Quantifying fluvial non linearity and finding self organized criticality? Insights from simulations of river basin evolution

A numerical study was undertaken to investigate non linearity and the potential for self-organized criticality (SOC) in the evolution of river basins. Twenty-three simulations were carried out, using the authors' CAESAR landscape evolution model, in which the magnitude of storm events, variability of storm events, sediment heterogeneity, sources of sediment supply, and catchment morphology are systematically varied to evaluate their importance as possible drivers for non linear behavior and SOC. Temporal fluctuations in simulated sediment yield show notable non linear behavior. Storm magnitude and occurrence of landslides appear to have little impact on variability of the sediment yield, when compared to the impacts of sediment heterogeneity, rainfall variability and catchment morphology. Particularly, it appears that the non linearity of sediment yields results from the manner in which the catchment processes the variable rainfall, rather than just the rainfall variability itself. The variations in sediment yield show a power law magnitude–frequency distribution, which is a possible, but inconclusive, indicator of SOC. However, several other, more qualitative arguments can be made to support the case for SOC in these simulations. Specifically, we identify the nature of the critical state and suggest two cascade mechanisms by which the system can organize itself around this critical state. Combined, these arguments indicate that simulated evolution of river basins indeed exhibits SOC, at least with respect to sediment yield. The critical state appears to be an indicator of the connectivity of the drainage network. Thus, the simulations indicate that, unlike traditional SOC systems, the critical state of the system can vary in time, as sudden changes in drainage network connectivity may result in sudden changes in the SOC behavior of the system.