Hydraulic Design of Bridges in Tennessee

The design of bridges over streams requires the application of relatively new analytical techniques to provide a structure capable of safely passing a large storm event as well as economically feasible to build. 

About 86% of all bridges in the United States are built over streams. In Tennessee there are 16,981 bridges that are twenty feet long or longer over water. The typical stream bridge in Tennessee is 92 feet long and has a main span length of 29 feet.

Tennessee is divided into 7 physiographic regions varying from the Blue Ridge Mountains of the upper east to the Mississippi River deltas in the west. Each of these regions has unique characteristics of topography, geology, and hydrology, that present obstacles in the hydraulic design of a bridge.

The design of bridges both by state and local officials in Tennessee for the first half of this century was based on two main methods. The first was Talbot's formula. This formula looked at the topography of the site and computed a minimum required area of opening for the bridge. The depth of flow was based on observed or historic high water marks. This method made no real mention of design flood (storm) frequency, and the hydraulic effectiveness of the bridge opening varied widely. The second method was to roughly span the stream channel. This method was fairly effective for low frequency designs, but during moderate storms these bridges were likely to wash out.

These two methods were the primary tool for bridge designers in Tennessee from the turn of the century to the late 1950's. At that time the Department started utilizing the United States Geological Survey (USGS) in determining major structure lengths, and focused more on Talbot's formula for routine designs. 

By the late 1960s it was obvious that the hydraulic design of bridges warranted further attention, and in 1970 the Tennessee Department of Transportation created a Hydraulic Engineering Section . 

Today the hydraulic design of bridges in Tennessee follows the Federal Highway Administration's (FHWA) guidelines. These guidelines focus on designing a structure that can pass a design storm event without catastrophic failure of the structure.

In the design of a bridge, nearby gage data is analyzed to determine the magnitude and frequency of storms that will pass the site. The magnitudes of peak discharges (usually measured in cubic feet per second or cubic meters per second) are associated with a frequency of occurrence (usually years).

These discharges are related to return frequencies (i.e., 2 year, 10 year, 50 year, 100 year, etc...). These events, commonly termed the 2, 10, 50 and 100 year floods, have a 50, 10, 2 and 1 percent chance, respectively, of being equaled or exceeded during any year. Although the recurrence interval represents the long term average period between floods of a specific magnitude, rare floods could occur at short intervals or within the same year. Additionally, the risk of experiencing a rare flood increases when periods of greater than one year are considered.

Survey information that defines the bridge and floodplain geometrical properties is analyzed. A mathematical hydraulic model is developed from this information and input into a computer for analysis. The Tennessee Department of Transportation utilizes the FHWA/USGS program WSPRO to analyze most bridges. 

The computer analysis will predict water surface elevations associated with the range of discharges analyzed. The discharge at the point just before the water overtops the roadway is designated as the Design Discharge. At this point all water collected in the drainage area upstream of the bridge must pass through the structure. The design discharge is generally the most stressful flood event a bridge and its roadway approaches will have to endure.

Following the hydraulic analysis, an analysis of the scour potential of the bridge is performed. Scour is the erosive action at the bridge due to the movement of water. As the water in the floodplain is constricted through the bridge opening, it increases in velocity. This increase in velocity allows the water to pick up sediment under and near the bridge and carry it away. In the late 1970's, the study of scour on bridge foundations began to produce results that could be applied to design. In September 1988, the FHWA published the first study dealing with scour that made recommendations on the analytical methods to predict scour. This publication on scour at bridges provided the designer with the methods to reasonably predict the scour at a bridge.

The scour analysis calculates the elevation that the design discharge will remove material to. Scour calculations are performed to determine the scouring effects due to the bridge constriction, piers and footings, and abutments. Long-term degradation of the stream is also studied. In areas where rock is near the streambed, scour is generally not a major concern. However, in cobble laden and alluvial streams, scour can have a major influence on the stability of the structure during a design storm.

The minimum criteria for the hydraulic design of a new structure is to provide a hydraulic replacement in kind. However, the replacement in kind must be designed to structurally withstand the calculated scour potential. In many cases the design scour elevation for a replacement in kind would require prohibitively expensive substructures. In these cases the bridge is lengthened and the hydraulic and scour calculations are recomputed until the economic efficiency of the structure are maximized.

In the design of a bridge over a stream, the final length of the structure is based not only on the size of the stream channel that the bridge crosses, but the overtopping elevation of the roadway, the calculated peak discharge, the calculated water surface elevation, the stream bed material and its resistance to scour and a host of other factors. These factors combine to present a unique problem to the Hydraulic Engineer. While some bridges may have similar appearances in design, no two hydraulic designs are exactly alike and every stream crossing must be individually analyzed and all the sites unique characteristics addressed in order to provide a safe and cost effective design.