Bedforms preserved in the rock record can provide detailed information on the morphologies and hydrodynamics of ancient fluvial systems on Earth and other planets. Existing processproduct relations for bedform preservation assume that fluvial cross strata reflect conditions under which bedforms were equilibrated with the prevailing flow, i.e., steady-state conditions. However, recent theoretical and experimental observations indicate that enhanced bedform preservation can occur in non-steady state, or disequilibrium, conditions, and it is currently unclear how prevalent disequilibrium dynamics are in preserved field-scale fluvial strata. Here we explore whether steady-state assumptions are appropriate for ancient fluvial systems by evaluating the nature of bedform preservation in well studied fluvial deposits of three Late Cretaceous (Turonian and Campanian) geologic formations in central Utah, USA: the Blackhawk Formation, Castlegate Sandstone, and Ferron Sandstone. In the field, we made systematic measurements of cross-set heights to quantify the extent to which preserved cross-sets reflect bedform preservation in steady-state conditions. Across the three formations, unanimously low coefficients of variation in preserved cross-set heights of 0.25-0.5 are inconsistent with bedform preservation in steady-state conditions, and instead point to fluvial systems in which enhanced preservation of bedforms occurred in disequilibrium conditions.Enhanced bedform preservation in fluvial strata can be explained by two independent hypotheses: the effect of flashy flood hydrographs on bedform preservation (flood hypothesis) or bedform preservation in the presence of a morphodynamic hierarchy (hierarchy hypothesis). We estimated bedform turnover timescales to quantitatively assess these competing hypotheses and contextualize their implications. Under the flood hypothesis, field measurements are consistent with enhanced bedform preservation driven by flashy flood hydrographs with flood durations ranging on the order of hours to a few days, which are consistent with perennial fluvial systems subject to heavy rains and tropical storms.Alternatively, under the hierarchy hypothesis, field measurements are consistent with bedform climb angles that range from 10 −2 to 10 −1 , reflecting rapid bar migration. Our work provides a novel way of investigating fluvial discharge variability in the geologic past, and we outline the potential next steps to disentangle the relative controls of flood variability and morphodynamic hierarchy in controlling bedform preservation in ancient fluvial systems.