Geomorphic Effects of Floods – Integrating Ancient, Modern and Experimental Data

Extreme river floods are among Earth’s most common and most destructive natural hazards and thus have a high impact on society. Their impact on shaping the landscapes and the sedimentary record is, however, less clear. The most common assumption is that moderate flood events build the sedimentary record, because (1) the high frequency moderate events are suggested to be more common and thus do more geomorphic work, (2) river recovery feedback loops are suggested to be negative and thus high-magnitude event effects are reworked, and (3) river sedimentary records are assumed to consist of ripple, dune and bar scale cross strata that indicate normal Froude subcritical flow, bedload transport, and equilibrium conditions for downstream bedform migration. However, there is a large and growing body of work that documents river deposits that are dominated by preserved high-magnitude flood deposits that consist of stacked flood event beds rather than equilibrium bedform strata, and indicate formative Froude supercritical flow conditions with suspension transport of sand and gravel.

Based on a synthesis of ancient and modern river records, modern river discharge records and experimental data, we propose that there are fundamental differences in magnitude-frequency relationships and relaxation times in rivers with distinct hydrology, such that in some rivers channel recovery may be virtually non-existent and larger floods may leave permanent and formative imprints on landscape. Only if the ratio of the mean relaxation time (normal conditions) to the mean recurrence interval of extreme channel disturbing events is <1, and the critical shear stress for sediment motion is exceeded during moderate (normal) conditions can a river recover from extreme flood-induced change before the next major disturbance occurs. This concept helps to explain the observed variability in the sedimentary record of rivers, as well as critical differences in river flood hazards.