RESPONSE TIMES IN LANDSCAPE EVOLUTION: PROMISES AND PITFALLS

The terrestrial landscape is composed of different entities such as hillslopes, bedrock and alluvial rivers, alluvial fans and floodplains for example. It is increasingly recognised that because these entities are dominated by different processes they might respond in different ways and at different rates to external forcings depending on the nature, magnitude and time scale of changes. Therefore, knowledge of those response times are fundamental if we want to “read” landscape forms and withdraw information on past climate or tectonic changes. In addition, because all these entities are connected inside the sediment routing system, the coupling of processes and their response times also control the response of the landscape as a whole, and the delivery of sediment flux to the basins. Investigating landscape evolution from this perspective gives a comprehensive framework in which many disciplines contribute to the general understanding of surficial mass redistribution. In one of such studies we attempted to quantify the plausible time scales of clastic sediment supply variations at the entrance of sedimentary basins. Our approach was based on the sedimentary system concept, which simplifies natural systems by dividing them into three zones of dominant processes: the erosion, the transfer, and the sedimentation subsystems. Considering that frequent climate changes can induce high-frequency sediment flux variations at the outlet of the source area put forward the crucial role of the transfer subsystem, which conveys sediment from the erosion zone to the basin. Applying a diffusive model to a number of worldwide rivers, we extended from large (>1000 km) to intermediate (>300 km) rivers the previous finding that the transfer subsystem could act as a buffer for short periods sediment pulses (tens to hundreds of kiloyears). This implied that high-frequency stratigraphic cycles in clastic accumulations fed by large drainage systems are unlikely to reflect sediment supply cycles of tens to hundreds of thousands of years of periodicities. Here I will review the promises and pitfalls of this approach in order to outline possible future research directions.