Scientific Investigations Report 2006-5023
U.S. GEOLOGICAL SURVEY
Scientific Investigations Report 2006-5023
Scour is a complex hydraulic process consisting of lowering a streambed by erosive forces. These forces are caused by acceleration of streamflow around obstructions or through channel contractions. Bridge scour is erosion of the streambed resulting from local flow accelerations caused by bridge piers, abutments, and, during extreme flows, the bridge structure and deck. According to Holnbeck and Parrett (1997), “the most common cause of bridge failure has historically been the scour or erosion of foundation material away from piers and abutments during large floods.”
To address scour concerns, the Federal Highway Administration (FHWA) established a national bridge-scour program in 1991. The purposes of this program are to evaluate existing bridges for scour potential and to conduct research in scour-related areas (Holnbeck and Parrett, 1997). In 1994, the U.S. Geological Survey (USGS), in cooperation with the Alaska Department of Transportation and Public Facilities (ADOT&PF) and the FHWA, began a cooperative study to analyze scour potential at bridges in Alaska. The initial phase of this project consisted of screening a large number of bridge sites for scour susceptibility using a one-dimensional hydraulic model at each bridge. Inputs for the models were derived from bridge plans provided by ADOT&PF, along with other historical data. Methodology and results of this study are available in a report by Heinrichs and others (2001).
The USGS, in cooperation with the FHWA and ADOT&PF, chose the Parks Highway crossing of Tanana River’s main channel, near the town of Nenana, Alaska (fig. 1), for a further, in-depth study because the results of the phase 1 study indicated large scour potential. This site is hydraulically complicated. Streamflow in the Tanana River at the site is divided between the main channel and a slough, and the Nenana River enters the Tanana River just downstream of the bridge. Tanana Slough re-enters the main channel downstream of the mouth of the Nenana River. Bridges cross both the main channel (ADOT&PF bridge 201, referred to in this report as the main bridge) and the slough (ADOT&PF bridge 202, referred to in this report as the slough bridge). The main bridge has a large pier, pier 3, that is oriented at an angle to streamflow. A one-dimensional hydraulic model used in phase 1 estimated a large scour value at the pier. Two types of scour were examined for the phase 2 study—contraction scour, caused by increased velocity in a reach where the flow area of the stream has been reduced, such as by bridge abutments or natural constrictions, and pier scour, erosion around piers caused by local flow acceleration and a resulting increase in turbulence (Richardson and Davis, 1995).
The primary concern at the Tanana bridge site is pier scour at pier 3 on the main bridge. Contraction scour is a secondary concern. Depth of pier scour is affected by several variables, including pier geometry, flow depth and velocity, and angle of attack. Pier geometry can be measured or taken from historical records such as bridge plans; flow depth and velocity can be measured or estimated with a computer model; and angle of attack must be estimated from visual observation or by using a two-dimensional model. In the phase 2 study, both one-dimensional and two-dimensional hydraulic models were used to simulate several discharge scenarios, to determine if a two-dimensional model might provide a better flow characteristic simulation. Hydraulics and scour estimates from each model were compared to each other to determine their applicability in studies at complex sites.
This report (1) presents depths of pier scour, calculated using output from one-and two-dimensional hydraulic models, and contraction scour, calculated using output from the one-dimensional model for five discharge scenarios in the Tanana and Nenana Rivers; (2) documents construction and calibration of the hydraulic models; and (3) compares effectiveness of both types of hydraulic models in scour analyses at complex sites. Using both model types with the same geometry and flows will aid in choosing the correct approach to scour investigations at other complex sites.
The study required establishing and surveying 20 cross sections on the Tanana River, developing water-surface and top-of-bank profiles, and measuring discharge and velocity at each cross section. Historical discharge data were used to calibrate and run the hydraulic models. The one-dimensional model consisted of 43 cross sections, 20 surveyed cross sections, and 23 interpolated cross sections. The two-dimensional model consisted of 8,198 nodes. The models simulated measured discharge for a large flood on the Tanana River and 100- and 500-year recurrence-interval flows on the Tanana River. Each flow was simulated with a high and a low discharge on the Nenana River.
The survey layout and setup of the models focused on accurately simulating high flows. This approach allowed larger spacing between cross sections and a coarser mesh for the two-dimensional model. Flow at the time of the survey was simulated to determine the applicability of the models to relatively low flows.
Tanana River is a glacially fed river with headwaters originating in the Alaska Range. The study reach includes the confluence with the Nenana River and extends from the Alaska Railroad Bridge to about 6,200 ft downstream of the highway bridge and about 1,600 ft up the Nenana River (fig. 1). Total length of the reach on the Tanana River main channel is about 10,000 ft.
As a glacially fed river, the Tanana has a distinctive hydrograph in which high summer flows begin with snowmelt runoff, but are sustained throughout the summer by glacial melt (fig. 2; Brabets and others, 2000). Extreme high flows result from rainfall and typically are in late summer. Flow in autumn and winter gradually decreases until snowmelt begins again in spring. Ice cover usually forms in October and the river remains frozen until spring breakup, usually in April or May. Breakup of the surface ice generally is caused by increased flow rather than melting in place.
As is common with glacially fed rivers, the sediment load of the Tanana River is quite high. A sediment sampling study in the late 1970s estimated annual sediment loads based on discharge records from USGS streamflow-gaging station Tanana River at Fairbanks (15485500) and on sediment samples taken near the gaging station and near the city of North Pole, Alaska (USGS site ID 644322147192900) (table 1, modified from Burrows and others, 1981). Average water-surface slope at the Fairbanks gaging station was 0.00051 ft/ft and at the North Pole gaging station was 0.00118 ft/ft (Burrows and Harrold, 1983). These gaging station sites are about 60 river miles upstream of Nenana. Major tributaries between these sites include the Chena River (non-glacial, relatively low sediment load) and the Wood River (glacial, relatively high sediment load).
Bed-material samples were collected by the USGS at Tanana River at Nenana on June 7, 1964 (fig. 3). The average grain diameter in these samples ranged from less than 0.05 to 10 mm. Average diameter of samples collected from Fairbanks and North Pole ranged from 0.12 to 34 mm, depending on sampling site and location in the channel (Burrows and others, 1981). Bed-material size is closely tied to the water-surface slope. When the water surface is steeper, more energy is available for sediment transport and larger particles can be moved. The largest grain sizes were from the steepest site, North Pole. The water-surface slope at Nenana is about 0.0003 ft/ft. This is similar to the Fairbanks sampling site, in which average grain diameter for the composite of the channel was 0.23 mm.
For more information about USGS activities in Alaska, visit the USGS Alaska Water Science Center home page.
U.S.
Department of the Interior | U.S. Geological
Survey
Persistent URL: https://pubs.water.usgs.gov/sir200695023
Page Contact Information: Publications Team
Page Last Modified: Thursday, 01-Dec-2016 18:59:20 EST
