Critical damage of bridges at river crossings frequently occurs during extreme flood events. This damage is due to various reasons, one of the most common being riverbed scour at bridge piers and abutments. Local scouring is a complex phenomenon induced by the 3D flow field generated around piers and abutments inserted in movable bed rivers. In Portugal, the tragic accident of Entre-os-Rios (59 casualties), was partly due to excess scouring at one of the bridge piers.

At present times, technical and economic considerations impose the choice of bridge foundations composed of a pier column founded on a pile cap, supported by an array of piles. Piers of this configuration are known as complex piers. Examples of the use of complex piers in Portugal may be found in Vasco-da-Gama Bridge, Lezirias Bridge or the bridge over the reservoir of Crestuma-Lever. The cost of large bridges, amounting to billions of Euro, justifies a rigorous prediction of scour depth, both for economic and safety reasons. Research on scouring at complex piers is rather more recent than at single-column piers; only four predictors of equilibrium scour depth at complex piers can be found in the literature; three of them were formulated in the present millennium. In spite of the complexity of the scour phenomena at complex piers, the four predictors were derived from limited experimental evidence. Florida’s method, being the most comprehensive and using the most robust data basis, relies on the results of only 49 tests, while the possible combinations of column, pile cap and pile group geometries are countless.

The complexity of the scouring process at piers and abutments is strongly related with the complexity of the associated 3D near-flow-field. It is well established that, around single-column piers, the flow field includes bow wave, down-flow, horseshow vortex and wake vortices. The strength of these organized structures depends on several parameters, which seem to include the relative sediment size. A challenge arises, which consists on explaining the effect of the relative sediment on the near-flow-field structure and, through this, on the non-dimensional scour depth.

The flow structure at single-column piers has been studied with increasing sophistication – yet, neglecting the influence of the relative sediment size – while no systematic study of the flow structure around complex piers is known until to the present. It seems obvious that the existence and strength of flow structures such as down-flow, horseshoe and wake vortices around complex piers depend on the pier geometry and stage of scour development. The mapping of the flow structure around complex piers remains an open question, deserving investigation as a step towards the understanding of the scouring processes and a better characterization of scour depth at complex piers.

The recent advances on CFD modelling of complex 3D flow fields constitute a promising venue for addressing the scouring processes and the characterization of ultimate scour depth at both single-column and complex piers. The key issue being the connection between fluid and solid phases, there is one immediate step that consists in the validation and improvement of existing CFD codes. This entails the systematic comparison of experimental and mathematical models’ outputs, developing adequate modelling strategies and introducing pertinent experimental evidence in the codes.