In cooperation with YRDOT
Prepared by: U.S. Geological Survey


Failures of bridges over water are caused primarily by scour and channel instability. Inadequate knowledge and understanding of bridge scour increases risk to bridges and the public who use them. Scour processes include local scour, contraction scour, and channel instability. These processes are functions of iInzerreiated factors which must all be analyzed for a bridge site selected for complete scour evaluation. Contraction scour is the general lowering of the channel section due to flow acceleration through a flow section. Local scour is the erosion due to redirected and contracted flow lines around piers or abutments. Channel instability may be lateral, as channel migration or bank failure, or vertical, as aggradation or degradation. The process may be initiated by man's activities (channelization, mining, damming) or naturally (natural meander cut-offs, landslides), and may be a process begun recently or decades earlier. Likewise, the cause of the instability may be local or many miles removed. Channel instability is the primary process causing bridge failures in some basins and regions, particularly where there has been extensive channelization for flood control or channel mining.

Hazards to bridges caused by contraction and local scour are compounded at sites experiencing significant vertical or lateral erosion from channel processes. It is, therefore, important in the evaluation of potential scour to assess not only local and contraction scour but also the role of general channel processes at a site.

Vertical channel changes leading to bridge failures are not limited to degradation. Basins with steep Mountainside slopes often deposit tremendous loads of gravel and cobbles where they drain into relatively flat, main-valley floors. Bridge openings in streams of this nature may become filled with coarse material even up to the floodplain level. This aggradation can cause contractions at the bridge which can result in the deflection of flood flows towards abutments and the possible undermining of piers and abutments. The reduction in the opening size also may contribute to bridge failure as a result of over topping.

An important consideration in a statewide or regional evaluation of channel stability and scour is the screening of bridge sites to identify those that require further study. Because a large amount of data are required for detailed scour potential, a method is needed to rapidly assess the channel stability and observed scour and to rank bridges according to analyses for observed scour or channel instability.


The objectives of this investigation are to provide information and understanding on stream stability and scour at bridge sites. One set of objectives applies to all of the bridges to be inspected while another set applies only to the bridges selected for further analysis.

The objectives of this investigation for all of the included bridges are:

The objectives of this investigation for bridges selected for further study are:

The U.S. Geological Survey (USGS) has considerable experience in assessing channel stability in a variety of stream environments. Channel stability will be assessed by field inspection of about [1,000) bridge sites. Data collected in this phase will be primarily qualitative and include information on stream characteristics, land use, lateral stability, and vertical stability. The channel-stability data will be collected using a form which will be developed based principally on Simon (1989) and HEC-20 (FHWA, 1991). These data will be used to make an overall assessment of channel stability. That will be summarized as a channel-stability index. The effectiveness of the index will be verified to the extent possible. This index as well as the inspection data will be entered into a GIS data base which can be used to map and identify regional changes in channel stability.

The field inspection also will collect information to identify bridges with observable, existing scour and observable, potential scour problems. This information will include assessment of bridge and stream characteristics as related to scour vulnerability. Data to be collected include pier width and shape and bed material characteristics. It also will include measuring of remnant, surficial scour holes which, in addition to identifying observable, existing scour problems, may be very useful in future evaluations.

The field inspection will not evaluate the structural integrity of the bridge and will not evaluate scour for extreme events where hydraulic and hydrologic analyses are required. Unless specified, the inspections may not have the bridge foundation data needed to assess bridge vulnerability due to observed scour conditions. Inspections will not compare bridge foundation data to observed scour conditions in order to assess the bridge vulnerability to scour.

Potential scour and channel instability will be evaluated in detail at about [30] sites selected in conjunction with the State Department of Transportation based on the field inspections. Procedures to be used are those described in the HEC-18, HEC-20 (FHWA, 1991), and/or those employing a pseudo two-dimensional sediment transport model.

The USGS has participated in the use of pseudo two-dimensional flow and sediment models for the prediction of sediment transport (erosion and deposition) through stream reaches. The model proposed for this study, BRI-STARS, has the ability to simulate loads of bed sediment at various points in a cross section for the specific channel and hydraulic characteristics of a stream reach. Both the flow and sediment load is then routed through the reach during the period of a storm hydrograph. At various locations along the reach, the configuration of the cross section is modified according to the modeled volumes of sediment transported. The model will predict deposition as well as scour. Development of the present form of this model was supported by the National Cooperative Highway Research Program with assistance in modeling bridge hydraulics from the USGS. An algorithm to model channel widening by mass-wasting processes (bank failures) is currently being added to the model as a USGS-sponsored effort and should be available for this study.

BRI-STARS is dependent on empirical equations for sediment transport and local scour, but it does permit considerable latitude in the selection of the equations, use of locally collected data, and calibration of the model to reflect observed scour. The model does not improve'upon the accuracy of existing empirical local scour equations nor reduce the immediate need to collect field bridge-scour data during floods to improve these design equations.

Although BRI-STARS represents an impressive state of the art approach to modeling sediment transport and bridge scour, it has had extremely limited field testing. FHWA currently regards BRI-STARS as a research product and not as an approved design tool. Thus, the process of calibrating and verifying this model in different stream environments will benefit future scour evaluation and sediment transport modeling. Substantial data are required to calibrate a sediment transport model. Calibration data sets should be required at only a few representative streams for each characteristic region, rather than for each stream where the model is applied. (State] has (three] characteristic stream regions for which BRI-STARS will be calibrated and verified. The detailed data collected also will have significant benefit to improving our understanding of scour and sediment transport processes.


Project activities to be described below incorporate both field and analytical techniques required for the successful completion of the proposed work. The activities are arranged in the order in which they will be done.

(1) Develop methods to assess channel stability

Field assessments of channel stability will be conducted by qualified persons using a form or forms to guide and record their assessment. The form used in the field assessment will include data on stream characteristics, land use, lateral stability, and vertical stability. Observable bridge scour data will be collected on remnant, surficial scour holes. Bed material characteristics and data relative to pier width, shape, location and orientation also will be obtained. Geomorphic factors that affect stream stability, as indicated by Figure 1 of the Federal Highways Hydraulic Engineering Circular No. 20 (FHWA, 1991), will be included in the data set.

Principal references for developing the methods to be used in the field inspection are: Simon (1989), HEc-20 (FHWA, 1991), HEC-18 (FHWA, 1991), State Highway Department inspections and the judgement of experienced engineers, hydrologists, and geomorphologists. Considerations in the development of the assessment methods include: effectiveness, time required for site visit, comparability of variables with data collected in past studies and inspections, and regional and temporal transferability.

Verification of inspection variables and indices is complicated by the time scale of channel changes, by the random occurrence of floods impacting this process, and by the cross-correlation and poor definition of the relation between explanatory variables and different scour processes. A comparison of past hydraulic-related maintenance at a representative group of bridges with the channel stability index may provide one form of verification. The objectivity of the form may be verified by comparing independent evaluations of the same group of bridge sites made by several qualified persons. Objectivity is required to support transferability because the evaluations will be made by more than one person.

(2) Assess overall channel stability

In this phase data will be collected for field assessments at [1,000] bridge sites throughout the State. Persons making the bridge site assessments will be qualified by their training and experience to make the field assessments. The forms developed in step (1) will be used at each site.

Separate subgroups of the stability and observed scour potential variables will be selected, weighted as to their importance, and summed into indices. The selection and weighting of variables will be analyzed to optimize the sensitivity and representativeness of the resulting index. Bridges will be ranked according to channel stability and according to observed scour indices. Maps of channel stability (and observed scour] indices will be produced using the GIS. Ranked lists and maps of stability and scour 4---Idices for each bridge site will assist the DOT in screening existing bridges as per TA 5140.20, "Scour at Bridges". Interim reports will provide the ranked listing and maps of stability [and observed scour] indices. The [State] DOT will be notified immediately of bridges observed to have high potential for failure. Regional analysis of stability indices and field data may reveal important regional tendencies in the scour processes.

(3) Select sites for detailed scour analysis

A percentage of those bridges that have been ranked as having high scour or fill potential will be selected in cooperation with State Highway officials for more detailed field work and for flow and sediment modeling. It is estimated that [101 sites will be selected. The mechanism for this selection process will be based on the scour-potential ranking in conjunction with other information obtained from the DOT such as the bridge structure. This information may include, but is not limited to, type of foundation, type of superstructure, provisions for relief flow, traffic volume, detour routes, and age of the structure.

(4) Detailed field work at sites selected for detailed scour analysis

Detailed field work will be done by USGS personnel at sites selected for scour evaluation. Data to be collected include:

a. Channel bed coring logs to define geologic substrata at the bridge site.

b. Bed material size and gradation distribution in the bridge reach, including armoring information.

c. Existing stream and flood plain cross-sections through the reach.

d. Stream geomorphic plan form (sinuosity, braiding, etc.).

e. Watershed characteristics.

f. Scour data on other bridges in the area.

g. Slope of energy grade line upstream and downstream of the bridge.

h. Estimation of the bed material sediment discharge for flow discharges.

i. History of flooding.

j. Channel roughness through the study area and range of depth.

k. Location of bridge site with respect to other bridges in the area, confluence with tributaries close to the site, bed rock controls, man made controls, and downstream confluences with other streams.

l. Character of the stream (perennial, flashy, intermittent, gradual peaks, etc.)

m. Geomorphology of the site.

n. Erosion history of the stream.

o. Development history of the stream and watershed.

p. Sand and gravel mining from stream bed up and downstream from site.

q. Other factors that could affect the bridge.

(5) Analyze magnitude and frequency of floods

Hydrologic analysis will be made to estimate the magnitude and frequency of floods at the bridge site for recurrence intervals up to 500 years. This analysis will use peak-flow records on the selected or adjacent basins, regional equations, or other techniques.

(6) Make scour analysis

Data obtained from the field work will be combined with hydrologic data to make the hydraulic analyses. The hydraulic, bed material, and bridge characteristic data will provide the input to evaluate scour and channel stability using methods outlined in HEC-18 (FHWA, 1991), and HEC-20 (FHWA, 1991). Local scour at piers and abutments, contraction scour, and channel instability will be estimated for the 100- and 500-year flood events. Where available, the known elevation of the pile tips will be compared with the elevation of the scoured channel to further assess the vulnerability of the bridge. For sites with potential fill problems, the estimated amount of filling will be added to the existing elevation of the channel bed and then compared to low-girder elevation to determine the extent of contraction.

(7) Collect model calibration data

Detailed data will be collected at representative sites in each region where sediment modeling is to be applied. If model results can be verified for these representative sites, it would be logical to assume the model predictions at other sites are reasonable. The selection of these sites will be based on representativeness and suitability for data collection and not scour potential. Selection of these sites and the detailed data collection plan will be implemented at the beginning of the investigation. Sets of calibration data (for one hydrograph) and verification data for two or more flood hydrographs will include: detailed, repetitive channel geometry at the bridge; stage, velocity and discharge data. For a few events, an attempt will be made to collect additional calibration data over the hydrograph including: bed material, bed load, suspended sediment, and detailed three-dimensional geometry throughout the study reach.

(8) Calibrate, verify, and use sediment-transport model

The BRI-STARS model will be calibrated and verified using the channel, bridge, and flood-magnitude data obtained in (6) to match the observed scour or fill. Sediment routing will be done by particle-size class using the BRI-STARS model. This model has the advantage of estimating scour or deposition variability across each cross-section without the massive data requirements of true two-dimensional models.

Modeled scour or fill depths will be compared with measured data collected during storm events. In some cases, calibration can be improved by adjusting the bed-material load entering the reach that is to be modeled. The model will compute and combine general scour (degradation), contraction scour, and local scour to estimate total potential scour at the site.

If verification is acceptable, scour will be predicted for the bank-full, or 2-year flow event, as well as the 100- and 500-year flow events.

(9) Report results

Reports will be prepared which summarize the channel stability and scour analysis results for each bridge analyzed. Summary data also will be input to the GIS data base for the purpose of identifying regional tendencies and particular problem areas. The reports will include GIS generated maps divided into regions delineated in consultation with the DOT.


Back to Reconnaissance Type Bridge-Scour Projects Memorandum