Experimental investigation of local scour around two cylinder-shaped piers in tandem arrangements

Abstract

Local scour is one of the most common causes of foundation failures in bridge piers, offshore and onshore structures, fish habitat structures, etc. In recent years, multilane or wide-span bridges are becoming popular in designing numerous engineering projects for geotechnical and economic reasons. Pier groups in tandem arrangements (in line with the flow) rather than a single pier are generally used to support those superstructures. Therefore, an experimental investigation was conducted to understand the influence of pier size and spacing on scour depth in tandem arrangements. Two identical cylinder-shaped piers were placed in two different sediment beds e.g., Sand-A (d50 =0.23mm) and Sand-B (d50 =0.89mm). Two different diameter piers (D = 8.3cm and 5.5cm) were used, and two identical-sized piers were placed in each run. A total of 20 experiments were conducted in single and tandem arrangements with four different pier spacings, S (2D, 3D, 6D, and 9D). During each experimental run, three-dimensional instantaneous velocity was measured using Acoustic Doppler Velocimeter (ADV) to investigate the turbulence structures along the plane of symmetry (POS) of the tandem arrangements. It was observed from the streamwise velocity component that reverse flow occurred in between the tandem piers for S/D ≤ 3. The turbulence parameters (turbulence intensity, turbulent kinetic energy, Reynolds shear stress) showed that for 2D and 3D pier spacings, strong turbulence structures were developed at the wake zones of the front pier. As the pier spacings further increased, the turbulence characteristics' effect decreased in the wake of the front pier. The temporal evolution of scour depths revealed that the rate of scour development at the front pier was higher than at the rear pier. For the front pier, approximately 90% of the maximum scour depth was observed during the first hours, whereas the rear pier took around 90 minutes. For Sand-A, the maximum scour depth was found in the front pier of 8.3cm pier, which is 12.66%, 4.40%, 13.89%, and 19.12% greater than 5.5cm diameter pier for 2D, 3D, 6D, and 9D spacings, respectively. The maximum scour depth at 8.3cm pier in Sand-B was 8.8cm for 2D spacing, which is 1.14% smaller than the 2D spacing of Sand-A. Again, the maximum scour depth for 5.5cm pier in Sand-B was found to be 6.70cm, which is 17.90% smaller than the maximum scour depth of Sand-A. Since the Sand-B is non-uniform, the larger particles were initially deposited into to scour hole (formation of armor layer); thus, the scour depth decreased. The maximum scour depth at the front pier decreases with the increase in spacing for 8.3cm pier, whereas, for 5.5cm pier, the scour depth slightly increases up to 3D, then decreases as the spacing increases. For 8.3cm pier, the maximum scour depth was found in 2D spacing, which is 9.9% and 9.3% greater than the minimum scour depth at 9D spacing for Sand-A and Sand-B, respectively. For 5.5cm pier, the maximum scour depth was found for 3D spacing, which is 16.2% and 2.3% greater than minimum scour depths at 9D spacing for Sand-A and Sand-B, respectively. The spatial variations of scour depths revealed that for 2D and 3D pier spacings, the front and rear piers scour hole was overlapped and formed an elliptical-shaped single scour hole. For 6D and 9D pier spacing, two separate scour holes were formed around the front and rear piers. The maximum near-bed shear stress variation in the plane of symmetry showed similar trends with the scour depths development. The larger scour depths occurred in the locations in which greater bed shear stress was found and vice versa. Three existing predictive equations were used to compare with the observed maximum scour depths. All the predictive equations overestimated the observed scour depths. In addition, a new correction factors have been introduced to the CSU single-pier equation for predicting scour depths in pier groups of tandem arrangement. The new equation gives a correlation coefficient (CC) is 0.92, and DR = 1.02. Findings of this study can be used to enhance the existing understanding of scour developments and the influence of turbulence parameters on sediment mobility around tandem pier arrangements.