Stochastic modelling of the sediment cascade in the Illgraben
Event details
| Date | 29.10.2013 |
| Hour | 16:15 › 17:15 |
| Speaker | Prof. Peter Molnar, Chair of Hydrology and Water Resources Management, ETH Zurich |
| Location | |
| Category | Conferences - Seminars |
Abstract:
Fluvial systems generally exhibit sediment dynamics that are strongly stochastic. This stochasticity comes from the randomness involved in sediment production, sediment transport pathways, and the transport triggering process, e.g. heavy rain or overland flow. Fully deterministic models of fluvial systems, even if they are physically realistic and very complex, are likely going to be unable to capture this stochasticity and as a result will fail to reproduce important features of long-term sediment dynamics. In this presentation I will show instead a simplified conceptual approach to modelling based on the notion of the sediment cascade and will apply it to the Illgraben. The model conceptualizes Illgraben as a spatially lumped cascade of connected reservoirs representing interconnected hillslope and channel storages where sediment is stochastically generated and goes through multiple cycles of storage and remobilization by surface runoff. All relevant hydrological processes that lead to runoff in an Alpine basin are included. Despite its simplicity, the model produces complex sediment discharge behaviour which is driven by the availability of sediment and antecedent moisture (system memory) as well as the triggering potential (climate). The argument will be made that this modelling approach has greater potential for statistical predictions of debris flow occurrence than methods based on a single rainfall threshold. I will conclude by arguing that in the context of stochasticity, traditional notions of stability and equilibrium, of the attribution of cause and effect, and of the timescales of process and form in geomorphic systems, become increasingly difficult.
Dr Peter Molnar is professor of hydrology and fluvial systems at ETH Zurich. His main research aim is to describe and quantify the propagation of stochasticity from rainfall into water and sediment fluxes on a river basin scale and the resulting effects on river morphology and processes in the riparian environment. He places high importance on high quality education in hydrology and water resources problems, both as a service to developing countries as well as in the graduate programme at ETH.
Fluvial systems generally exhibit sediment dynamics that are strongly stochastic. This stochasticity comes from the randomness involved in sediment production, sediment transport pathways, and the transport triggering process, e.g. heavy rain or overland flow. Fully deterministic models of fluvial systems, even if they are physically realistic and very complex, are likely going to be unable to capture this stochasticity and as a result will fail to reproduce important features of long-term sediment dynamics. In this presentation I will show instead a simplified conceptual approach to modelling based on the notion of the sediment cascade and will apply it to the Illgraben. The model conceptualizes Illgraben as a spatially lumped cascade of connected reservoirs representing interconnected hillslope and channel storages where sediment is stochastically generated and goes through multiple cycles of storage and remobilization by surface runoff. All relevant hydrological processes that lead to runoff in an Alpine basin are included. Despite its simplicity, the model produces complex sediment discharge behaviour which is driven by the availability of sediment and antecedent moisture (system memory) as well as the triggering potential (climate). The argument will be made that this modelling approach has greater potential for statistical predictions of debris flow occurrence than methods based on a single rainfall threshold. I will conclude by arguing that in the context of stochasticity, traditional notions of stability and equilibrium, of the attribution of cause and effect, and of the timescales of process and form in geomorphic systems, become increasingly difficult.
Dr Peter Molnar is professor of hydrology and fluvial systems at ETH Zurich. His main research aim is to describe and quantify the propagation of stochasticity from rainfall into water and sediment fluxes on a river basin scale and the resulting effects on river morphology and processes in the riparian environment. He places high importance on high quality education in hydrology and water resources problems, both as a service to developing countries as well as in the graduate programme at ETH.
Practical information
- General public
- Free
- This event is internal
Organizer
- EESS - IIE
Contact
- Prof. Paolo Perona, AHEAD