Modelling the impact of seismic landslides on landscape evolution

Xie, Jun

Abstract

Mass movements such as landslides are prominent sources of sediment in active mountain belts. As evidenced by recent earthquakes in Chi-Chi, Wenchuan and Gorkha, seismic triggered landslides can have a major impact on river basins and landscape evolution. “Landslide-landscape relationships” are dominated by factors including the hydrological (climate) and geomorphic characteristics of the landscape, the frequency and location of landslides, and landslide properties (e.g., grain size and runout length). Although a wide variety of studies have attempted to quantify or determine the exact contribution of landslides to basin sediment budgets, no established technique can explicitly fingerprint/track landslide derived sediment. Thus, the key factors that control the impact of landslides and their role on landscape evolution still remain poorly understood.
In this thesis, the landscape evolution model, CAESAR-Lisflood (CL), was modified to enable sediment from landslides to be tracked throughout the basin, and to include spatial changes in slope failure angle to trigger landslides. The modified CL was then used to investigate the impact of landslides on the Hongxi basin, which suffered tremendous damage from the 2008 Wenchuan earthquake (Ms 8.0). Three scenarios were modelled. First, the evolution of the Hongxi basin was simulated with and without landslide activity under two future climate scenarios to identify the contribution of landslides to the basin sediment budget in response to climate change. Second, three different landslide grain size distribution (GSD) scenarios (Original, Coarser, Finer) were simulated to study the impact of landslide grain size variation on basin sediment transport and landscape evolution. Third, landslides were generated at twenty random locations within four distinct regions within the basin- upstream and downstream sub-basins were further divided into low and high elevation regions using the proximity to the stream network- to investigate the impact of landslide location on basin sediment yield, dynamics and connectivity.
Results show that landslides can greatly increase sediment yield (up to 35% in the study basin) and that climatic perturbations can lead to variations in sediment dynamics. The interplay and relative importance of landslides and climate is dependent on the landslide density within a specific basin. The annual landslide-derived sediment yield stabilizes after 20 years under both the RCP 4.5 and RCP 8.5 climate change scenarios in the study basin; this adjustment timescale is controlled by grain coarsening as a result of fine-grained sediment evacuation. Modelling results also indicate that the location of landslides could alter fluvial sediment yields and the spatial pattern of landform change, with low elevation landslides being more likely to link or connect to the drainage system (higher connectivity), thus leading to greater sediment yields at the basin outlet. Conversely, material generated by high elevation landslides tend to accumulate on hillslopes, which could lead to a sediment evacuation lag from the basin.
These findings are of great importance for understanding the characteristics of landslides and the interplay between landslides and climate change, thus providing support for landslide risk prevention and mitigation. More importantly, it is evident that the modelling approach, considered as a conceptual and effective tool, is able to identify the first-order controls on the post-seismic evolution of landslide-generated sediment transport in good agreement with observations and conclusions from the literature. This approach can be considered as a prototype study to be extended and improved using a more robust and reliable modelling framework in studies targeting larger areas and to provide more comprehensive insights in post-earthquake risk assessments.

Citation

Xie, J. (2023). Modelling the impact of seismic landslides on landscape evolution. (Thesis). University of Hull. https://hull-repository.worktribe.com/output/4923124

• Thesis Type: Thesis
• Deposit Date: Nov 25, 2024
• Publicly Available Date: Aug 2, 2025
• Keywords: Physical geography
• Public URL: https://hull-repository.worktribe.com/output/4923124
• Additional Information: Physical Geography; Department of Geography, Geology & Environment; University of Hull
• Award Date: Jan 1, 2023