Experimental investigation of the effect of hydrograph and steady flow properties on group pier bridge scouring for different spacing

Introduction
The pier group is one of the important hydraulic structures that the scouring around them is affected not only by the flow characteristics, but also by their number and arrangement. Each pile in a group has an individual scouring mechanism that can influence the other piles in the group. The following are mechanisms that make pier group scouring more complicated than a single pile: 1) sheltering, 2) reinforcement, and 3) horse-shoe vortex compression (Nazariha, 1996).
More recent attention has focused on the arrangement and geometric variables and their effects on the scouring size and process in steady flow. Several reviews of the angle of attack, spacing, numbers, and pier diameter have been undertaken (Hannah, 1978; Nouh, 1986; Vittal et al., 1994).
As seen, most studies in the field of scouring around pier groups have only focused on steady current, and there is a relatively small body of literature that is concerned with scouring in pier groups in unsteady flow, while floods, waves and unsteady flows are the most destructive phenomena in rivers and coastal environments. This paper uses the experimental investigation of three piers in the tandem arrangement as a pier group and analyses the impact of hydrograph unsteadiness on scouring with different pier spacing in the clear-water regime and investigate the time variation of the scouring depth to understand the scouring process around pier groups in unsteady flow.

Methodology
The experiments were conducted in a flume 10 m in length, 0.74 m in width and 0.6 m in depth at the Hydraulic Laboratory of Shahid Chamran University. In the flume a pump was used to drive the water from an underground reservoir to a head tank. A false bed was built at the bottom of the flume with 0.15 m height, with a 1.7 m length sand bed located 2.8 m from the inlet. The sediment part of the bed was filled with d50=0.7 mm uniform sand (Geometric standard deviation of the sand size(σ)=1.3) and the other parts were covered with the materials as rough as the sand.
In this study, 36 experiments were done to evaluate the pier group scouring, which included 9 tests of steady flow in different discharge and pier spacing, and 27 tests of unsteady flow in different peak discharge, time duration and pier spacing. It is worth mentioning that all experiments were performed in the clear water regime.
During the experiments, four cameras (Full High Definition (FHD) resolution) recorded the scouring process from four different angles to investigate the temporal changes (Fig. 4). All piles were scaled to extract the scouring depth from the videos and remove the light deflection effect in the water.

Results and Discussion

Steady flow results
What stands out from the steady flow experiments investigation is that the scour depth around the first pier was more than the next one due to the flow attack. In other piers, because of the previous pier's protection, the scour depth was less than the first one. As shown in Figure 4, the scour depth changing rate decreased earlier at lower discharges, which may be due to the lower flow intensity to continue the scouring process.
Early in the second and third pier scouring process, the scouring depth remains constant for a while, which is due to the eroded sediments from the previous pier into the next pier's scouring hole and the equal amount of deposited and eroded sediment. This issue is also seen in figures of Mahjoub et al.'s (2014) research.

Unsteady flow results
The scouring depth around each pier gradually increases with increasing flow discharge during the hydrograph's rising limb. This increase occurred at the beginning of the process slower than the steady flow due to a gradual increase in flow discharge and, consequently, a gradual increase in flow intensity and shear stress to erode sediments around the pier (fig 5-8).
In the unsteady flow experiments, the first pier's scouring process was gradually stopped after the peak discharge and during the falling limb due to the decrease in flow discharge. However, evaluating the recorded videos from the scouring process angles and the extracted data showed that the process proceeded differently for the second and third pier in some experiments. In these cases, four conditions around the pier were occurring by reducing the flow discharge in the falling limb. These conditions caused new phenomena called backfilling in this study.
• Sufficient height of deposition region resulting from previous pier scouring
• Sufficient flow intensity for erosion and moving the sediment
• Short distance for sediment to reach the next pier scouring hole
• weak vortex of the next pier to re-erode the entered sediments into the hole

Conclusion
Evaluating the scouring hole's temporal variation during the unsteady flow shows that the scouring process around the pier group in unsteady flow can differed from steady flow around rear piers and caused a new phenomenon, which is called backfilling. This difference is due to the extra mechanisms in the pier group scouring process, and the flow changes during the hydrograph and changed with changing the pier spacing.