Scientific Investigations Report 2006-5023

**U.S. GEOLOGICAL SURVEY
Scientific Investigations Report 2006-5023**

Two forms of scour studied at the main pier on the bridge crossing the main channel of the Tanana River were contraction scour and pier scour. The primary concern is pier scour, which was calculated using the one- and two-dimensional models for the five high discharge scenarios. Contraction scour was calculated using output from the one-dimensional model. Because contraction scour was not of primary concern and was of minimal magnitude, the two-dimensional model was not used to further evaluate contraction.

Procedure and scour estimation equations used in this analysis are described in “Evaluating Scour at Bridges; 3d ed.” (Richardson and Davis, 1995). This publication is also known as HEC-18, and describes the scour estimation methods used by FHWA.

Contraction scour was calculated with the live-bed equation from HEC-18 (Richardson and Davis, 1995). Sediment samples taken at Fairbanks by Burrows and others (1981) show that the Tanana River carries a considerable bed load during all stages; the same can be inferred for the Nenana River. This means that the bed is always ‘live’. Contraction-scour calculation results from the HEC-RAS models and variables used in the calculations are shown in table 5.

Pier scour was calculated using the methods in HEC-18 (Richardson and Davis, 1995) and hydraulic data output from both the one- and two-dimensional models. The equation is applicable to both live-bed and clear-water scour conditions. Input variables from both models are in table 5.

Angle of attack can make a significant difference in the calculated pier scour. Partly for this reason, the two-dimensional model was used. The estimated angle of attack from the discharge measurement of August 17, 1967, was used for the one-dimensional model pier scour calculations. The angles of attack simulated by the two-dimensional model (table 5) were used for the calculations with two-dimensional data. The two-dimensional model demonstrated that the angle of attack is affected by discharge in the Nenana River.

The one-dimensional model divided the channel into 20 subsections of equal conveyance. Depth and velocity for the calculation can be taken from directly upstream of the pier or from the deepest, fastest channel subsection. The subsectioning is based on the assumption that flow is evenly distributed across the channel. If this is not the case, then the depth and velocity values could be in error.

The two-dimensional model calculated the depths and velocities at many points across the channel, so the values just upstream of the pier are presumed to be more reliable. The values directly upstream of the pier were used to calculate pier scour for both models. This does not give the most conservative answer, but did provide a basis for comparison of the models.

For more information about USGS activities in Alaska, visit the **USGS
Alaska Water Science Center home page**.