13 Geohazards: Methods for Mapping, Monitoring, and Mitigating

Thursday, 10 April 2014
1:00 PM - 5:00 PM
Solitude

IMPLICATIONS OF SAMPLING METHODOLOGY ON SCOUR DEPTHS AND TIME RATES OF SCOUR AT BRIDGE FOUNDATIONS

Paper number: 13-5
Program: Theme Sessions
Time: 2:20 PM – 2:40 PM

Authors

Abstract

The purpose of this research was to find a relationship between field sampling methodology and calculated scour depth at bridges; a sampling protocol is proposed based on the results. Scour is the term for fluvial erosion of soil surrounding a bridge foundation (piers and abutments). Current guidance regarding scour analysis is provided by the Federal Highway Administration document Hydraulic Engineering Circular 18 (HEC 18), “Evaluating Scour at Bridges.” HEC 18 derives equations from flume studies that are used to calculate scour depth, and scales those equations to foundation materials of different gradations. The analysis takes into account channel geometry, flow parameters, and the grain size of the sediment (D50). While HEC 18 describes where to take samples, it does not address sampling methodology; hydrologists rely on a wide variety of manual and mechanical sampling methods which are further complicated because sediments to be sampled are often under water.

The study selected as field area an armored stream channel at a bridge southwest of Harrisonburg, VA, for which a steady-state flow model had been developed. The bridge sits on Quaternary alluvium consisting of sand, pebbles, gravel, cobbles, and boulders, with the largest particle size of 180 mm in diameter. The bedload was sampled using various manual techniques, including shoveling, auguring, aerial cobble counts, walking cobble counts, and volumetric sampling behind a barrier device, with samples taken from the stream at the locations prescribed by HEC 18. Each method was compared based on its error, bias, and precision. Scour was then calculated using the varying gradations provided by each sampling method. Counter-intuitively, scour only varied a small amount over a wide range of D50s. Based on the error, bias, and precision of each sampling method a preferred sampling method and sample size was determined.