Scour at piers embedded in a bed of cohesive sediment mixture is investigated to quantify the effects of bed compaction. Experiments were conducted in a large rectangular laboratory flume with natural clay sediment collected from a stream located at the Chilean coastal range. Nonintrusive and high-resolution topographic measurements of scoured bed in the vicinity of piers were performed during the experimental runs with an installation of a laser distance sensor (LDS) that was traversed by the precision step motors. Measurements were taken by the LDS in different azimuthal half-planes to study the spatiotemporal variation of the topography of evolving scour holes. Experiments were conducted over 40h, until an advanced stage of scour was reached, with the approaching flow having a bed shear stress equaling 95% of the Shields critical condition for the initiation of motion of sand in a sand-clay mixture. Results show that the bed is successively scoured in three different forms, chunks of aggregates, aggregate by aggregate, and particle by particle, when increasing the ratio of actual to Proctor's optimum molding water content, . Maximum scour depth was mainly observed at the wake region (downstream of the pier) and ranged from 10 to 58% of that in the sand alone case for mixtures with the same size of the sand fraction. For a given compaction energy, scour depths first decrease with , up to a value of , as fine clay particles are flocculated and increasingly dispersed with water content. For values of , the sediment matrix behaves as a fluid mud. In the latter case, scour depths increase with water content as the sediment bed becomes increasingly liquid.