Introduction. Scour at bridge piers is a relevant hazard to bridges, as it causes an important proportion of the bridge collapses worldwide. The mechanisms causing scour have been extensively studied in the past, but are still not fully understood and important knowledge gaps are detected. In particular, scour at bridge piers in supercritical flows remain nearly unexplored, even when it represents the case of bridges in mountain rivers and streams. For the same specific energy and flow rate, supercritical flows (Froude number Fr > 1) has much higher velocities and shallower depths than subcritical flows (Fr < 1). Therefore, intuitively, it is expected that local scour around bridge piers will be greater in supercritical flows. However, the evidence, although not systematic, indicates that bridges with piers in supercritical flows do not have a higher collapse rate than bridges in subcritical flows, and measurements of scour at piers in supercritical flows show similar magnitudes as those in subcritical flows. Methods. The analysis of the flow around bridge piers in supercritical flows is presented for the two known flow patterns, namely: detached hydraulic jump and wall jet like bow wave. The analysis of scour is presented for a bridge over the Rucúe river, where first scour measurements during floods with supercritical flow were performed. Flow field at piers in supercritical flow. The flow field around a pier in supercritical flow was performed on the basis of numerical simulations obtained by solving the Reynolds equations with an SST turbulence closure, using OpenFoam software. The free surface was simulated with the Volume of Fluid, VoF model. Flow visualization was done with the software ParaView. Field measurements of scour during floods. Scour caused by supercritical flow was measured after 3 floods during the year 2023 around a pier of the Rucúe bridge in Chile. The flow was permanently measured during the year 2023 by means of a fluviometric station, the topo bathymetry of the section was carried out, as well as a characterization of the bed sediments that correspond to coarse gravel and boulders. Results. The simulations of the flow around a pier show that the turbulent kinetic energy and Reynolds stresses over the bed reach similar magnitudes to the case of subcritical flow, but the size of the horseshoe vortex decreases with Froude number and the ratio of depth versus pier width. This at least, does not contradict the idea that in the supercritical regime scour is not substantially greater than that in the subcritical regime. The field measurements that could be made confirm that the scour formulas are on the safe side, giving slightly higher estimates than what is observed in real cases.

wu