Jump to main content or area navigation.

Contact Us


Field Calibration of Bankfull Discharge, Cross-Sectional Area


Field determination of bankfull stage

The most consistent bankfull determination is associated with the top of the floodplain. This is the elevation where incipient flooding begins for flows above the bankfull discharge. Many floodplains are constructed when rivers abandon their point bars. The elevation of the tops of point bars and the bankfull stage thus share a common elevation that is related to the development of the floodplain within the valley under the current climate regime. Since alluvial channels with well developed floodplains also often have river terraces (abandoned floodplains) adjacent to the channel. For correct bankfull stage determination, it is important for the field observer to know the difference between a terrace and floodplain. For stream types that have a well-developed floodplain — "C," "D," "DA," and "E" types — the bankfull stage is easily and reliably identified as the elevation of the floodplain.

Where floodplains are not well developed, the identification of the bankfull stage must be determined by field indicators that combine corroborating evidence for a consistent common elevation. The appropriate use of any bankfull indicators requires adherence to four basic principles:

  • Seek indicators in the locations appropriate for specific stream types,
  • Know the recent flood and/or drought history of the area to avoid being mislead by spurious indicators (e.g., colonization of riparian species within the bankfull channel during drought, or flood debris accumulations caught in willows that have rebounded after flood flows have receded),
  • Use multiple-indicators wherever possible for reinforcement of a common elevation, and
  • Where possible, calibrate field determined elevations and corresponding bankfull dimensions to known recurrence interval discharges at gaged stations. This verifies the difference between the active floodplain and a low terrace.

There are several visual indicators of the bankfull stage that enable field determination of this important parameter for areas where streamflow records are not available. These indicators vary in their importance and discriminating power for different stream types. A partial listing of these indicators follows:

  • The presence of a floodplain at the elevation of incipient flooding.
  • The elevation associated with the top of the highest active channel depositional features (e.g., point bars, central bars within the bankfull channel). These are especially good indicators for channels in the presence of terraces or channels adjacent to colluvial slopes.
  • A break in slope of the banks and/or a change in the particle size distribution, (since finer material is associated with deposition by overflow, rather than deposition of coarser material within the active channel).
  • Evidence of an inundation feature such as defined benches inside of incised rivers.
  • Staining of rocks.
  • Exposed root hairs below an intact soil layer indicating exposure to frequent erosive flow.
  • Certain vegetation species only for specific locales and/or stream types.

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.

Before selecting a reference reach for obtaining bankfull cross-section data, it is essential to survey a longitudinal profile for at least 20 widths both upstream and downstream to determine representative bankfull indicators for that reach. When corroborating, representative indicators have been identified consistent measured values for bankfull width can be determined. Identifying corroborating indicators helps establish good field techniques and greatly improves the resulting field estimates. Again, it is important to first measure bankfull cross-sections at gaged reaches to calibrate your interpretation of geomorphic features to known streamflow quantities and corresponding return period.

Appropriate locations to determine bankfull widths are summarized for riffle/pool and step/pool channels as shown in Figure 101 and Figure 102, respectively.

Figure 101. Recommended cross-section locations for bankfull stage measurements in "riffle/pool" systems.
Figure 102. Recommended location for measurement of bankfull stage in "step/pool" systems.

In general, for all stream types, the best location to measure bankfull width is the narrowest portion of the cross-section where the channel can freely adjust its boundaries under existing streamflow conditions. The position of the narrowest portion varies by stream type. For example, the best locations for determining bankfull dimensions are at the riffle or "cross-over" reach of "C," "E," and "F" stream types; the middle of the "rapid" for "B" stream types; and the narrow width of the transition from the "step" into the head of the pool for "Aa+," "A," and "G" stream types. Deflectors such as rocks, logs, other debris, nickpoints or unusual constrictions that make the stream unusually narrow or that create unusually wide backwater conditions must be avoided. Additional examples of field methods for locating measurements of bankfull stage for riffle/pool channels are summarized by Lowham (1976), and described for step/pool channels by Osterkamp (1994 personal communication).

Many species of riparian plants are widely distributed, occurring across a variety of hydro-physiographic provinces. Using vegetation to identify bankfull stage must be done cautiously, since some species and age classes can establish themselves within the bankfull channel. Bankfull stage is frequently underestimated when determined solely on the basis of vegetation, and such unilateral determinations should be avoided. Nonetheless, some common riparian species can be used as indicators of bankfull stage such as certain mature species of birch (Betula spp.), dogwood (Cornus spp.), and alder (Alnus spp.), which can colonize and become established at levels very close to bankfull stage. Similarly, mature alders generally are not found within the bankfull channel, unless undercut and "slumped in" on a steep erodible bank. Conversely, smaller woody plants, grasses and forbs can colonize within the bankfull channel; especially during drought as can certain species and age classes of willows (Salix spp.). Such species should not be used as indicators. In cases where there is little choice but to use vegetation as an indicator, it is best to seek the advice of a riparian ecologist familiar with the study area and to verify the relations of these species to stream stage at gaged sites within the same hydro-physiographic province.

Calibrating Bankfull Stage to Known Streamflows

A common error is the failure of field observers to calibrate the appropriate field indicators of bankfull stage to known streamflows. Such calibration is essential until you gain sufficient field experience in a given locale to validate the proper interpretation of the bankfull stage. The recommended procedure for calibrating field identified bankfull stage with known streamflows and return period is as follows:

  • Locate all current and discontinued stream gaging stations within the study basin and/or in nearby similar basins.
  • Make a field visit to each station to collect supplemental data needed to interpret existing hydrologic records at each station. Note that these field visits are not an unnecessary extravagance, nor are they likely to be a major encumbrance on your time. You will be fortunate if you can find a half-dozen gaging stations within your study area, and often it may be necessary to travel outside your study area to obtain representative data for extrapolation.
  • The use of Table 17 will serve as a checklist for procedures to be performed at the gaged site. A portion of this data to be collected is not necessary for stream classification, but is necessary to perform the sediment and hydraulic analyses.
Table 17. Checklist of recommended procedure at USGS gage or other streamflow measurement locations.
Describe Site
  • Geomorphic Setting - Valley Types I through XI.
  • Channel Materials (Pebble Count) (D16, 35, 50, 84, 95)
    • Bed Material - Pebble Count
    • Bank Material - Pebble Count and core sample
    • Bar Material - core sample
  • Locate on Topographic Map
  • Photo Document - Up/Downstream
  • Compute percentage of Watershed Hydraulically Impacted

Longitudinal Profile
  • Measure Average Water Surface Slope
    • Riffle Slope
    • Pool Slope
  • Measure Valley Slope
  • Sequence of Riffle/Pool or Step/Pool as a function of Bankfull Width
  • Locate Bankfull Stage along Longitudinal Profile

Plan View
  • Measure Sinuosity (SL/VL) (VS/CS), where: SL=stream length; VL=valley length; VS=valley slope; and CS=channel slope
  • Meander Geometry
    • Meander Length (LM)
    • Belt Width (BW)
    • Radius of Curvature (RC)
    • Meander Arc Length (ML)
    • Meander Width Ratio

Cross-Section (Dimension)
  • Cross-section of Channel + Valley Features - Terrace/Floodplain (to be identified on cross-section plot).
    • Bankfull Width (Wbkf)
    • Bankfull Mean Depth (dbkf)
    • Bankfull Maximum Depth (dmbkf)
    • Flood Prone Area Width (Wfpa)
    • Entrenchment Ratio (Wfpa/Wbkf)
    • Bankfull Cross-sectional Area (Abkf)
    • Bankfull Velocity (Ubkf)* – estimated from various sources
    • Estimated Bankfull Discharge (Qbkf)
  • Calibrate Bankfull Estimates
    • Survey Estimated Bankfull Stage
    • From gage plate, extrapolate stage reading associated with estimated "Bankfull"
    • Read Discharge from Rating Curve @ Gage (Stage/Discharge Relation)
    • Determine recurrence interval in years from flood frequency curves at station (Should be 1-2 years or average of 1.5 year Q).
    • Analyze hydraulic geometry data from 9-207 forms (discharge notes) for width, depth, velocity and cross sectional area vs. stream discharge. Plot data on log-log paper and run a regression to obtain slope and intercept values for each variable.
    • Develop dimensionless hydraulic geometry relations. This is to be applied for extrapolation purposes to rivers of the same stream type, but for various sizes. W/Wbkf vs Q/Qbkf (Complete for depth, velocity, and cross sectional area).
Worksheet 10 (PDF, 53 kb, 1 p.) provides a form for recording gaged and field data.

3st3tabV2_thThe primary use of this data is to calibrate field-estimated bankfull stage to a corresponding measured discharge and associated return period. The return period of bankfull discharge is most frequently between 1-2 years with 1.5 as an average. If field determined bankfull yields a return period greater than this, the field indicators selected may be indicating a low terrace or other features not typical of the bankfull stage. This form is completed using flood frequency data and rating curve information (stage-discharge relationships) published by the USGS in their annual water resources summaries, and from stream discharge notes (hard copy available for the USGS on Form 9-207; note that for some states this information is now available in electronic format for recent years). To this you must add your own field-data describing the reference reach slope, particle size distribution, bankfull characteristics, and hydraulic geometry, and, of course, stream type.

All gaging stations have a permanent benchmark installed for elevation control and corresponding cross-sections. These cross-section should be re-surveyed and expanded to include the active floodplain, low terraces, and other valley features.

A field guide of bankfull stage determination and conducting a stream channel survey was recently published by the USDA Forest Service (Harrelson et al, 1994). This guide is very helpful in describing stream survey methods, Bankfull stage surveys, pebble counts, and other channel inventory methods. A bankfull video produced by the USFS (1994) is also very helpful in selecting bankfull stage.


Return to Main Flowchart | Proceed to Next Step

Jump to main content.