Periodic Reporting for period 1 - SCALEES (Signature of sediment CAscades following Landslides triggered by Extreme Events in the Stratigraphy)
Periodo di rendicontazione: 2024-02-01 al 2026-01-31
Catastrophic sediment release in fluvial systems is largely driven by landsliding, which occurs naturally during large earthquakes (e.g. November 2016 Mw 7.8 Kaikōura earthquake, New Zealand) or climatic events in mountain belts (e.g. October 2020 Alex storm, France). Landslides are geological hazards that occur worldwide and landslide debris, such as sediment of heterogeneous grain sizes, cascade from hillslope to lake or sea where they are preserved in the stratigraphy. Sedimentary records can reveal crucial information on paleo-climate and tectonics and thus contribute to unravelling what happens upstream in catchments. However, environmental signals of such forcing, i.e. changes in sediment production, transport and deposition, are difficult to identify since sediment transport modifies the initial signals. Spatial and temporal scales are key factors that make such signals detectable in sedimentary records. Non-linear sediment transport leads to signal shredding and even to its loss depending on the system length, and signal frequency and amplitude. Landslides cause hazards such as aggradation and the formation of fans and floodplains. Therefore, it is important to develop a method that copes with buffered, incomplete and shredded signals to recognise landslides in sedimentary records and thus to better understand past extreme events that triggered them and their amplitude and frequency.
Despite more frequent and severe extreme events worldwide, only few studies connect landslide and reservoirs and none directly unravels how landslide signals propagate through a river system and appear in sedimentary records, which is the aim of the SCALEES (Signature of sediment CAscades following Landslides triggered by Extreme Events in the Stratigraphy) project . We identify that the lack of numerical models able to fully describe the entrainment, transport and deposition of multiple grain sizes and the related channel morphodynamics is a key bottleneck in progressing on this topic.
The specific research objectives of the SCALEES project were: 1) to assess how considering multi-grain sized sediments affects the magnitude and duration of fluvial morphodynamic response (change in grain size and topography) to post-earthquake increases in sediment supply; 2) to predict the signature (amplitude and grain size) of landslides induced by catastrophic events in lacustrine sedimentary records and to identify parameters that control the landslide signal propagation through a river system, and, 3) to predict the signature of landslides induced by catastrophic events in fluvial sedimentary records, to assess how vegetation affects this signature and to determine the role played by landslide in the dynamics of alluvial fans, floodplains and terraces.
Despite more frequent and severe extreme events worldwide, only few studies connect landslide and reservoirs and none directly unravels how landslide signals propagate through a river system and appear in sedimentary records, which is the aim of the SCALEES (Signature of sediment CAscades following Landslides triggered by Extreme Events in the Stratigraphy) project . We identify that the lack of numerical models able to fully describe the entrainment, transport and deposition of multiple grain sizes and the related channel morphodynamics is a key bottleneck in progressing on this topic.
The specific research objectives of the SCALEES project were: 1) to assess how considering multi-grain sized sediments affects the magnitude and duration of fluvial morphodynamic response (change in grain size and topography) to post-earthquake increases in sediment supply; 2) to predict the signature (amplitude and grain size) of landslides induced by catastrophic events in lacustrine sedimentary records and to identify parameters that control the landslide signal propagation through a river system, and, 3) to predict the signature of landslides induced by catastrophic events in fluvial sedimentary records, to assess how vegetation affects this signature and to determine the role played by landslide in the dynamics of alluvial fans, floodplains and terraces.
Regarding the first objective, an existing multi-grain size sediment transport and storage model previsouly developed was improved to meet that objective. To better capture the behaviour of fine grain sizes (typically sand), intensive work was carried out to adjust the existing model, both its transport and storage components.
Simulations were run on an artificial river reach inspired by the Hāpuku River, New Zealand, which was impacted by a major landsliding event during the 2016 Mw 7.8 Kaikōura earthquake and is studied in detailed by the Host Institution in New Zealand. Promising results were obtained for a single event (water discharge peak), as the grain-size-specific responses were reproduced by the improved multi-grain-size sediment transport and storage model, which predicted different patterns of hysteresis. This highlights the critical role played by the formation and break-up of the bed armour layer made of coarse grain sizes in controlling the pace of evacuation of the sediment pulse. Fine and coarse sizes do not respond in the same way to a given forcing (water discharge), and interactions between grain sizes (blocking effects such as armouring) result in a delay in the response of the fine grain sizes as it closely linked to the response of the coarse grain sizes.
Further simulations were run using a time series of flood events (water discharge peaks of various magnitudes) and revealed the great potential of the validated model to predict the propagation of landslide signal through the river system and its record in the stratigraphy.
Regarding the second objective, the lacustrine sedimentary record was successfully inverted for the first time using the Quakos (e.g. Croissant et al., 2019) model to predict the location and volume of landslide generated by the 1717 CE Mw 8 Alpine Fault earthquake in New Zealand, and to calculate the sediment fluxes related to the evacuation of the landslide sediment by the fluvial system on the West Coast of the Southern Alps. The landslide density and the fine grain size content of the landslides are two parameters that significantly control the fine signal in lacustrine records. Post-earthquake coarse sediment fluxes were reconstructed for the first time based on the set of parameters that best explain the recorded fine signal.
Regarding the third objective, a new numerical model was developed: the multi-grain size sediment transport and storage model improved in the first scientific objective was coupled with a model of river width evolution (SSTRIM; e.g. Lague et al., 2010) to make it suitable to predict sedimentary records on floodplains/terraces. The theoretical development of this 2.5D model (vertical, streamwise and “simplified” spanwise directions) was completed while its implementation in Python is near completion (done for a single-grain size approach but challenging numerical decisions for the multi-grain size approach remain). This model was designed to include the effects of vegetation present on the floodplains/terraces on the hydraulics and consequently the sediment fluxes. Preliminary simulations were run to test the model performance and helped to refine key sediment flux formulation.
Simulations were run on an artificial river reach inspired by the Hāpuku River, New Zealand, which was impacted by a major landsliding event during the 2016 Mw 7.8 Kaikōura earthquake and is studied in detailed by the Host Institution in New Zealand. Promising results were obtained for a single event (water discharge peak), as the grain-size-specific responses were reproduced by the improved multi-grain-size sediment transport and storage model, which predicted different patterns of hysteresis. This highlights the critical role played by the formation and break-up of the bed armour layer made of coarse grain sizes in controlling the pace of evacuation of the sediment pulse. Fine and coarse sizes do not respond in the same way to a given forcing (water discharge), and interactions between grain sizes (blocking effects such as armouring) result in a delay in the response of the fine grain sizes as it closely linked to the response of the coarse grain sizes.
Further simulations were run using a time series of flood events (water discharge peaks of various magnitudes) and revealed the great potential of the validated model to predict the propagation of landslide signal through the river system and its record in the stratigraphy.
Regarding the second objective, the lacustrine sedimentary record was successfully inverted for the first time using the Quakos (e.g. Croissant et al., 2019) model to predict the location and volume of landslide generated by the 1717 CE Mw 8 Alpine Fault earthquake in New Zealand, and to calculate the sediment fluxes related to the evacuation of the landslide sediment by the fluvial system on the West Coast of the Southern Alps. The landslide density and the fine grain size content of the landslides are two parameters that significantly control the fine signal in lacustrine records. Post-earthquake coarse sediment fluxes were reconstructed for the first time based on the set of parameters that best explain the recorded fine signal.
Regarding the third objective, a new numerical model was developed: the multi-grain size sediment transport and storage model improved in the first scientific objective was coupled with a model of river width evolution (SSTRIM; e.g. Lague et al., 2010) to make it suitable to predict sedimentary records on floodplains/terraces. The theoretical development of this 2.5D model (vertical, streamwise and “simplified” spanwise directions) was completed while its implementation in Python is near completion (done for a single-grain size approach but challenging numerical decisions for the multi-grain size approach remain). This model was designed to include the effects of vegetation present on the floodplains/terraces on the hydraulics and consequently the sediment fluxes. Preliminary simulations were run to test the model performance and helped to refine key sediment flux formulation.
Regarding the first objective, the main outcome is a validated multi-grain size sediment transport and storage numerical model. Preliminary numerical simulations at the reach scale illustrate the model capabilities to reproduce key grain-size sorting patterns and thus the great potential of this new tool to interrogate sediment cascades.
Regarding the second objective, the main outcome is the demonstration that a simplified modelling framework, while intentionally reduced in complexity, proves effective for capturing first-order sediment (fine and coarse) dynamics at the scale of the Alpine Fault and the Southern Alps, New Zealand. This represents a significant step forward in quantifying post-earthquake sediment cascades. Importantly, this work demonstrates the potential of using sedimentary records preserved in natural sinks to invert for landscape-scale sediment transport processes.
Regarding the third objective, the main outcome is the development of a new model that includes the lateral sediment dynamics, i.e. lateral exchanges of sediment between a main channel and the adjacent floodplains/terraces. This work is undergoing (return phase of the project) and will contribute to unravel the role played by floodplain vegetation on the recorded landslide signals.
The results obtained through the SCALEES project are a significant contribution to predict the full signal (all grain sizes) of sediment cascades preserved in stratigraphy in response to an extreme event at various spatial scales. This project also paves the way for inverting the stratigraphic record of landslide induced sediment cascades for quantitative insights into their response amplitudes and relaxation times.
Regarding the second objective, the main outcome is the demonstration that a simplified modelling framework, while intentionally reduced in complexity, proves effective for capturing first-order sediment (fine and coarse) dynamics at the scale of the Alpine Fault and the Southern Alps, New Zealand. This represents a significant step forward in quantifying post-earthquake sediment cascades. Importantly, this work demonstrates the potential of using sedimentary records preserved in natural sinks to invert for landscape-scale sediment transport processes.
Regarding the third objective, the main outcome is the development of a new model that includes the lateral sediment dynamics, i.e. lateral exchanges of sediment between a main channel and the adjacent floodplains/terraces. This work is undergoing (return phase of the project) and will contribute to unravel the role played by floodplain vegetation on the recorded landslide signals.
The results obtained through the SCALEES project are a significant contribution to predict the full signal (all grain sizes) of sediment cascades preserved in stratigraphy in response to an extreme event at various spatial scales. This project also paves the way for inverting the stratigraphic record of landslide induced sediment cascades for quantitative insights into their response amplitudes and relaxation times.