Kansas Water Science Center
Fluvial Geomorphology Studies in Kansas
An understanding of river and stream-channel geomorphic responses to various human-caused and natural disturbances is important for effective management, conservation, and rehabilitation of rivers and streams to accommodate multiple, often conflicting, needs. For example, channel changes may have implications for various needs including the protection of property and structures, water supply, navigation, and habitat. The channel-bank erosion that accompanies natural channel migration across a flood plain represents a constant threat to property and structures located in or near the channel (see photo). Various human-caused and natural disturbances introduce additional instability to which rivers and streams will adjust. Human-caused disturbances include reservoirs, channelization, in-channel sand and gravel extraction, and urbanization. A common natural disturbance is a flood. Possible geomorphic responses of a channel to disturbances include channel-bed degradation (erosion), channel-bed aggradation (deposition of material), and (or) channel widening. These adjustments represent the channel's attempt to establish a new approximate equilibrium condition.
Channel adjustments are a concern for several reasons. A significant lowering of the channel bed poses an immediate threat to bridge pier foundations as well as buried pipelines and cables. In addition, significant bed lowering increases bank height and bank instability that may trigger channel widening. Channel aggradation raises the bed elevation, reduces channel capacity, and increases the likelihood of flooding. Any channel-bed changes that occur on the main-stem rivers and streams also may migrate upstream on the tributaries where additional property, structures, and habitat may be at risk. Finally, any long-term channel adjustment processes also may instigate or worsen local scour problems.
Geomorphic investigations by the U.S. Geological Survey, in cooperation with various agencies, are providing some of the information needed to better understand and manage river and stream systems in Kansas.
The USGS has completed several studies that have investigated the response of river and stream channels in Kansas to a variety of human-caused and natural disturbances.
Reservoir construction and operation can have a substantial effect on the stability of the river channel downstream from the dam. Reservoirs can trap and permanently store virtually the entire sediment load delivered from the upstream basin (Brune, 1953; Petts, 1979; Williams and Wolman, 1984). Thus, immediately downstream from a dam, a river's sediment load is greatly reduced. In addition, typical downstream changes in the flow regime include a reduction in the magnitude of peak flows and a possible increase in the magnitude of low flows (Williams and Wolman, 1984). In response, the downstream river may adjust in an attempt to re-establish an approximate equilibrium between the channel and the discharge and sediment load being transported. Possible adjustments include channel-bed erosion or deposition, channel widening or narrowing, and changes in channel pattern or shape.
Streamgage data and multi-date aerial photography were used to assess the stability of the Neosho River downstream from John Redmond Reservoir (completed in 1964) in east-central Kansas in response to concerns that the reservoir was causing channel widening. For two gages located downstream from the dam, long-term data on channel width, channel cross-sectional area and flow velocity representing a range of in-channel flows were plotted against discharge in successive 5-year intervals. For the near-dam gage, located about 5 miles downstream, an initial period of modest channel widening followed by stability was indicated. At the downstream gage, located about 56 miles from the dam, no change in width was apparent; however, a decrease in channel cross-sectional area associated with an increase in flow velocity was indicated. The aerial photography also indicated minimal post-dam changes in channel width (Juracek, 2000).
John Redmond Dam on the Neosho River near Burlington, Kansas. (photo by Kyle Juracek, USGS)
An investigation of channel-bed elevation change downstream from 24 large reservoirs in Kansas indicated that channel-bed erosion was a common geomorphic response (table 1). The amount of channel-bed erosion ranged from negligible for a few sites to a maximum of about nine feet for the Republican River downstream from Milford Dam in northeast Kansas. The gage sites for which channel-bed erosion of at least one foot was indicated were located within four miles downstream of the dam (Juracek, 2001).
An interesting example was the Smoky Hill River in central Kansas for which both a downstream and upstream geomorphic response to the presence of Kanopolis Lake (completed in 1948) was indicated. Before the dam was completed, a downstream increase in the stage (i.e.,elevation of the water surface) for the mean annual discharge (hereafter referred to as the reference stage) of about one foot indicated deposition that may have been a result of the disturbance (and associated increased sediment load) caused by the construction of the dam. Once the dam was completed, the downstream reference stage decreased about six feet with most of the indicated erosion occurring during the first 30 years. During that 30-year period, the erosion was interrupted by an interlude of modest deposition that occurred during a prolonged drought in the 1950s (fig. 1). Upstream from Kanopolis Lake, changes in the reference stage indicated that the channel bed slowly aggraded for several decades before stabilizing (fig. 2) (Juracek, 2001). The aggradation likely was a response to the artificial base level created by the reservoir.
Channelization typically involves several types of channel modification including realignment, widening, deepening, and straightening. Typically, channelization results in channel shortening that increases channel slope and flow velocity which can cause substantial channel degradation. The degradation can migrate long distances upstream from the original site of disturbance. In addition, channel aggradation may occur downstream.
Juracek (2004) investigated the geomorphic response of Soldier Creek, Kansas, to channelization using streamgage data. In the late 1950s, about 10 miles of the stream was channelized for the purpose of flood control in the vicinity of Topeka, Kansas. The project, which was completed in 1961, involved channel realignment, widening, deepening, and straightening. The channelization-caused disturbance resulted in channel-bed degradation that migrated several miles upstream.
At a gage located about 12 miles upstream from the upstream end of the channelized section of Soldier Creek, changes in the reference stage clearly indicated the geomorphic response of the channel to the downstream channelization (fig. 3). Initially, from 1958 to 1970, a relatively stable channel bed was indicated as the reference stage fluctuated in response to scour and fill processes. Following the construction of a low-water crossing downstream from the gage in 1971, the reference stage increased about 1.5 feet indicating deposition. Then, following the partial washout of the low-water crossing in 1978, the reference stage decreased about 6.6 feet by 1999 indicating a long-term period of channel-bed erosion (fig. 3). From 1978 to 1999, the decrease in reference stage indicated a channel-bed degradation rate of about 0.3 feet per year.
A question that sometimes arises is whether a decrease in stage with time for a specific discharge is caused by channel-bed degradation or another cause such as channel widening or increased flow velocity. In this case, the pronounced channel-bed degradation indicated by the progressive change in the stage-discharge relation (fig. 3) was confirmed by an assessment of temporal changes in the width-discharge relation for a range of in-channel flows. Specifically, a substantial decrease in the widths associated with the higher discharges (fig. 4) indicated that the channel had entrenched.
The correspondence of the beginning of channel-bed degradation with the partial washout of the low-water crossing indicated that the crossing temporarily may have prevented the upstream migration of channel-bed degradation that would have reached the gage site as early as 1971 or as late as 1978. This translates to a total travel time of 10 to 17 years for the degradation to migrate from the upstream end of the channelized section of Soldier Creek (located 12 miles downstream) to the gage site. Thus, the degradation migrated upstream at an average rate of 0.7 to 1.2 miles per year (Juracek, 2004).
The geomorphic effectiveness of a flood can be defined as the amount of channel morphological change caused by the flood and the subsequent time required for the channel to recover (Wolman and Gerson, 1978). Geomorphic effects caused by floods, which may range from negligible to substantial, include channel widening, channel-bed erosion or deposition, and channel straightening (i.e., avulsion) (Baker, 1988; Knighton, 1998). Factors that can determine geomorphic effectiveness include channel bed and bank composition, channel morphology, channel slope, valley confinement, sediment load, flood duration, stream power, the temporal ordering of floods, climate, and vegetation (Baker, 1988; Kochel, 1988; Costa and O'Connor, 1995; Osterkamp and Friedman, 2000; Emmett and Wolman, 2001; Fuller, 2007).
The geomorphic effects (short-term change and subsequent recovery) of the record-setting 1951 floods in eastern Kansas were assessed using streamgage data for 23 sites in a study by Bowen and Juracek (in press). For the sites studied, the 1951 flood was the largest discharge for the period of record at least through the year 2000. Flood-related, channel-bed elevation change was indicated at 17 sites and a substantial increase or decrease in bed elevation was indicated for seven sites. Substantial channel widening was indicated for three sites.
At a gage located along the Kansas River at Wamego, Kansas, a sudden drop in the reference stage indicated that the channel bed was eroded about 1.5 feet as a result of the 1951 flood. Subsequently, the channel bed never recovered to its pre-flood elevation (fig. 5).
At another gage located along the Kansas River at Lecompton, Kansas, channel widening of 150 feet or more (about a 20 percent increase) was indicated for moderate flows following the 1951 flood. Subsequently, during the late 1950s, the channel width for moderate flows decreased by as much as 50 to 80 feet (fig. 6). Due to a lack of high flows during the mid-1950s, impacts at higher flows could not be assessed. Channel width then stabilized and remained relatively stable at least through 2007 (data not shown). Thus, channel width did not completely recover to its pre-flood width.
John Redmond Reservoir, Kansas: Sedimentation, sediment quality, and upstream channel stability, 2009-2012.
In response to concerns about sedimentation in John Redmond Reservoir, this study was undertaken in cooperation with the U.S. Army Corps of Engineers with objectives being to: (a) determine the amount and quality of sediment deposited in the reservoir, (b) assess channel stability at selected sites upstream from the reservoir, (c) characterize sediment transport through an area of streambank stabilization, and (d) characterize the potential of altered reservoir management to decrease sediment accumulation in the reservoir.
For additional information contact: Kyle Juracek (firstname.lastname@example.org)