A Comparison of Bridge Modeling and Scour Prediction Using 1-D and 2-D Hydraulic Models

Since scour is the leading cause of bridge failures in the US, accurately estimating bridge scour depths for extreme flow conditions is necessary for engineers and managers to ensure public safety and protect valuable infrastructure. Climate change is further impacting our infrastructure sustainability by introducing uncertainty and potentially increased flow intensities. The current state of the practice for predicting bridge scour is detailed in the Federal Highway Administration’s (FHWA’s) Hydraulic Engineering Circular No. 18 (HEC-18). For decades, one-dimensional (1-D) hydraulic models have been the preferred method of calculating flow conditions despite their limited ability to replicate complex flows present in river systems. With the evolution of technology, two-dimensional (2-D) hydraulic models have become a more viable method for computing flow conditions. The purpose of this study is to compare the modeling methods and results between different numerical models and modeling approaches used to predict bridge scour. For this study, one bridge site located on Dry Fork Creek in central Missouri was examined using three numerical hydraulic methods: WSPRO (1-D) (data from a previous USGS study from the 1990s with only cross sections immediately adjacent to the bridge), HEC-RAS (1-D), and SRH-2D. LiDAR data were combined with ADCP bathymetric data to develop a comprehensive channel surface that was used for both the HEC-RAS and SRH-2D models. This paper contrasts the modeling challenges, discrepancies, and scour predictions between the different methods. The use of 2-D numerical models proved to be the optimal approach given the vast availability of ground terrain data from public domain sources, limitations with establishing 1-D cross sections to accurately represent highly 2-D flow conditions, and the reasonable computational demand of 2-D modeling methods. Given the advancements in modeling tools and data access that have facilitated simulating 2-D flow conditions, this study also highlights the need for advancements in methods for estimating or modeling bridge scour in 2-D flow environments.