This report provides the results of a detailed Level II analysis of scour potential at structure
CHESTH00030007 on Town Highway 3 which is also State Route 35 crossing the South
Branch Williams River, Chester, Vermont (figures 1–8). A Level II study is a basic
engineering analysis of the site, including a quantitative analysis of stream stability and
scour (U.S. Department of Transportation, 1993). Results of a Level I scour investigation
also are included in Appendix E of this report. A Level I investigation provides a qualitative
geomorphic characterization of the study site. Information on the bridge, gleaned from
Vermont Agency of Transportation (VTAOT) files, was compiled prior to conducting Level
I and Level II analyses and is found in Appendix D.
The site is in the New England Upland section of the New England physiographic province
in southern Vermont. The 10.4-mi2 drainage area is in a predominantly rural and forested
basin. In the vicinity of the study site, the surface cover is pasture on the upstream right
bank while the immediate bank has some trees. Downstream of the bridge and the upstream
left bank are forested.
In the study area, the South Branch Williams River has an incised, sinuous channel with a
slope of approximately 0.03 ft/ft, an average channel top width of 65 ft and an average bank
height of 4 ft. The channel bed material ranges from gravel to boulder with a median grain
size (D50) of 70.5 mm (0.231 ft). The geomorphic assessment at the time of the Level I and
Level II site visit on August 26, 1996, indicated that the reach was laterally unstable. There
are cutbanks on both the left and right banks alternating with point bars in the upstream
The Town Highway 3 (VT 35) crossing of the South Branch Williams River is a 74-ft-long,
two-lane bridge consisting of one 72-foot steel-beam span (Vermont Agency of
Transportation, written communication, March 30, 1995). The bridge is supported by spill-
through abutments. The channel is skewed approximately 5 degrees to the opening and the
opening-skew-to-roadway is also 5 degrees.
Three channel scour holes 1.0 ft deeper than the mean thalweg depth were observed during
the Level I assessment in the upstream reach. There are no scour protection measures at the
site. Additional details describing conditions at the site are included in the Level II
Summary and Appendices D and E.
Scour depths and recommended rock rip-rap sizes were computed using the general
guidelines described in Hydraulic Engineering Circular 18 (Richardson and others, 1995).
Total scour at a highway crossing is comprised of three components: 1) long-term
streambed degradation; 2) contraction scour (due to accelerated flow caused by a reduction
in flow area at a bridge) and; 3) local scour (caused by accelerated flow around piers and
abutments). Total scour is the sum of the three components. Equations are available to
compute depths for contraction and local scour and a summary of the results of these
Contraction scour for all modelled flows ranged from 0.0 to 0.4 ft. The worst-case
contraction scour occurred at the 100-year. Abutment scour ranged from 4.1 to 15.5 ft. The
worst-case abutment scour occurred at the 500-year. Additional information on scour
depths and depths to armoring are included in the section titled “Scour Results”. Scoured-
streambed elevations, based on the calculated scour depths, are presented in tables 1 and 2.
A cross-section of the scour computed at the bridge is presented in figure 8. Scour depths
were calculated assuming an infinite depth of erosive material and a homogeneous particle-
It is generally accepted that the Froehlich equation (abutment scour) gives “excessively
conservative estimates of scour depths” (Richardson and others, 1995, p. 47). Usually,
computed scour depths are evaluated in combination with other information including (but
not limited to) historical performance during flood events, the geomorphic stability
assessment, existing scour protection measures, and the results of the hydraulic analyses.
Therefore, scour depths adopted by VTAOT may differ from the computed values