analysis of channel width and area, we used the extent of water to represent the channel.

All map data were compiled into a geodatabase for analysis (available at U.S. Geological Survey, 2013: Aside from the broadscale geomorphic surface mapping, the analysis domain was restricted to the geomorphic flood plains of the Sprague River, and the Sycan, North Fork Sprague, and South Fork Sprague Rivers within the broad alluvial valleys of the study area. Spatial and temporal metrics of channel and flood-plain characteristics were assessed by segment and more continuously by adopting measurement frameworks based on channel and flood-plain centerlines, following the approach of O’Connor and others (2003c).

A flood-plain centerline frame of reference, which was static for the entire analysis period, simplified comparison among different years and was used to evaluate overall flood-plain characteristics and channel migration rates. Channel centerlines were mapped for each source data set and were the basis for determining channel migration rates as well as providing a frame of reference for measuring channel characteristics such as width, bar frequency, and near-channel riparian conditions. For both the flood-plain and channel centerline reference frames, intersection points were created along the centerline at 200-m spacing. At each intersection point, transects perpendicular to the centerline were defined, and measurements were extracted from points along the transect line or, for some attributes, from flood-plain polygons within the area defined by adjacent transect lines and the geomorphic flood-plain boundary. These flood-plain polygons were used to extract areas of sand and gravel bars, water surface, vegetation types, and other flood-plain attributes. The flood-plain transects extended to the limits of the geomorphic flood plain. The channel transects extended 50 m in each direction from the channel centerline. To simplify analysis, the flood-plain centerline was smoothed in highly sinuous valley sections such that the resulting transects did not cross each other. No such smoothing was conducted for the channel centerlines.


For more information, contact:

Director, Oregon Water Science Center
U.S Geological Survey
2130 SW 5th Ave
Portland, OR 97201
Phone: (503) 251-3200

Part or all of this report is presented in Portable Document Format (PDF); the latest version of Adobe Reader or similar software is required to view it. Download the latest version of Adobe Reader, free of charge.

Accessibility FOIA Privacy Policies and Notices logo U.S. Department of the Interior | U.S. Geological Survey
Page Contact Information: Oregon Webteam
Page Last Modified: Tuesday, 11-Jun-2013 20:09:13 EDT