The Washington State Department of Transportation (WSDOT) presently uses an empirical approach in estimating the depth to which riverbed scour is apt to occur around bridge piers. The empiricism arises at least in part, from the use of prediction equation of the form ds/b=K(y0/b)n where ds is the predicted scour depth, b is the width of piers, y0 is the depth of the approach flow, K is a multiplier that incorporates geometry of piers and their orientation to the flow path in streams and n is a factor reflecting erosive characteristics of streambeds.
More than 35 different formulae, having form similar to the one above, have been proposed for scour estimation since 1949. All apply most appropriately to cohesionless streambed materials that are uniform in size. Many site conditions in Washington and other states have graded material with some armoring characteristics. Prediction equations of the above type will estimate scour much deeper than what actually occurs in these latter type streambeds.
Bridge pier construction can be overly costly if they penetrate the streambed unnecessarily deep. At the same time, however, designs must be safe. Research reported here examined whether existing WSDOT scour estimating practices are appropriate or whether other methodology should be used. Results point out that where uniform- sized cohesionless streambeds exist, correct estimating methods are satisfactory. However, at bridges over streams having graded bed materials current estimates of scour depths are excessive. A procedure for estimating these latter depths has been developed and is presented herein for use by WSDOT and others.