Landslide Episodes Research Articles

Back-analysis of a rainfall-induced landslide case history using deterministic and random limit equilibrium methods

Because recurrent rainfall-induced landslide episodes mimic prior failures, this study builds a back-analysis approach to identify those prior failure events and assess the prospective slope performance going forward. Engineers use the results of this approach to project the slope performance based on the quantity of rainfall accumulation during the anticipated rainfall event. A landslide case that occurred in 2004, as a result of a typhoon event, was selected for the back-analysis procedure. Four typhon events were chosen to simulate the pore water pressure conditions, which may have caused damage of varying severity (including landslide failure) to the slope. The numerical model was initially analyzed via the deterministic limit equilibrium slope stability approach to evaluate the most critical failure surface. By adopting the Monte Carlo simulation approach, the following step assessed the random limit equilibrium slope stability on the most critical failure surfaces. The probabilistic distributions of the safety factor were then generated for each typhoon event. Subsequently, the simulation outcomes provide safety indicator values incorporating the mean safety factor, failure probability, and reliability index. From the failure probability and reliability index, a zonation approach with various site-specific performance levels were established by using the corresponding accumulated rainfall to define the exceedance threshold values for each zone. In order to confirm the site-specific performance, we compared the normalized error values acquired from the analysis technique to those measured under two recent typhoon occurrences. An acceptable consistency was attained, validating the procedure's appropriateness for future rainfall events.

THE NEXUS BETWEEN LAND USE CHANGES AND LANDSLIDES OCCURRENCES ON THE SLOPES OF MT. ELGON, BUNGOMA COUNTY IN KENYA

Land use and land cover has been recognized throughout the world as one of the most important factors influencing landslides occurrence. Although many landslide episodes have been reported and observed on the slopes of Mount Elgon both in Uganda and Kenya, little literature exists describing the relationship between land use/land cover and landslide occurrences. The overall objective of the study was to establish the link between land use changes and landslide occurrence. The study adopted mixed method research design. The research used both primary and secondary data. Primary data was collected through interview schedules and questionnaires. Purposive sampling was used to select sites from the four wards for purposes of primary data collection and transect walk. Simple random sampling was used to select 381 households. Landsat satellite images for land cover change analysis was sourced from Regional Centre for Mapping of Resources for Development (RCMRD). Primary data which included quantitative data as analyzed using the statistical package for social scientist (SPSS) (version20). A linear regression model was run to get the predictor values between the variables. The results of this study revealed that; there is a positive link between land use changes characterized by agricultural practices, human activities, increased population, semi-urbanization and landslide occurrences in the Mt. Elgon region. On basis of the above findings, the study recommends that there is an urgent need by appropriate stakeholders from Government, both national and County, to restore forest cover on the deforested steep slopes and restrain the communities from encroaching on the pristine mid slopes of Mt Elgon.

Volcano‐Tectonic Setting of the Pisciarelli Fumarole Field, Campi Flegrei Caldera, Southern Italy: Insights Into Fluid Circulation Patterns and Hazard Scenarios

AbstractPhreatic and hydrothermal eruptions are small energetic explosive events that generally have few to no precursors and represent a considerable hazard in tourist and urban areas. At the Campi Flegrei caldera, these events have occurred at the Solfatara volcano and have likely occurred at the nearby Pisciarelli site, where the most powerful hydrothermal phenomena are located. Here, increased hydrothermal activity has caused relevant morphological changes that has led local authorities to deny access to the site. Stratigraphic, structural, and geophysical investigations have allowed us to reconstruct the volcano‐tectonic setting of the area. In particular, we have recognized a fault system and related damage zones that act as the preferred pathway for hydrothermal fluids in the caldera. At the surface, these faults control the migration and/or accumulation of deep‐seated gases into the subsoil and the formation of fumaroles and mud pools. We have recognized two main fault systems with different ages that show variable displacements. The electrical anomalies identified by electrical resistivity tomography further highlight the main fault pattern and show the interplay between volcano‐tectonic structures and fluid circulation. Host rocks and fault zones may be involved in self‐sealing processes and/or rock failure phenomena capable of modifying the fluid pathways and establishing favorable conditions, leading to overpressure and/or rapid decompression of fluids and triggering an explosive event. Furthermore, stratigraphic mapping shows fossil mud pool sediments embedded in an old debris flow located above the modern hydrothermal system. This implies that they were at a higher elevation when they formed. The morphotectonic evolution and intense rock alterations in the area could promote further landslide episodes, producing debris‐flow deposits that can cover the active area and possibly trigger hydrothermal/phreatic events.

Landslide detection and susceptibility mapping using geological and remote sensing data: A case study of Azad Kashmir, NW Sub-Himalayas

Azad Kashmir has undergone sever landslide episodes in the past. The main reasons behind the occurrence of landslides in this area are heavy rainfall, weak geological strata, seismic zones and active faults. Several episodes of roads blockage due to landslide is one of the major problem in this area, therefore, this study is carried out for landslide susceptibility analysis along the roads of Kotli to Trar Khel, Azad Kashmir, NW Sub-Himalayas. Geographic information system (GIS) is used to interpret remote sensing, geological strata and topographical data. Total ten anthropogenic and physical factors (aspects, fault, lithology, drainage density, curvature, distance to roads, landuse landcover, slope, seismic and elevation) were examined by Multi-criteria Decision Analysis (MCDA) technique. The thematic layers of all factors were prepared for the preparation of the final susceptibility map by using a weighted overlay method (WOM). The susceptibility map classified the study area into four susceptible classes of low, moderate, high and very high. The spatial analysis and result outcomes depicted that active faults, slope gradient and landuse landcover may facilitate landslide phenomena. The results were verified by landslide density analyses (LDA), receiver operator characteristics (ROC) and area under curve (AUC) with predictive accuracy of 77 % that is rated as satisfactory by many researchers.

Analysis of landslide-induced fatalities and injuries in Bangladesh: 2000-2018

Landslides are one of the most substantial damaging and recurrent disasters in the hilly areas of Bangladesh. In the last decades, devastating landslides have repeatedly hit the hilly regions of the country, typically due to climate change along with other anthropogenic causes such as high population density, unplanned land use, and rampant hill cutting. However, there is a lack of an official database to analyze the trends of the landslide disaster in Bangladesh. In this vein, this paper compiled a catalog of 204 reported landslides in Bangladesh that caused 727 deaths and 1017 injuries during the period 2000–2018. The average number of annual landslides is 19, with a 4% yearly growth rate, which results in 38 fatalities and 54 injuries yearly on average. Using data from three national and two regional newspapers, the study finds that monsoon summer is the most damaging season and 83% of landslides are primarily triggered by heavy rainfall. A Hot Spot Analysis tool (Getis-Ord- Gi* algorithm) was used to identify the spatial concentration of high and low landslide affected areas in Bangladesh. Of 64 districts in the country, eight were exposed by landslide episodes, five were recognized as profoundly affected. Moreover, three sub-districts (Chattogram City Corporation, Ukhiya and Teknaf) have been identified as hot spot areas. Finally, the study will assist academicians for continuing risk studies through understanding the trends of landslide related causalities; policymakers for formulating suitable policies considering the high landslide affected areas; consequently, help local communities to protect their life and properties through raising awareness.

Investigating the effects of different landslide positioning techniques, landslide partitioning approaches, and presence-absence balances on landslide susceptibility mapping

The Ziarat Watershed, located in the south of the Golestan Province, Iran, has witnessed several destructive landslide episodes, prompting a number of researchers to aspire to improve landslide susceptibility modeling (LSM) techniques. We constructed three scenarios focusing on landslide positioning techniques (pixel-based, centroid, crown, and toe), training/test sampling strategies (Mahalanobis distance (MD), and random sampling (RS)), with alternative landslide/non-landslide data balances (1:1, 1:2, and 1:3). The data mining boosted regression trees (BRT) model was used for the landslide susceptibility modeling, using landslide data and 13 landslide controlling factors for the Ziarat Watershed. The performance of the scenarios was assessed using the areas under the success and prediction rate curves (AUSRC and AUPRC). A combination of pixel-based–MD–1:2 showed the highest learning capability and goodness-of-fit with an AUSRC value of 0.87, and the highest predictive power and generalization capacity with an AUPRC value of 0.79. Conversely, centroid-based–1:3–RS, crown-based–1:3–RS, and toe-based–1:3–RS performed less well. Comparatively, the pixel-based, MD, and 1:2 data balance scenarios surpassed their counterparts and outperformed the other models. The results indicated a high spatial differentiation with a significant chi-square value of 4549.46 at 95% confidence level. Moreover, 15.21% of the study area, containing almost 50% of the landslides, was found to have a high susceptibility to landslides. According to the premier scenario (pixel-based–MD–1:2), lithological formation, distance from roads, and NDVI, with respective contributions of 31.4%, 12.9%, and 12%, are the main spatial controlling factors leading to landslide occurrences in the study area.

Hydrothermal eruptions at unstable crater lakes: Insights from the Boiling Lake, Dominica, Lesser Antilles

The Boiling Lake - a major touristic attraction in the Morne Trois Pitons National Park, Dominica's (West Indies) World Heritage site - is located in an area of high hydrothermal activity in the southern part of the island. This spectacular lake is characterized by unstable behavior which has been interpreted to result from the presence of a water body above a fluctuating gas-filled, or multi-phase vent system. The system produces a near-boiling lake that occasionally drains and refills with water, without any simple periodicity. Despite the fact that such unstable draining lakes are thought to be non-eruptive, the Boiling Lake is known to have sourced occasional small hydrothermal eruptions as well as toxic gas emissions which accompany the refilling phase of these drainage events. Similar explosive events seem also to be likely after landslide slumped rocks from steep altered crater walls cover the vent area in the lake leading to overpressurized conditions. In both cases, rock expulsion and gas release continue to pose serious threats for tourists and tour guides. While conceptual models exist describing possible mechanisms of the Boiling Lake drainage itself, little information exists to date regarding the trigger mechanisms related to the explosive events. Further, no estimation of the eruption processes and dynamics of these events has previously been attempted. To address this knowledge gap, we employ here a combined field- and laboratory-based approach in order to i) characterize the properties of the dominant lithologies from the crater walls at Boiling Lake, ii) determine how the ongoing alteration may affect their petrophysical properties (e.g., porosity and permeability), and iii) evaluate how water-saturated, unconsolidated wall rock materials behave when rapidly decompressed from a slightly elevated pressure and temperature conditions.Here, we propose that it is the petrophysical state of the materials involved (e.g. porosity, permeability and alteration-related strength) which controls the style and energy of these eruptions, as well as the potential for landslides from the crater walls. The experimentally simulated scenarios included in this study are: i) decompression from only slightly elevated pressure conditions inferred for small hydrothermal eruptions occurring after lake drainage events, and ii) decompression from rapidly enhanced pressure conditions produced by the capping and overpressurization of the vent area following landslide episodes. Ejection velocities obtained from rapid decompression experiments were used as input parameters for ballistic trajectory calculations, allowing for estimation of the potential area impacted by ejected debris.We anticipate that this study will serve as a contribution to the understanding of the hazard potential of future hydrothermal eruptions at unstable crater lakes in general, as well as providing direct insights into the ongoing hydrothermal activity at the Boiling Lake.

Probabilistic Rainfall Thresholds for Landslide Episodes in the Sierra Norte De Puebla, Mexico

The Sierra Norte de Puebla, Mexico, has a record of hundreds of mass removal processes triggered by rainfall, where the intensity and duration of the rain are the main mechanisms. In order to determine threshold values for precipitation as a cause of a landslide, the prior, marginal and conditional probabilities were calculated. A Bayesian method was used for one-dimensional (precipitation intensity) and two-dimensional (precipitation intensity and duration) analysis. This suggested a high probability of mass movement when the precipitation exceeds 60 mm within ten days. A proposed warning system is based on classes in which the threshold is exceeded.

Integrated use of GIS and remote sensing for monitoring landslides in transportation pavements: the case study of Paphos area in Cyprus

This study considers the impact of landslides on transportation pavements in rural road network of Cyprus using remote sensing and geographical information system (GIS) techniques. Landslides are considered to be one of the most extreme natural hazards worldwide, causing both human losses and severe damages to the transportation network. Risk assessment for monitoring a road network is based on the combination of the probability of landslides occurrence and the extent and severity of the resultant consequences should the disasters (landslides) occur. Factors that can trigger landslide episodes include proximity to active faults, geological formations, fracture zones, degree and high curvature of slopes, water conditions, etc. In this study, the reliability and vulnerability of a rural network are examined. Initially, landslide locations were identified from the interpretation of satellite images. Different geomorphological factors such as aspect, slope, distance from the watershed, lithology, distance from lineaments, topographic curvature, land use and vegetation regime derived from satellite images were selected and incorporated in GIS environment in order to develop a decision support and continuous landslide monitoring system of the area. These parameters were then used in the final landslide hazard assessment model based on the analytic hierarchy process method. The results indicated good correlation between classified high-hazard areas and field-confirmed slope failures. The CA Markov model was also used to predict the landslide hazard zonation map for 2020 and the possible future hazards for transportation pavements. The proposed methodology can be used for areas with similar physiographic conditions all over the Eastern Mediterranean region.

Complex rupture mechanism and topography control symmetry of mass-wasting pattern, 2010 Haiti earthquake

The 12 January 2010 Mw 7.0 Haiti earthquake occurred in a complex deformation zone at the boundary between the North American and Caribbean plates. Combined geodetic, geological and seismological data posited that surface deformation was driven by rupture on the Léogâne blind thrust fault, while part of the rupture occurred as deep lateral slip on the Enriquillo–Plantain Garden Fault (EPGF). The earthquake triggered >4490 landslides, mainly shallow, disrupted rock falls, debris-soil falls and slides, and a few lateral spreads, over an area of ~2150km2. The regional distribution of these slope failures defies those of most similar earthquake-triggered landslide episodes reported previously. Most of the coseismic landslides did not proliferate in the hanging wall of the main rupture, but clustered instead at the junction of the blind Léogâne and EPGF ruptures, where topographic relief and hillslope steepness are above average. Also, low-relief areas subjected to high coseismic uplift were prone to lesser hanging wall slope instability than previous studies would suggest. We argue that a combined effect of complex rupture dynamics and topography primarily control this previously rarely documented landslide pattern. Compared to recent thrust fault-earthquakes of similar magnitudes elsewhere, we conclude that lower static stress drop, mean fault displacement, and blind ruptures of the 2010 Haiti earthquake resulted in fewer, smaller, and more symmetrically distributed landslides than previous studies would suggest. Our findings caution against overly relying on across-the-board models of slope stability response to seismic ground shaking.

Rainfall patterns and critical values associated with landslides in Povoação County (São Miguel Island, Azores): relationships with the North Atlantic Oscillation

AbstractSão Miguel Island (Azores) has been affected by hundreds of destructive landslide episodes in the last five centuries, triggered either by earthquakes, volcanic eruptions or rainfall episodes, which were responsible for many deaths and very important economic losses.Among the instability causes, meteorological factors are of primary importance on Povoação County, namely the high recurrence rate of calamitous rainfall triggering landslides. The most recent catastrophic episode took place on the 31st October 1997 when almost 1000 soil slips and debris flows were triggered, and 29 people died in the Ribeira Quente village.The role of rainfall on regional landslide activity was analysed applying cumulative rainfall methods. The method comprises the reconstruction of both absolute and calibrated antecedent rainfalls associated with each major landslide event. The critical rainfall combination (amount‐duration) responsible for each landslide event was assessed and a rainfall critical threshold for landslide occurrence was calculated. Rainfall‐triggered landslides in the study area are ruled by the function I = 144·06 D−0·5551, and they are related both to short duration precipitation events (1–3 days) with high average intensity (between 78 and 144 mm/day) and long‐lasting rainfall episodes (1–5 months) with a lower intensity (between 9 and 22 mm/day).The impact of the North Atlantic Oscillation (NAO) on the regional precipitation regime was evaluated. It is shown that the monthly precipitation of São Miguel is largely modulated by the NAO mode presenting a significant negative correlation with the NAO index. This result arises from the NAO control on the travelling latitude of most storm tracks that cross the Northern Atlantic Ocean. Copyright © 2007 John Wiley & Sons, Ltd.

Growth and collapse of the Reunion Island volcanoes

This work presents the first exhaustive study of the entire surface of the Reunion Island volcanic system. The focus is on the submarine part, for which a compilation of all multibeam data collected during the last 20 years has been made. Different types of submarine features have been identified: a coastal shelf, debris avalanches and sedimentary deposits, erosion canyons, volcanic constructions near the coast, and seamounts offshore. Criteria have been defined to differentiate the types of surfaces and to establish their relative chronology where possible. Debris avalanche deposits are by far the most extensive and voluminous formations in the submarine domain. They have built four huge Submarine Bulges to the east, north, west, and south of the island. They form fans 20–30 km wide at the coastline and 100–150 km wide at their ends, 70–80 km offshore. They were built gradually by the superimposition and/or juxtaposition of products moved during landslide episodes, involving up to several hundred cubic kilometers of material. About 50 individual events deposits can be recognized at the surface. The landslides have recurrently dismantled Piton des Neiges, Les Alizes, and Piton de La Fournaise volcanoes since 2 Ma. About one third are interpreted as secondary landslides, affecting previously emplaced debris avalanche deposits. On land, landslide deposits are observed in the extensively eroded central area of Piton des Neiges and in its coastal areas. Analysis of the present-day topography and of geology allows us to identify presumed faults and scars of previous large landslides. The Submarine Bulges are dissected and bound by canyons up to 200 m deep and 40 km long, filled with coarse-grained sediments, and generally connected to streams onshore. A large zone of sedimentary accumulation exists to the north–east of the island. It covers a zone 20 km in width, extending up to 15 km offshore. Volcanic constructions are observed near the coast on both Piton des Neiges and Piton de la Fournaise volcanoes and are continuations of subaerial structures. Individual seamounts are present on the submarine flanks and the surrounding ocean floor. A few seem to be young volcanoes, but the majority are probably old, eroded seamounts. This study suggests a larger scale and frequency of mass-wasting events on Reunion Island compared to similar islands. The virtual absence of downward flexure of the lithosphere beneath the island probably contributes to this feature. The increased number of known flank–failure events has to be taken into consideration when assessing hazards from future landslides, in particular, the probability of landslide-generated tsunamis.

Lichenometric dating of landslide episodes in the Western part of the Polish Flysch Carpathians

Landslides are common in the Polish part of the Flysch Carpathians. They are controlled mostly by lithology and tectonics of the bedrock and by climate. They are a major factor of slope transformation. Rock cliffs within deep landslides are overgrown with lichen thalli of the Rhizocarpon genus, so these cliffs may be dated using lichenometry. This paper presents the basics of lichenometry as well as its applications in geomorphological research in Poland and worldwide. Also presented are results of lichenometric dating of landslide–rockfall walls in the western part of the Polish Flysch Carpathians. The study helps to identify landslide episodes from historical times. The results obtained indicate that on the slopes studied there coexist several generations of landslide–rockfall forms of various age, including recent and active ones.

The Influence of the North Atlantic Oscillation on Rainfall Triggering of Landslides near Lisbon

The majority of landsliding episodes in the area north of Lisbon are associated with rainfall events of short (less than 5 days) medium (5-20 days) or long duration (more than 20 days). The precipitation regime in Portugal is highly irregular, with large differences between wet and dry years. We have assessed the impact of the North Atlantic Oscillation (NAO) on both the winter precipitation and the timing and magnitude of associated landslide events. Results show that the large inter-annual variability of winter precipitation is largely modulated by the NAO mode. The precipitation composite corresponding to high NAO index presents a considerable lower median value (47 mm/month) than the corresponding low NAO index class (134 mm/month). The entire precipitation distribution associated with the low NAO index composite encompasses a wider range of values than the corresponding high NAO index composite. This non-linear behavior is reflected in the probability of occurrence of a very wet month (precipitation above the 90% percentile) that is just 1% for the positive NAO class and 23% for low NAO index months. Results for the low NAO class are crucial because these months are more likely associated with long-lasting rainfall episodes responsible for large landslide events. This is confirmed by the application of a 3-month moving average to both NAO index and precipitation time series. This procedure allowed the identification of many months with landslide activity as being characterized by negative average values of the NAO index and high values of average precipitation (above 100 mm/month). Finally, using daily data we have computed the return periods associated with the entire set of landslide episodes and, based on these results, obtained a strong linear relationship between critical cumulative rainfall and the corresponding critical rainfall event duration.

Inventory of landslides and susceptibility mapping in the Dessie area, northern Ethiopia

A comprehensive study of landslide processes was carried out in the city of Dessie and its environs in northern Ethiopia. The study area consists of a graben bounded by steep fault scarps and characterized by highly weathered Tertiary basaltic rocks covered with Quaternary deposits. A method of landslide susceptibility mapping that adapts to the area of the hillside (about 16 km 2) is presented. First, a preliminary zoning was made by dividing the area into a discrete number of smaller units having similar susceptibility to sliding, followed by semi-quantitative field investigation of the geological, geotechnical, geomorphological, hydrogeological, and anthropogenic factors that contribute to the susceptibility of the units to landslide activity. The final landslide susceptibility map was accomplished by using classical overlay procedures of established maps representing the different influencing factors. The validity of the method was checked by means of a landslide-hazard map that showed the actual distribution of landslide episodes that have occurred in the past. Finally, four broad landslide susceptibility zones and 22 specific active landslide sites were identified. The most important landslide types were complex earth and debris slides and flows in silty clay soils associated with alluvial and colluvial deposits overlying highly weathered basalts. Intact basaltic rocks experienced rock fall and toppling in steep slope areas. Large-scale landslides have been initiated mostly by surface water and groundwater during the wet season.

Cultural Collapses in the Northwest: A Reply to Ian Kuijt

We previously suggested that the abandonment of large housepit villages along the Mid-Fraser River ca. 1200 B.P. was due to landslides that decimated salmon stocks and the main subsistence staples of the villages. Kuijt has questioned the plausibility of such a scenario. However, he confuses the two landslide episodes we discuss and raises unrealistic expectations about lake formation, migration behavior, and residual populations. He also confuses the lacustrine vs. fluvial effects of landslide blockages. While he attributes to us specific interpretations to which we are not tied and which are not critical for our basic argument, he offers no real alternative explanation for why the large Mid-Fraser villages were abandoned.

Prediction of rainfall‐triggered landslides: a test of the Antecedent Water Status Model

A rainfall-based landslide-triggering model, developed from previous landslide episodes in Wellington City, New Zealand, is tested for its ability to provide a 24-hour forecast of landslide occurrence. The model, referred to as the Antecedent Water Status Model, calculates an index of soil water, by running a daily water balance and applying a soil drainage factor to excess precipitation, over the preceding ten days. Together with the daily rainfall input, the soil water status has been used empirically to identify a threshold condition for landslide triggering. The prediction process provides a daily update of the soil water status and thereby the amount of rainfall required on the following day to equal or exceed the triggering threshold. The probability that this triggering rainfall will occur is then determined from the frequency/magnitude distribution of the local rainfall record. The model produces a satisfactory level of prediction, particularly for periods of concentrated landslide activity. Copyright © 1999 John Wiley & Sons, Ltd.

Prediction of rainfall-triggered landslides: a test of the Antecedent Water Status Model

A rainfall-based landslide-triggering model, developed from previous landslide episodes in Wellington City, New Zealand, is tested for its ability to provide a 24-hour forecast of landslide occurrence. The model, referred to as the Antecedent Water Status Model, calculates an index of soil water, by running a daily water balance and applying a soil drainage factor to excess precipitation, over the preceding ten days. Together with the daily rainfall input, the soil water status has been used empirically to identify a threshold condition for landslide triggering. The prediction process provides a daily update of the soil water status and thereby the amount of rainfall required on the following day to equal or exceed the triggering threshold. The probability that this triggering rainfall will occur is then determined from the frequency/magnitude distribution of the local rainfall record. The model produces a satisfactory level of prediction, particularly for periods of concentrated landslide activity. Copyright © 1999 John Wiley & Sons, Ltd.