Do Precipitation-Induced Shallow Landslides Occur under Unsaturated Conditions?

点击翻译按钮获取中文摘要

The socioeconomic impacts of natural landslides are enormous worldwide. In the U.S., natural landslides result in the loss of life, as well as property and environmental damages exceeding US$2 billion annually. Shallow landslides, typically translational slope failures a few meters thick of unlithified soil mantle or regolith, may dominate mass-movement processes in hillslope environments. They are particularly destructive when they initiate or coalesce to form debris flows. Shallow landslides and debris flows are commonly triggered by intense precipitation or strong ground shaking and may affect extensive areas during a single meteorological or seismic event. According to a recent survey, about half of the 40 most destructive landslide disasters worldwide in the past century resulted from prolonged or intense rainfall. Recent advances in the scientific understanding of landslide initiation, particularly for those landslides that occur under intense or prolonged precipitation in hillslope environments around the world, indicate that the failure surface may be above the water table and under nearly saturated conditions. The classic methodology for landslide analysis assumes that earthen materials are either fully saturated or completely dry neglecting the contribution of soil suction to the stability of slopes. Thus this methodology is overly conservative and incapable of accurately forecasting shallow landsliding. Recent advances in soil mechanics have shed light on the state of stress in partially saturated soil masses. Furthermore, physical evidence and scientific understanding in both geomechanics and geomorphology all point to the likelihood that the failure surface of infiltration-induced landslides may occur above the water table and under nearly saturated conditions.This project seeks further physical evidence and uses the new knowledge in defining stress in partially saturated soil to identify the likelihood of landslide initiation under partially saturated soil conditions. The overall goals of this research are: to test the hypotheses that (1) landslides, particularly shallow and translational slope failures under heavy precipitation conditions, can occur above the water table; and, (2) that the failure occurs when materials are nearly saturated and suction stress is reduced to nearly zero. To test these hypotheses, we will: (1) measure the unsaturated hydrologic and mechanical properties under laboratory settings, (2) conduct landslide initiation tests using an intermediate-scale shallow landslide simulator, and (3) implement a site-specific program for long-term, real-time monitoring for shallow landslide occurrence, thus examining the validity of both hypotheses at the field scale. By clarifying whether shallow landslides occur under saturated or unsaturated conditions, this work provides a physical basis for answering the important question regarding whether we need a new paradigm that includes either soil suction or effective stress or both to predict shallow landslides. Knowledge gained here is transformative to advancing the understanding of the interactions among climate change, topography, and sustainable land use. Mass movement due to shallow landslide initiation under heavy precipitation is part of long-term landform evolution, and presents a great challenge for society subject to climate change and evolving land use. Testing the concept of effective stress variation above the water table for landslide initiation analysis provides a sound physical basis for future predictive models and will enhance our ability to predict and forecast shallow landslide hazard. Collaboration with the USGS National Landslide Information Center ensures broad dissemination via outreach activities targeted toward researcher, K-12 students, and the general public. Integrating the research mission with an educational and diversity mission via graduate and undergraduate student training, proactive recruitment of underrepresented groups and incorporating the research activities into existing courses ensure broad impact to a diverse audience of students, researchers, and practitioners.