Thesis
An approach for the development and validation of physically based models for rainfall-induced shallow landslides and their application to understand key initiation mechanisms
- Creator
- Rights statement
- Awarding institution
- University of Strathclyde
- Date of award
- 2018
- Thesis identifier
- T17252
- Person Identifier (Local)
- 201358249
- Qualification Level
- Qualification Name
- Department, School or Faculty
- Abstract
- Rainfall-induced shallow landslides are one of the major natural hazards and have a significant impact on society and economy. The areas affected by this phenomenon are often very extended and unfortunately, sometimes, crossing human activities. Hence, the hazard to people and infrastructures might be significantly high. Landslides initiate with the failure of sloping earth materials which happens when forces or stresses overcome the strength of the soil cover. Shallow landslides usually involve the first few meters of soils and are mainly characterized by a translational movement on a failure surface parallel to the undisturbed ground. One of the main landslides triggering factors is heavy and/or prolonged rainfall events. As rainwater infiltrates into the slope, the pore-water pressure increases resulting in a decrease of the shear strength of the soil eventually triggering the slope instability. The destructiveness of rainfall-induced shallow landslides may increase when, in adverse weather conditions, rainwater may saturate the soil leading to a flow slides or debris flow. Flow slides and/or debris flow are characterized by a fluid mass of a mixture of soil and fragments of rocks, flowing down the slope at very high velocities. Landslides risk is defined as the likelihood of a slope failure to happen and adversely impact the society. It is the result of the combination between the hazard, i.e. the likelihood of landslide occurrence, and vulnerability, i.e. the degree of loss of one or more elements at risk, resulting from the landslides occurrence. Risk, hazard and vulnerability are human concepts. Indeed, they can be only applied in those instances where human beings and goods could be adversely impacted by the landslides. The landslides risk assessment consists of a series of procedures aiming to quantify the possible consequences related to the landslides phenomena, to assess their probability of occurrence and, finally, what can be done to mitigate them. The first step of a landslides risk assessment consists of the evaluation of the hazard. Landslide hazard can be quantified using several approaches. A well-known method is the use of hazard maps, also known as landslides potential or susceptibility maps. They are built by combining the study of past landslides events and the identification of controlling factors over the territory, i.e. slope inclinations, soil profiles, past landslides. They divide a region into zones reflecting the intensity of landslide hazard. The hazard zones are usually classified into at least four categories, for example high, moderate, low and no susceptibility. Landslide hazard maps are a very useful tool for the development planning. Their potential to illustrate regional area with different level of potential landslides hazard may help decisions about the appropriate land-use, building regulation and engineering practice. Landslide hazard maps, together with information on existing or expected vulnerability, can be used to estimate the risk associated with critical facilities like road and rail networks, hospitals, or water pipelines. Such information may be used to make decisions regarding the ‘acceptable risk’ for one or more facilities, the need for relocation or the application of appropriate remedial measures. However, landslide hazard maps are not able to predict when or exactly where landslides will occur during a specific triggering event. They don’t take into account the physics related to the phenomenon. Moreover, they are implicitly based on the assumptions that geomorphological and meteorological conditions causing failure in the past will not change in the future. This assumption may not hold for the case where climate pattern is expected to change. Another class of approaches aiming to quantify the landslides hazard are based on the use of physically-based models. They mainly attempt to extend spatially the slope stability theories. Physically-based models can potentially estimate the rainfall event which is likely to trigger slope failures, the location and the time of occurrence of the event. They usually combine a mechanical model for landslide initiation and a hydraulic model for rainwater infiltration. However, physically-based models require detail spatial information about the hydrological and mechanical properties of the soils and about the morphology of the slopes. Furthermore, physically-based models are calibrated using rainfall events registered at the time and the location of the landslides event. This information might not be always available and it can be expensive to obtain. When Hazard assessment is combined with vulnerability assessment, i.e. the assessment of losses that may be incurred through the impact of landslides hazards on vulnerable elements, it is hence possible to assess the risk. The last step of the risk assessment procedure consists in design procedures and methodologies with the aim of reducing the risk associated with rainfall induced shallow landslides. The risk can be mitigated by either reducing the hazard or the vulnerability. The hazard can be mitigated by acting on those factors which make the slopes susceptible to fail and/or enhancing mechanisms which would lead the slope to remain stable. With this aim, vegetation is often seen as one of the few practicable solutions to mitigate landslide hazard at the catchment scale. It’s well established in literature how the vegetation may positively contribute to the slope stability by reinforcing the slope thanks to the anchoring of the root systems. It is also well known how plants can give a hydrological contribution too, by promoting low pore water pressures via the transpiration process. However, there is an other hydrological effect which the plants are responsible for and it consists in the rhizosphere hydrological contribution. The rhizosphere may have beneficial effects on slope stability by preventing downward rainwater infiltration. Indeed, the rhizosphere acts as a natural lateral drainage. In turn, this implies that shallow landslide hazard can be potentially mitigated by promoting plants with root-system architecture that enhances lateral subsurface flow as a remedial measure. The vulnerability of landslides affected area can be mitigated via early-warning systems. They are usually based on threshold values of rainfall, which is the most ‘accessible’ and frequently used as landslide precursors. Empirical rainfall thresholds are defined through the study of rainfall events registered at the occurrence of landslides phenomena. The thresholds curves are usually obtained by drawing lower-bound lines to the rainfall conditions that resulted in landslides plotted in Cartesian, semilogarithmic, or logarithmic coordinates. They usually build correlation between either the intensity and duration of the rain event registered at the slope failure or between the cumulated rain in the period antecedent the day of the landslides and the daily event at the failure. For a reliable early warning system, a significant amount of data on past failure is needed. Threshold curves are usually defined as a lower bound of the registered critical rainfalls. Unfortunately, this approach often leads to over-conservative thresholds, resulting in false alerts and therefore a malfunctioning of the early warning system. This thesis work proposes a methodology for the study of rainfall-induced shallow landslides, with the aim to gain a deep understanding of the processes related to their triggering mechanisms and provide a reliable tool to improve the landslides risk assessment procedures.
- Advisor / supervisor
- Tarantino, A. (Alessandro)
- Resource Type
- Note
- This thesis was previously held under moratorium from 4th June 2018 until 4th June 2023.
- DOI
Relations
Items
| Thumbnail | Title | Date Uploaded | Visibility | Actions |
|---|---|---|---|---|
|
|
PDF of thesis T17252 | 2025-03-07 | Public | Download |