Junction flows: more than "water under the bridge"...

What is it about?

Scouring around bridge piers is a complicated process. This study overcame some major obstacles to a better understanding such as: limited amount of data (in both space and time), simplifications in the texture of the soil around the pier, simplistic techniques for data analysis and numerical models with poor predictive capabilities. Using an advanced technique for flow measurement, the laboratory tests delivered a wealth of information at numerous points of the flow domain. These data were collected at a high rate, so that the dynamics of the phenomenon was fully captured. The experiments were designed in a way that a realistic bed geometry could be used. In this way, it became possible to evaluate how the flow changes in the presence of a rough bed (as opposed to the smooth bed case, which had been the benchmark for bridge scour studies).

Featured Image

Photo by Fabian Schneiter on Unsplash

Photo by Fabian Schneiter on Unsplash

Why is it important?

Scouring is the leading cause of bridge failure in the USA. Due to the complexity of the phenomenon, there is incomplete knowledge about its characteristics. As a result, engineers have limited confidence in existing methodologies for scour prediction and frequently end up over-designing the scour protection measures at the expense of skyrocketing costs. The much needed insight to develop more reliable techniques against bridge scour can only be provided through more and better data. This Work furnished a wealth of high-quality datasets. New aspects of the phenomenon were investigated that will allow, for example, to more accurately apply results from small, laboratory scale models to real-life structures. Or to account for the roughness of sediments and refine existing models, which for simplicity had so far considered smooth surfaces. This Work also took our understanding of the scouring phenomenon to the next level through the application of advanced methodologies for data analyses, which can extract the most important information (scour-inducing flow structures in both space and time) from the massive data sets.