Combined abutment and contraction scour in compound channels for extreme flood events

Abstract

In this PhD project, the combination of abutment and contraction scour is investigated to better understand the scour mechanisms and scour patterns for extreme floods. Close-toreality scour events were physically simulated using models built at 1:45 and 1:30 geometric scales of two-lane bridge prototypes. Scour and flow-measurement experiments under submerged orifice and overtopping flows were carried out. To better understand the effect of vertical contraction on abutment scour, free surface flows were also investigated for similar experimental conditions. The majority of the experiments were carried out in compound channels, simulating abutments set back from the main channel. Spill-through abutments were used in the live-bed scour regime, and both spill-through and wing-wall abutments were used in the clear-water regime. In addition, to better understand the effect of contraction length on abutment scour, and also to verify the long contraction theory for apron-protected, abrupt abutments, a series of long contraction experiments were carried out with vertical-wall abutments. Several major conclusions can be drawn from the results of this project. For the investigated conditions, results show that vertical contraction significantly affects the flow pattern, the temporal development of scour and the final scour bathymetry. Comparing with submerged orifice flows, flow relief of the overtopping flows has a small effect on the near-bottom turbulence and the scour. Flow patterns at the initial state are found to correlate with scour patterns at the equilibrium state. In the bridge section, a “retreating” behaviour of the main channel bank is observed; at the equilibrium state, the side slope of the “retreated” main channel bank is observed to be invariant, presenting a simple geometric relationship between the depth of the scour hole and its location. For unprotected abutments, scour is centred at the upstream corner of the abutment, regardless of contraction length; and for apron-protected abutments, scour differs significantly with contraction length. Numerical and physical modelling work is required in the future to broaden the knowledge of abutment and contraction scour. Also, further research is required to improve the scour countermeasure design for abutments under pressure flows.