Jump to main content or area navigation.

Contact Us


Channel Processes: River Classification

Note: If you are unfamiliar with the Rosgen stream classification system or would like to review it again, go to the one-hour tutorial on this classification at the Watershed Academy Web training site.

Since streams in their stable form take on many various combinations of dimension, pattern, profile, and materials amidst a wide range of valley slopes, sediment size, sediment load, and streamflow, a stream classification system is used to stratify and describe the various river types (Rosgen 1994, 1996). This river classification system integrates individual variables into a morphological description that combines various forms of the existing as well as "probable state" variables. The objectives of this stream classification scheme are:

  1. Predict a river's behavior from its appearance, based on documentation of similar response from similar types for imposed conditions,
  2. Stratify empirical hydraulic and sediment relations by stream type by state (condition),
  3. Provide a mechanism to extrapolate site-specific morphological data,
  4. Describe physical stream relations to complement biological inventory and assist in establishing potential states, and
  5. Provide a consistent frame of reference for communicating stream morphology and condition among a variety of disciplines.

Examples of broad level stream morphology (broad level geomorphic characterization) that are shown in Figure 14 (PDF, 192 kb, 1 p.) and summarized in Table 2 (PDF, 168 kb, 1 p.)(Rosgen, 1994,1996). A more detailed stream classification (morphological description) is shown in Figure 15 (PDF, 210 kb, 1 p.) (Rosgen, 1994, 1996).

Reference reach data that represent the stable form of various morphological types are used in WARSSS to compare degree of departure (from reference) of the assessed channel's attributes (Rosgen 1998, 1999, 2001b and 2001c).

The classification system is developed with an understanding that stability evaluation has to be conducted at a higher degree of resolution than morphological classification. The channel stability assessment, however, must be stratified by stream type for extrapolation purposes. Barbour et al. (1991) commented on the role of stream classification in bio-assessment as follows:

About PDF Files...
Many of the documents listed on this site are PDF files. Viewing a PDF file requires use of Adobe's free Acrobat Reader software.  *EPA's PDF page  provides information on downloading the software.

"One function of classification is to increase the resolving power or sensitivity to biological surveys to detect impairment by partitioning variation within selected environmental parameters or among sites. The importance of minimizing variation… clearly it is easier to distinguish impairment if the parameters have low variability. Formal statistical tests (parametric and nonparametric) indicate greater resolution and power exist if there is low variance within elements being compared. Effective classification leads to improving resolving power by partitioning and accounting for variability. A coarse classification yields higher variance and therefore lower resolving power; vice versa for finer classifications."

Reference reach data describe the stable morphological form of dimension, pattern and profile for a range of valley slopes, channel materials, riparian vegetation and other measurable variables (Rosgen 1998). Dimensionless ratios are developed from reference reach data as stratified by stream types in order to quantitatively describe specific channel attributes. Examples are the dimensionless parameters of depth and slope of bed features such as pools, runs, riffles, and glides. Dimensionless ratio relations of meander and hydraulic geometry are also developed by stream type for the purpose of determining departure as well as providing information for natural channel design (Rosgen 1998). The advantage of developing dimensionless ratio relations is for extrapolation to rivers of similar channel and valley types but different sizes.

Applications of regime equations are most appropriate if the river where the data were empirically derived is similar to the river to which it is applied. Stratification of such relations by stream type/condition would provide insight as to how the relations were developed. For example, the relation of stream width/discharge as influenced by bank stability as shown in Figure 16 and Figure 17 (Copeland et al. 2001) assumes there are no differences in the width/discharge relation due to morphological stream types.

Figure 16

Figure 16. Downstream width hydraulic geometry for United Kingdom gravel bed rivers, W = a Qb0.5 with confidence bands. Based on 36 sites in the United Kingdom with erodible banks. S.I. units, m and m3/sec (English units, ft and ft3/sec). (Copeland et al., 2001)

Figure 17

Figure 17. Downstream width hydraulic geometry for United Kingdom gravel bed rivers, W = a Qb0.5 with confidence bands. Based on 43 sites in the United Kingdom with resistant banks. S.I. units, m and m3/sec (English units, ft and ft3/sec). (Copeland et al, 2001)

Hydraulic geometry relations as predicted by regime equations can be improved by the integration of stream type data. The variability of hydraulic geometry relations such as width versus discharge can be minimized by stratification by stream type. For example, a stable E4 stream type that has a width/depth ratio of 3 (Figure 18) has a much lower width for the same discharge than a C4 stream type which has a stable width/depth ratio of 20 (Figure 19).

Figure 18

Figure 18. Example of a typical E4 stream type. (photo by J.D. Kurz)

Figure 19

Figure 19. Example of C4 stream type similar in size to the river in Figure 18.

Both streams are gravel-bed, low gradient, meandering streams with floodplains and both are stable, but their widths for the same discharge are much different. Smaller rivers that exhibit the same small width/depth ratios for similar flows are shown in the E4 stream type (Figure 20) and for a higher width/depth ratio C4 (Figure 21).

Figure 20

Figure 20. An E4 stream type with a bankfull width of 4 feet for a bankfull discharge of 75 cfs.

Figure 21

Figure 21. A stable C4 stream type with a bankfull width of 15 feet (w/d ratio 14) for approximately 80 cfs.

Even though both rivers are stable, they have much different widths for the same discharge. The data points as shown in Figure 22 for North American gravel-bed rivers (Copeland et al. 2001) show great variability and a range of widths from 5-30 meters for a bankfull discharge of 10 cms.

Figure 22

Figure 22. Downstream width hydraulic geometry for North American gravel bed rivers, W=3.68 Qb0.5, and U.K. gravel bed rivers, W=2.99 Qb0.5 (Copeland et al, 2001).

Much of the observed differences in width for the same discharge can be explained by morphological stream type. The relations of width/discharge in Figures 16, 17, and 22 may be improved if stratified by stream type or at least by width/depth ratio, a delineative criterion for stream classification.

Hydraulic geometry relations including width are presented for various stream types of the same size, depicting large differences in width for the same discharge (Figure 23) (Rosgen 1994, 1996). This relation contrasts the differences in width for the same discharge between an E3 (cobble) and C3 stream type, both being stable channels.

Figure 23 Figure 23 Figure 23 Figure 23

Figure 23. Hydraulic geometry relations for selected stream types of uniform size (Rosgen, 1994,1996).

To General Principles Home Page | Go to Type & Stability

Jump to main content.