Extending HYRISK to Predict Scour Risk as a Function of Soil Erodibility Characteristics

Click to enlarge




Terry Sturm


Emma Bones
Travis Harris

The majority of bridge failures in the U.S. are the result of bridge foundation scour. However, determining which bridges are most vulnerable to scour is challenging – particularly in Georgia where more than 5,000 bridges have unknown foundations. Financially, it is not feasible to GDOT to thoroughly evaluate scour risk for all of these bridges – to do so would cost at least $100 million.  GDOT must prioritize bridges for scour screenings in a way that identifies those bridges that are most likely to benefit from the installation of monitoring or counter-scour measures.

FHWA developed a risk-assessment tool called HYRISK that calculates the probability of bridge failures due to scour. In a prior GDOT project, Dr. Garrow and her colleagues customized the HYRISK model for Georgia and developed a “scorecard” of bridge vulnerabilities. This scorecard includes an extensive economic loss model due to bridge failures that accounts for time lost by individuals and shippers due to detouring around a bridge closure, rebuilding costs that are linked to construction times, the cost of lives lost due to a bridge collapse, and other factors.

Currently, HYRISK does not incorporate risk adjustment factors for soil factors associated with erodibility. In fact, no State DOT has incorporated risk adjustment factors for soil factors into HYRISK, despite the fact that information about soils is clearly one of the most important factors influencing scour. Intuitively, this is because the original HYRISK model was built exclusively from data inputs available in the National Bridge Inventory; information about soils is maintained in state databases and was excluded from FHWA’s original HYRISK model.

This project proposes to extend HYRISK to include risk adjustment factors that account for soil erodibility based on some results from Dr. Sturm’s prior GDOT-sponsored research. In this research, a relationship was developed for soil erodibility as defined by measured critical shear stress of Shelby tube soil samples taken from fifteen bridge sites in Georgia.

The capability to relate critical shear stress to soil erodibility properties is conceptually appealing, but unfortunately information that is needed to estimate critical shear stress (such as soil size distribution) is available only for a small subset of existing bridges in Georgia which hinders the development of soil risk adjustment factors for HYRISK. However, qualitative descriptions of soil properties based on visual and tactile assessments are available for each bridge in boring logs. In this project, we will develop a method to relate these qualitative soil descriptions to categories of known soil erodibility for incorporation into HYRISK. The methodology will be validated with existing critical shear stress data from previous Ga Tech research and tested on a minimum of 100 bridges.

Given this broad classification of a soil into a small number of erodibility categories, we will develop a set of risk adjustment factors that account for soil erodibility properties for HYRISK. In turn, this will enable us to rank bridges that should be targeted for further scour evaluation; importantly, this ranking will be based on the probability of failure (including soil erodibility properties determined in this project) and expected economic losses.


Publications resulting from this research:

1. Khelifa, A., Garrow, L.A., Higgins, M.J., and Meyer, M.D. (2013). The impacts of climate change on scour-vulnerable bridges: An assessment based on HYRISK. ASCE Journal of Infrastructure Systems 19(2):138-146.

2. Bones, E., Garrow, L.A. and Sturm, T.W. (accepted). Incorporating soil erodibility properties into scour risk assessment tools: An application to HYRISK.  Journal of Infrastructure Systems.