Time Of Slope Failure Research Articles

Machine Learning: A Novel Approach to Predicting Slope Instabilities

Geomechanical analysis plays a major role in providing a safe working environment in an active mine. Geomechanical analysis includes but is not limited to providing active monitoring of pit walls and predicting slope failures. During the analysis of a slope failure, it is essential to provide a safe prediction, that is, a predicted time of failure prior to the actual failure. Modern-day monitoring technology is a powerful tool used to obtain the time and deformation data used to predict the time of slope failure. This research aims to demonstrate the use of machine learning (ML) to predict the time of slope failures. Twenty-two datasets of past failures collected from radar monitoring systems were utilized in this study. A two-layer feed-forward prediction network was used to make multistep predictions into the future. The results show an 86% improvement in the predicted values compared to the inverse velocity (IV) method. Eighty-two percent of the failure predictions made using ML method fell in the safe zone. While 18% of the predictions were in the unsafe zone, all the unsafe predictions were within five minutes of the actual failure time, all practical purposes making the entire set of predictions safe and reliable.

Hydro-mechanical analysis of rainfall-induced fines migration process within unsaturated soils

Seepage-induced fines migration under rainfall infiltration is a main cause leading to shallow failures in loose colluvial slopes. To describe the full process of fines migration within unsaturated soils during rainfall infiltration and the associated hydro-mechanical behaviors, a seepage-erosion-deformation coupled formulation is proposed in this paper. The governing equations proposed are implemented into a finite element code and used to investigate the influences of skeleton deformation on the rainfall infiltration process through unsaturated soil columns. The numerical results were presented in detail for a better understanding of the rainfall-induced fines migration process within unsaturated soils. Further, the obtained results are integrated into an infinite slope model for slope stability analysis. The results show that, the skeleton deformation will affect the rainfall infiltration rate and hence the timing of slope failures; meanwhile their influences are more evident if the fines deposition process is taken into account. Moreover, the slope stability could be reduced gradually due to the soil strength loss along with loss of fine particles. Therefore, particular attentions should be paid to analyzing the stability of soil slopes susceptible to internal erosion.

Real-Time Prediction Method for Slope Failure Caused by Rainfall Using Slope Monitoring Records

The failure time of a slope is predicted by a method based on creep failure theory for slope displacement on natural slopes, embankments, and cutting slopes. These prediction methods employ several equations based on the relationship between the displacement rate (displacement velocity) and time. However, such methods harbor problems because the shape of the tertiary creep curve is affected by many conditions, and it is difficult to identify the phase of tertiary creep. This study examines the time change in the displacement rate of the slope and derives an index for identifying the phase of tertiary creep. Two models of large-scale composite granite slopes were tested by using a large-scale rainfall simulator. In the experiments, the slope displacements were monitored in real time. From these results, inflection points were found in the velocity of the slope displacement. It was found that the corresponding inflection points at different locations in the sliding soil mass occurred with the same timing. This paper discusses the effectiveness of the prediction method for slope failure time by using the inflection points of displacement rate in real-time monitoring records.

FULL-SCALE MODEL TEST FOR PREDICTING COLLAPSE USING INVERSE OF VELOCITY OF SLOPE SURFACE DURING EXCAVATION

Predicting the onset of slope failure or rock falls is important for occupational safety because they kill approximately ten to 20 workers every year in Japan. Approximately half of such victims are from slope failure during slope excavation and construction. In order to predict the time of slope failure during an excavation, a full-scale model slope was built, and the displacement of the slope surface was monitored during slope excavation. The surface displacement rapidly increased with the elapsed time after the excavation, and the relationship between the displacement and elapsed time included an exponential function just before collapse. Based on the results, by computing the inverse of the velocity of the slope surface displacement obtained in ten-second intervals, a warning signal can be provided five min before collapse.

A new prediction model for rock slope failure time: a case study in West Open-Pit mine, Fushun, China

Establishing an accurate method for predicting the failure times of rock slopes subject to creep deformation is challenging, but at the same time crucial for preventing damage to properties and loss of life. In this paper, the Medium–short Term Prediction of Landslide by Polynomial (MsTPLP) model is proposed based on the Levenberg–Marquardt (LM) algorithm. The West Open-Pit mine in Fushun, NE China is currently the largest open-pit coal mine in Asia. The landslide on the southern slope of the West Open-Pit mine was selected as the study case. Global Positioning System (GPS) monitoring is employed in landslide displacement monitoring. Based on the analysis process of the MsTPLP model, the displacement time series derived from GPS monitoring points is selected as the input. The model parameters of the MsTPLP model are obtained using the Levenberg–Marquardt (LM) algorithm. The predicted failure time of a landslide, which is the output, can be determined according to the prediction criteria of the model. The prediction results show that the MsTPLP model can provide accurate landslide displacement predictions (correlation coefficient R 2 > 0.98 and average relative error ARE < 17 %). The forecasting results of the landslide show that the estimated failure time is Mar 5, 2014. Based on field investigation and displacement analysis, the landslide on the southern slope of the West Open-Pit mine occurred on Mar 9, 2014. The predicted and actual failure times are significantly close, demonstrating the potential of the new method in landslide prediction.

Characteristics of landslides in unwelded pyroclastic flow deposits, southern Kyushu, Japan

Abstract. We investigated 184 landslides that occurred in unwelded pyroclastic flow deposits (Shirasu) on southern Kyushu Island, Japan, that included detailed data on the rainfall characteristics and the timing of slope failure. Localized rainfall intensity, antecedent precipitation index (API), and topography affected the hydrologic processes that triggered landslides. API (adjusted for evapotranspiration losses) for large ( &gt; 200 mm) storms that triggered landslides was much lower than for smaller ( ≤ 200 mm) storms. Mean storm intensity and 7-day API (API7) thresholds of &gt; 5 mmh−1 and ≤ 30 mm (or API30 ≤ 60 mm), respectively, were useful to identify landslides triggered by rapid pore water pressure response, especially for shorter ( &lt; 20 h) duration events. During smaller storms with lower intensity, landslides are likely affected by a combined increase in soil weight and loss of suction when API30 ≥ 150 mm; simulations indicated that these weight and suction changes due to rainfall accumulation decreased the factor of safety in steep Shirasu slopes, but did not necessarily trigger the landslides. Most of the landslides that were plotted below a general rainfall intensity–duration threshold for landslide initiation occurred during smaller storms with API30 values &gt; 200 mm, indicating that they were highly influenced by the combined effects of the accumulated weight of rainfall and loss of suction. Our findings show that both event rainfall characteristics and API affect the hydrogeomorphic processes that trigger different types of landslides in Shirasu. This knowledge and the thresholds we have identified are useful for predicting the occurrence of different types of landslides in unwelded Shirasu deposits and improving sediment disaster prevention practices, including real-time warning systems.

Surface roughness dating of long-runout landslides near Oso, Washington (USA), reveals persistent postglacial hillslope instability

Establishing regional landslide chronologies is necessary to advance from hazard recognition to risk assessment, and to understand the evolution of landslide-prone terrain. Despite recent advances in landslide mapping due to the availability of high-resolution lidar imagery, estimating the timing of slope failures remains a challenge. Here we present a new integrated approach to dating landslides on a regional scale by augmenting quantitative surface roughness analysis with radiocarbon dating and numerical landscape modeling. We calibrate a roughness-age curve, which we use to date 25 deep-seated landslides in glacial sediment surrounding the catastrophic A.D. 2014 Oso landslide along the North Fork Stillaguamish River in Washington State (USA). Lidar bare-earth images show a high density of long-runout landslides in this region. Using our roughness-age curve, we estimate an average Holocene landslide frequency of 1 every 140–500 yr, and show that the 2014 Oso landslide was the latest event in an active history of slope failures throughout the Holocene. With each landslide, substantial sediment is delivered to the North Fork Stillaguamish River, driving shifts in the active channel that ultimately affect the pattern of landslides across the valley. The high frequency of landslides in this area, where river incision and isostatic uplift rates have dropped dramatically since peaking soon after ice retreated from the region, shows that landscapes inundated by glacial sediment do not require dramatic changes in base level to remain highly unstable for tens of thousands of years.

Rainfall-induced infiltration, runoff and failure in steep unsaturated shallow soil deposits

In unsaturated shallow deposits, rainfall infiltration and runoff may cause either slope failure or erosion processes depending on the combination of rainfall intensity and duration. Consequently, different flow-like mass movements may occur, whose distinction is fully necessary for the management and mitigation of the posed risk. To provide a contribution to this topic, the paper proposes an engineering reference framework to evaluate the amount of both rainfall infiltrating the ground surface and runoff flowing as wash out and remarks are outlined as far as the time to runoff and the slope failure time. This framework is validated through a numerical parametric analysis based on seepage and slope stability analysis. The obtained results show that time to runoff, time to failure and runoff rates are strongly affected by soil water characteristic curves, soil initial conditions, rainfall intensity and slope angle. Furthermore, slope stability analyses show that time to failure can be either shorter or longer than time to runoff depending on soil mechanical parameters. Finally, it is outlined that the proposed framework provides more accurate estimates of time to runoff and runoff rates compared to simplified standard procedures.

Estimating the Age of Near-Shore Carbonate Slides Using Coral Reefs and Erosional Markers: A Case Study from Curacao, Netherlands Antilles

Carbonate mass failures are widespread, particularly along steep tropical shorelines, however, the age, frequency, and cause of these failures are often poorly constrained. As shorelines undergo intensive human development, understanding when carbonate slopes failure occur is becoming increasingly important. Using both land and marine-based observations, we present methods for making valuable first-order age estimates of near-shore slope failures. As a case study, we analyze the partially submerged lower carbonate terrace along the southern shoreline of Curacao, Netherlands Antilles. This terrace sits adjacent to a well-documented carbonate slope failure. Our combined land-sea study constrains the slide contact, the amount of post-slide coral reef growth, and the location of erosional shoreline notches. From these observations, we place limits on the timing of slope failure, and conclude that the Caracas Bay slide did not occur recently (within the last ~3000 years) but more likely at least ~116,000 years ago, near the end of the last interglacial sea level high stand. Slope failure likely occurred along the unconformable contact separating the base of the coastal limestone with underlying Cretaceous basalt, and may have been triggered by a large earthquake along the Caribbean-South American plate boundary.

Using simulated hydrologic response to revisit the 1973 Lerida Court landslide

Hydrologically driven mass wasting in the form of landslides on steep slopes is a worldwide occurrence. High-profile events in, for example, Brazil, Chile, the Philippines, Puerto Rico, and Venezuela during the last three decades all clearly illustrate, based upon significant losses of life and property, that hydrologically driven slope instability in developed (urban) areas can be a major geologic/environmental hazard. The focus of this study is the 1973 hydrologically driven Lerida Court landslide in Portola Valley, CA, USA. Physics-based hydrologic-response simulation, with the comprehensive Integrated Hydrology Model, was employed to forensically estimate the spatiotemporal pore pressure distributions for the Lerida Court site. Slope stability, driven by the simulated pore pressure dynamics, was estimated for the Lerida Court site with the infinite slope/Factor of Safety approach. The pore pressure dynamics for the Lerida Court site were reasonably captured by the hydrologic-response simulation. The estimated time of slope failure for the Lerida Court site compares well with field observations. A recommendation is made that hydrologically driven slope stability estimates including variably saturated subsurface flow be standard protocol for development sites in steep urban settings.

A Practical Numerical Model for Seepage Behavior of Unsaturated Soil

ABSTRACT It is important to estimate the seepage properties of unsaturated soils when performing an unsaturated-saturated seepage analysis, which is used to clarify the slope failure mechanism and predict the occurrence time of slope failure due to rainfall. Kitamura et al. (1998) proposed a numerical model to quantitatively estimating the seepage properties of unsaturated soils. And they have compared numerical results with laboratory soil tests in order to examine the validity of the numerical model. Consequently, it has been found that the numerical model must be improved moreover. In this paper a practical numerical model for seepage behavior of unsaturated soil is proposed to improve the previous model proposed by Kitamura et al. (1998). Firstly, the basic theories of the previous model are explained. Calculation results are compared with those obtained from the laboratory soil tests in order to examine the validity of the previous model. Then, we perform some improvements of the previous model based on hydro-mechanical properties, and propose a new parameter, named “parallel translation index (Ipt)”. From the results we showed that the relationship between the new parameter and the uniformity coefficient (Uc), expressed by logarithm is linear relation. Therefore, the reasonable seepage properties of soils can be computed from the improved numerical model using only the popular soil parameters.

Slide structure and role of gas hydrate at the northern boundary of the Storegga Slide, offshore Norway

The Storegga Slide off the coast of Norway is one of the largest underwater slide complexes known and has been proposed as a significant source of past methane release into the atmosphere. We present pre-stack depth-migrated images from a new multi-channel seismic data set in the Storegga Slide. The northern scarp of the Storegga Slide has previously been interpreted as a single, large recent slope failure; however, our images show strong evidence for a composite structure consisting of a much older event and recent slumping. We observe onlapping features onto slide deposit highs, and layer thickening as post-slide sediments fill in accommodation space created at the slide scarp, both of which support this conclusion. Displaced fault blocks are overlain by undeformed, flat-lying sediments, also indicating considerable time between slide events. According to dating of the base of the Naust at this location, this older slide event occurred at a minimum of ∼ 250 ka. The causes of submarine slope failure are poorly understood, but previous studies have proposed both earthquakes and dissociation of gas hydrates as triggering mechanisms. Pressure/temperature modeling shows that, assuming steady-state conditions, the bottom simulating reflector (BSR) would have been deeper than the glide plane at the time of slope failure. The base of the gas hydrate stability zone, and any gas that may have been present, likely played only a minor role, if any, in slide initiation at this locale.

INFLUENCE OF SLOPE FAILURE PROCESS ON FLOOD AND SEDIMENT RUNOFF IN A MOUNTAIN RIVER

A heavy rainfall often causes slope failures and landslides. The earth mass produced by a large scale slope failure buries a valley, and sometimes forms a landslide dam. Water and sediment runoff process in the river channel are strongly affected by the formation and deformation of landslide dams. In a report of the natural disaster 2004, the earth mass sometimes breaks into several blocks, and each block starts to move separately. Such slope failure process seems to influence on the landslide dam formation. I apply 2-D hybrid debris flow model to the landslide dam formation process, and investigate the influence through the calculations in different occurrence timing of slope failure. The model was proposed to explain the runoff process of the soil block including the fluidization phenomenon, which is composed of the usual two-dimensional debris flow model and a tracking model for soil block movement as solid. The deformation process of landslide dam and the water and sediment runoff process are calculated by a usual two-dimensional bed variation model.

3-D Seismic Characterization of Submarine Landslides on a Miocene Carbonate Platform (Luconia Province, Malaysia)

3-D seismic reflection data and a variance cube are used to determine the architecture and investigate the triggering processes of submarine landslides affecting the flanks of a Miocene carbonate platform in the Luconia Province, Malaysia. The slide masses exhibit, in time-slice displays, chaotic, patchy seismic patterns, in otherwise smooth reflections. They lie basinward of the slide scar, tend to widen in the transport direction, and end in indistinct lobes. Slide scars appear as crescent-shaped embayments in otherwise straight or oval platform-edge contours. However, slide scars show planar morphology where they coincide with a fault zone. In vertical sections, the basal surfaces of the slides are steep concave slope segments (slide scar) that rapidly flatten where they dip under the slide masses. Slide masses appear as zones of discontinuous wavy reflections that wedge out in both upslope and downslope directions, extend for 1.5 km into the basin, and have a maximum thickness of 130 m. The slide deposit on the western flank is the result of at least two individual sliding events. Two seismically chaotic bodies are separated by a smooth reflection interpreted as an intercalation of hemipelagic mud between carbonaterich slide masses. Syndepositional faulting affects the geometry of the platform and the platform margins, particularly at the time of slope failure. We suggest that the slides were generated by the interplay of steep-slope progradation and faulting accompanied by seismic shocks.

Synergetic-bifurcated prediction model of slope occurrence and its application

Landslide prediction is one of the most important aspects of prevention and control for geological hazards and the environmental protection. In order to study the nonlinear methods for landslide prediction, the synergetic-bifurcated model of predicting the timing of slope failure is established by combining Synergetics with Bifurcation Theory based on single-state variable friction law in this paper. The synergetic effects and bifurcated process of the factors in the slope evolution can be characterized in the model. Taking the Xintan Landslide as an example, the prediction of landslide is carried out based on the model suggested.

Hydro-Geological Study of Landslides Caused by Heavy Rainfall on August 1998 in Fukushima, Japan.

Heavy rainfall over a wide area of northern Japan from August 26 to 31, 1998, triggered many landslides which caused serious damage to human lives as well as houses and civil engineering structures, particularly in Fukushima Prefecture. We report preliminary finding of a hydro-geological study of two landslide sites, Nishigo and Taishin, Fukushima. When the stability of slopes during a rainstorm is discussed, both the thickness and hydraulic conductivity of the surface soil must be taken into account, in addition to the rainfall intensity and cumulative rainfall. These hydro-geological factors are important, in particular for the prediction of the time of slope failure. In certain cases, the hydro-geological structure of the surface soils can be used as a diagnostic factor for slope stability. For example, an impermeable zone can obstruct the outflow of seepage water from a slope, so that a kind of underground dam is accidentally constructed in situ. In such a case, a site investigation, in which much attention is paid to the hydro-geological structure, is crucial for the discussion of slope stability.

The significance of rainstorm variations to shallow translational hillslope failure

AbstractLandsliding in eastern Scotland results from high‐magnitude rainstorms generated under either cyclonic or anticyclonic conditions, particularly during the summer. Data from Aviemore indicated that cyclonic storms produce higher rainfall totals than anticyclonic storms, as well as being of longer duration and lower intensity. The distribution of rain during individual storms also varies with the synoptic conditions under which the storms are produced. These different rainfall characteristics produce different geomorphic responses, which can be investigated in detail using physical based modelling. In this paper, a physically based coupled hydrology–stability model is used to assess the significance of these rainfall characteristics to soil moisture response and slope instability for mature podsols. The results provide evidence that rainstorms of different synoptic origin produce varying hydrological response, involving both the extent and the timing of moisture content change. This affects the depth and timing of slope failure, with anticyclonic storms promoting a large, rapid response in the factor of safety at shallow depths within the soil. Cyclonic storms produce a more gradual response, with the region of probable failure being deeper. Futhermore, each of these storm types is associated with different rainfall distributions, and this is also shown to have a significant effect on the timing and depth of slope instability.

クリープ変形式を基にした斜面崩壊時刻の予測について

クリープ変形式を基にした斜面崩壊時刻の予測について

Forecast Methods of Slope Failure Time and Their Case Studies

Two forecast methods of slope failure time and results of their applications for seven landslide cases are presented. One is the three points method on the cumulative displacem nt curve, the other is the method of least squares for discrete points on that curve . The equations used to forecast the slope failure time are as follows:(1) Equations for the method of three pointsSaito's equations: (1)Fukuzono's equation: (5) and (6)Kawamura's equation: (8)Omura and Tsuchiya's equation as: (11) and (12)(2) Equations for the method of least squaresSaito's equation: (13), (16) and (17)Fukuzono's equation: (19), (16) and (17)Kawamura's equation: (21), (16) and (17)Omura and Tsuchiya's equation: (23), (16) and (17)Both methods are useful to forecast the failure time of slope . In a strict sense, the latter method in the tertiary stage can forecast more accurately than the former.

Volcanic hazards from Bezymianny- and Bandai-type eruptions

Major slope failures are a significant degradational process at volcanoes. Slope failures and associated explosive eruptions have resulted in more than 20 000 fatalities in the past 400 years; the historic record provides evidence for at least six of these events in the past century. Several historic debris avalanches exceed 1 km3 in volume. Holocene avalanches an order of magnitude larger have traveled 50–100 km from the source volcano and affected areas of 500–1500 km2. Historic eruptions associated with major slope failures include those with a magmatic component (Bezymianny type) and those solely phreatic (Bandai type). The associated gravitational failures remove major segments of the volcanoes, creating massive horseshoe-shaped depressions commonly of caldera size. The paroxysmal phase of a Bezymianny-type eruption may include powerful lateral explosions and pumiceous pyroclastic flows; it is often followed by construction of lava dome or pyroclastic cone in the new crater. Bandai-type eruptions begin and end with the paroxysmal phase, during which slope failure removes a portion of the edifice. Massive volcanic landslides can also occur without related explosive eruptions, as at the Unzen volcano in 1792. The main potential hazards from these events derive from lateral blasts, the debris avalanche itself, and avalanche-induced tsunamis. Lateral blasts produced by sudden decompression of hydrothermal and/or magmatic systems can devastate areas in excess of 500km2 at velocities exceeding 100 m s−1. The ratio of area covered to distance traveled for the Mount St. Helens and Bezymianny lateral blasts exceeds that of many pyroclastic flows or surges of comparable volume. The potential for large-scale lateral blasts is likely related to the location of magma at the time of slope failure and appears highest when magma has intruded into the upper edifice, as at Mount St. Helens and Bezymianny. Debris avalanches can move faster than 100 ms−1 and travel tens of kilometers. When not confined by valley walls, avalanches can affect wide areas beyond the volcano's flanks. Tsunamis from debris avalanches at coastal volcanoes have caused more fatalities than have the landslides themselves or associated eruptions. The probable travel distance (L) of avalanches can be estimated by considering the potential vertical drop (H). Data from a catalog of around 200 debris avalanches indicates that the H/L rations for avalanches with volumes of 0.1–1 km3 average 0.13 and range 0.09–0.18; for avalanches exceeding 1 km3, H/L ratios average 0.09 and range 0.5–0.13. Large-scale deformation of the volcanic edefice and intense local seismicity precede many slope failures and can indicate the likely failure direction and orientation of potential lateral blasts. The nature and duration of precursory activity vary widely, and the timing of slope faliure greatly affects the type of associated eruption. Bandai-type eruptions are particularly difficult to anticipate because they typically climax suddenly without precursory eruptions and may be preceded by only short periods of seismicity.