Effect of hydrologic and hydraulic calculation approaches on pier scour estimates

Abstract

Bridge scour, a phenomenon characterized by the erosion and removal of sediment from the vicinity of bridge foundations, is a significant concern in the fields of civil engineering and infrastructure management. There are various hydrologic and hydraulic approaches to calculating the peak flow that are used to determine the water depth and velocity in the vicinity of a bridge, which are variables used for scour estimates. Depending on the assumptions, limitations, and boundary conditions, each approach can yield significantly different flow results that influence water depth and velocity estimates. Also, even when methods estimate similar flow magnitudes, different velocity distributions can result from bridge configurations between these methods. The extent to which these methods can influence pier scour depth estimation is not well understood due to a lack of systematic investigations. This research addresses this question by evaluating pier scour on four bridges with 12 combinations of hydrological and hydraulic approaches for a total of 48 simulations. Each simulation was analyzed to assess the potential pier scour depth using the FHWA HEC-18 and the Observation Method for Scour methodologies. There are various alternatives to calculating the peak flow: Regional Regression Equations (RRE), Flood Frequency Analysis (FFA), and distributed models using HEC-HMS 4.9 were evaluated using the SCS Curve Number for abstractions and different antecedent moisture conditions. The peak flow was estimated for a 100-yr event, and the hydrological models were simulated for one event-based. HEC-RAS 6.1/6.2 was utilized for the hydraulics analysis, and 1D-WSPRO, 1D-Energy, 2D SA connection, and 2D terrain modification with raised piers were used as bridge modeling approaches. The models in HEC-RAS were created using Lidar data with a resolution of 1 meter x 1 meter (3.28 x 3.28 ft). The results showed that the regression equations, often used by state DOT's, do not always yield the worst-case hydrological scenario when compared with hydrological models’ simulation. The results of the 1D models are very similar, and in most cases, they produce less scour depth. The 2D approaches better represent the approach channel for the bridges with complex configurations and depict large velocities and therefore more scour depth than the 1D models. Lastly, it was found that the moisture conditions can influence determining the worst-case scenario for peak flow determinations, which in turn impact the scour calculations.