Discriminant analysis of the geomorphic characteristics and stability of landslide dams

Landslides can cause the formation of dams, but these dams often fail soon after lake formation. Thus, rapidly evaluating the stability of a landslide dam is crucial for effective hazard mitigation. This study utilizes discriminant analysis based on a Japanese dataset consisting of 43 well documented landslide dams to determine the significant variables, including log-transformed peak flow (or catchment area), and log-transformed dam height, width and length in hierarchical order, which affect the stability of a landslide dam. The high overall prediction power (88.4% of the 43 training cases are correctly classified) and the high cross-validation accuracy (86%) demonstrate the robustness of the proposed discriminant models PHWL (with variables including log-transformed peak flow, and log-transformed dam height, width and length) and AHWL (with variables including log-transformed catchment area, and log-transformed dam height, width and length). Compared to a previously proposed “DBI” index-based graphic approach, the discriminant model AHV – which uses the log-transformed catchment area, dam height, and dam volume as relevant variables – shows better ability to evaluate the stability of landslide dams. Although these discriminant models are established using a Japanese dataset only, the present multivariate statistical approach can be applied for an expanded dataset without any difficulty when more completely documented worldwide landslide-dam data are available.     

Predicting landslides for risk analysis — Spatial models tested by a cross-validation technique

A landslide-hazard map is intended to show the location of future slope instability. Most spatial models of the hazard lack reliability tests of the procedures and predictions for estimating the probabilities of future landslides, thus precluding use of the maps for probabilistic risk analysis. To correct this deficiency we propose a systematic procedure comprising two analytical steps: “relative-hazard mapping” and “empirical probability estimation”. A mathematical model first generates a prediction map by dividing an area into “prediction” classes according to the relative likelihood of occurrence of future landslides, conditional by local geomorphic and topographic characteristics. The second stage estimates empirically the probability of landslide occurrence in each prediction class, by applying a cross-validation technique. Cross-validation, a “blind test” here using non-overlapping spatial or temporal subsets of mapped landslides, evaluates accuracy of the prediction and from the resulting statistics estimates occurrence probabilities of future landslides. This quantitative approach, exemplified by several experiments in an area near Lisbon, Portugal, can accommodate any subsequent analysis of landslide risk.     

Seven landslide dams of old seismic origin in southeastern Sicily (Italy)

This paper focuses on origin, morphology and evolution of seven landslide dams in southeastern Sicily. These landforms are part of a set of 146 landslides recently recognised in this area, which was hitherto considered to have little or no slope instability. Southeastern Sicily consists of a plateau (the Hyblaean Mountains) incised by canyons and surrounded by lower lands. It is underlain mostly by subhorizontal, moderately to well-lithified carbonate rocks. Relief is low.Several lines of evidence justify the assumption of a seismic trigger for the landslides in this area: (1) the geo-climatic environment is not favourable to landsliding, (2) low-angle basal shear surfaces are very frequent, (3) landslide distribution is consistent with the known magnitude–distance relationships for earthquake-induced landslides, (4) historical documents testify to earthquake-triggered slope instability and (5) a specific landslide can be exactly dated.The phenomena illustrated here include six rock slides (one with a debris-flow component) and one rock fall. Slip surfaces are mostly non-circular. Landslide volume ranges from about 50×10³ to 34×10⁶ m³.With reference to the Costa and Schuster [Geol. Soc. Am. Bull. 100 (1988) 1054] classification of landslide dams, five cases belong to type II (spanning the entire valley), and two to type IV (failures from both valley sides, with frontal or side contact between failed masses). With reference to Crozier and Pillans [Catena 18 (1991) 471] classification of landslide lakes, all cases show a main valley lake while tributary valley, back and supra lakes are sporadically present. One damming is attributable to the 1693 earthquake with certainty; another damming, to the same earthquake with high probability. Three dams were reincised, one breached or reincised, one is slightly reincised and two more or less intact; correspondingly, five silting up deposits were reincised, one is being reincised at present and two are still under formation.     

Landforms and landscape evolution in the Skardu,Shigar and Braldu Valleys,Central Karakoram

The Central Karakoram, which includes K2 in Pakistan, is one of the most rapidly rising areas on Earth and exhibits complex topography and extreme relief. Impressive valley fills and glacial landforms are present throughout the valleys. The dynamics of landscape evolution of the region are currently not well understood. Consequently, the landforms were mapped and assessed in the Skardu, Shigar, and Braldu valleys, to elucidate the spatio-temporal scale dependencies of surface processes active in the region. These valleys were examined using geomorphic field methods, remote sensing, geomorphometry, and terrestrial cosmogenic nuclides (TCNs) surface exposure dating. The glaciers in this region have oscillated considerably throughout the Late Quaternary, and four glacial stages have been recognized including at least six glacial advances. Surface processes readjusted after glacier retreat, and ubiquitous mass movements and catastrophic landsliding transported material from steep slopes to valley bottoms, while glaciofluvial meltwater and glacier outburst floods redistributed sediment down valley. Glacier geochronology and late Holocene ages of the outburst flood deposits indicate that landscape evolution has been dominated by glaciation and paraglaciation during the late Quaternary.     

The 1786 earthquake-triggered landslide dam and subsequent dam-break flood on the Dadu River, southwestern China

Chinese historic documents recorded that on June 1, 1786, a strong M=7.75 earthquake occurred in the Kangding-Luding area, Sichuan, southwestern China, resulting in a large landslide that fell into the Dadu River. As a result, a landslide dam blocked the river. Ten days later, the sudden breaching of the dam resulted in catastrophic downstream flooding. Historic records document over 100,000 deaths by the flood. This may be the most disastrous event ever caused by landslide dam failures in the world. Although a lot of work has been carried out to determine the location, magnitude and intensity of the 1786 earthquake, relatively little is known about the occurrence and nature of the landslide dam. In this paper, the dam was reconstructed using historic documents and geomorphic evidence. It was found that the landslide dam was about 70 m high, and it created a lake with a water volume of about 50×10⁶ m³ and an area of about 1.7 km². The landslide dam breached suddenly due to a major aftershock on June 10, 1786. The peak discharge at the dam breach was estimated using regression equations and a physically based predictive equation. The possibility of a future failure of the landslide seems high, particularly due to inherent seismic risk, and detailed geotechnical investigations are strongly recommended for evaluating the current stability of the landslide.     

美国西部滑坡坝的合作研究

沟谷的滑坡成坝已成为美国西部山地中一个值得注意的问题。1985年作者们考察了美国西部六州九个滑坡坝的特点。这是执行1979年中美科学技术合作协议(特别是1980年地学合作补充议定书)的部分内容。中美两国科学家和工程专家考察九个滑坡坝后,取得了一些研究成果,即论述了滑坡坝(尤其是近代滑坡坝)的成因与稳定性,并列举了灾变性溃坝防止措施的若干成功经验。     

Flood assessment and early warning of the reoccurrence of river blockage at the Baige landslide

On 10th Oct.and 3rd Nov.2018,two successive landslides occurred in the Jinsha River catchment at Baige Village,Tibet Autonomous Region,China.The landslides blocked the major river and formed the barrier lake,which finally caused the huge flood disaster loss.The hillslope at Baige landslide site has been still deforming after the 2018 slidings,which is likely to fail and block the Jinsha River again in the future.Therefore the investigation of 2018 flood disaster at the Baige landslide is of a great significance to provide a classic case for flood assessment and early warning for the future disaster.The detailed survey revealed that the outstanding inundations induced bank collapse disasters upstream the Baige landslide dams,and the field investigations and hydrological simulation suggested that the downstream of the Baige landslide were seriously flooded due to the two periods of the outburst floods.On these bases,the early warning process of potential outburst floods at the Baige landslide was advised,which contains four stages:Outburst Flood Simulating Stage,Outburst Flood Fore-casting Stage,Emergency Plan and Emergency Evacuation Stage.The study offers a con-ceptual model for the mitigation of landslides and flood disasters in the high-relief mountain-ous region in Tibet.     

Response and recovery of the Eel River, California, and its tributaries to floods in 1955, 1964, and 1997

Northwestern California is prone to regional, high magnitude winter rainstorms, which repeatedly produce catastrophic floods in the basins of the northern Coast Ranges. Major floods on the Eel River in 1955 and 1964 resulted in substantial geomorphic changes to the channel, adjacent terraces, and tributaries. This study evaluated the changes and the effects of a moderate flood in 1997 through field observations and examination of aerial photographs that spanned from 1954 to 1996. The purpose was to document the nature and magnitude of geomorphic responses to these three floods and assess the rates and controls on the recovery of the Eel River and its tributaries. Channel widening from extensive bank erosion was the dominant geomorphic change along the lower Eel River during major floods. As a result of the 1964 flood, the largest amount of widening was 195 m and represented an 80% change in channel width. Channel narrowing characterized the periods after the 1955 and 1964 floods. More than 30 years after the 1964 flood, however, the river had not returned to pre-flood width, which suggests that channel recovery required decades to complete. A long recovery time is unusual given that the Eel River is located in an area with a “superhumid” climate and has an exceptionally high sediment yield. This long recovery time may reflect highly seasonal precipitation and runoff, which are concentrated in 3–5 months each winter. In contrast to the main stem of the Eel River, the dominant effects of floods on the tributaries of the Eel River were rapid aggradation of channel bed and valley floor followed by immediate downcutting. Dendrogeomorphic data, aerial photographs, and field observations indicate that thick wedges of gravel, derived largely from hillslope failures in upper reaches of the tributaries, are deposited at and immediately upstream of the mouths of tributaries as the stage of the Eel River exceeded that of the tributaries during major floods. In the waning stages of the flood, the tributaries cut through the gravel at a rate equal to the lowering of the Eel and generated unpaired terraces and nickpoints. The complete process of deposition and incision can occur within a few days of peak discharge. Although reworking of some sediment on the valley floor may continue for years after large floods, channel morphology in the tributaries appears to be a product of infrequent, high magnitude events. The morphology of the tributary channel also appears to be greatly influenced by the frequency and magnitude of mass wasting in headwater areas of small basins.     

Magnitude–frequency relationships of debris flows and debris avalanches in relation to slope relief

Probability of occurrence, hazard intensity and encounter probability are key parameters in the quantitative risk analysis (QRA) of landslides. All are strongly dependent on magnitude of the landslides. As a result, magnitude–frequency analysis should be a part of QRA. Deriving representative magnitude–frequency relationships for debris avalanches and debris flows, however, is difficult. One key problem is illustrated with the example of a unique database from the coastal region of British Columbia, Canada, which was compiled entirely from detailed ground investigations. The magnitude of debris avalanches and debris flows is not an independent statistical quantity, but a function of the scale of a given slope, as characterized by the slope length. Thus, attempting to derive probability and magnitude for a given location or sub-region from a regionally-derived magnitude–frequency curve may lead to incorrect predictions. The same problem is pertinent to the application of the same approach to any type of landslide in which the largest combined dimension of the source volume (including entrainment) is of the same order as the length of the slope. It is recommended that greater emphasis be placed on site-specific geological observations, at the expense of generalized statistics.     

Analysing the influence of upslope bedrock outcrops on shallow landsliding

A model for the prediction of topographic and climatic control on shallow landsliding in mountainous terrain is enhanced to analyse the impact of upslope rocky outcrops on downslope shallow landsliding. The model uses a ‘generalised quasi-dynamic wetness index’ to describe runoff propagation on bare rock surfaces connected to downslope soil-mantled topographic elements. This approach yields a simple enhanced model capable of describing the influence of upslope bedrock outcrops on the pattern of downslope soil saturation. The model is applied in both diagnostic and predictive modes to a small catchment in the eastern Italian Alps for which a detailed inventory of shallow landslides in areas dominated by rocky outcrops is available. In the diagnostic mode, the model is used with satisfactory results to reproduce the pattern of instability generated by an intense short-duration storm occurred on 14 September 1994, which triggered a large percentage of the surveyed landslides. In the predictive mode, the model is used for hazard assessment, and the return time of the critical rainfall needed to cause instability for each topographic element is determined. Modelling results obtained in the predictive mode are evaluated against all the surveyed landslides. It is revealed that the generalised quasi-dynamic model offers considerable improvement over the non-generalised quasi-dynamic model and the steady-state model in predicting existing landslides as represented in the considered landslide inventory.     

February floods in the lower North Island, 2004: Catastrophe – causes and consequences

Abstract: The February 2004 Manawatu floods in New Zealand were the result of a naturally occurring, although unusual, storm. Up to 300 mm of rain fell on the already saturated ground of the lower North Island over two days, generating substantial and rapid runoff from catchment slopes. Rivers rose quickly, inundating unprotected farmland and properties and in places breaching stopbanks. There was widespread slope failure in the hill country of the lower North Island, affecting an area of ca. 7500 km². Slopes under scrub, plantation forest and native bush were not as badly affected as those under pasture, where slopes typically failed by shallow translational landsliding. Flooding caused catastrophic channel change in a number of small to medium sized channel systems in the upland fringes. Whilst the occurrence of landsliding and channel changes during an extreme event such as this is natural, the intensity of both landsliding and channel erosion was exacerbated by human activity within the catchments.     

Formation and failure of volcanic debris dams in the Chakachatna River valley associated with eruptions of the Spurr volcanic complex, Alaska

The formation of lahars and a debris avalanche during Holocene eruptions of the Spurr volcanic complex in south-central Alaska have led to the development of volcanic debris dams in the Chakachatna River valley. Debris dams composed of lahar and debris-avalanche deposits formed at least five times in the last 8000–10,000 years and most recently during eruptions of Crater Peak vent in 1953 and 1992. Water impounded by a large debris avalanche of early Holocene (?) age may have destabilized an upstream glacier-dammed lake causing a catastrophic flood on the Chakachatna River. A large alluvial fan just downstream of the debris-avalanche deposit is strewn with boulders and blocks and is probably the deposit generated by this flood. Application of a physically based dam-break model yields estimates of peak discharge (Q_p) attained during failure of the debris-avalanche dam in the range 10⁴<Q_p<10⁶ m³ s⁻¹ for plausible breach erosion rates of 10–100 m h⁻¹. Smaller, short-lived, lahar dams that formed during historical eruptions in 1953, and 1992, impounded smaller lakes in the upper Chakachatna River valley and peak flows attained during failure of these volcanic debris dams were in the range 10³<Q_p<10⁴ m³ s⁻¹ for plausible breach erosion rates.Volcanic debris dams have formed at other volcanoes in the Cook Inlet region, Aleutian arc, and Wrangell Mountains but apparently did not fail rapidly or result in large or catastrophic outflows. Steep valley topography and frequent eruptions at volcanoes in this region make for significant hazards associated with the formation and failure of volcanic debris dams.     

Hydro-geomorphic hazards and impact of man-made structures during the catastrophic flood of June 2000 in the Upper Guil catchment (Queyras, Southern French Alps)

The Guil River Valley (Queyras, Southern French Alps) is prone to catastrophic floods, as the long historical archives and Holocene sedimentary records demonstrate. In June 2000, the upper part of this valley was affected by a “30-year” recurrence interval (R.I.) flood. Although of lower magnitude and somewhat different nature from that of 1957 (>100-year R.I. flood), the 2000 event induced serious damage to infrastructure and buildings on the valley floor. Use of methods including high-resolution aerial photography, multi-date mapping, hydraulic calculations and field observations made possible the characterisation of the geomorphic impacts on the Guil River and its tributaries. The total rainfall (260 mm in four days) and maximum hourly intensity (17.3 mm h⁻¹), aggravated by pre-existing saturated soils, explain the immediate response of the fluvial system and the subsequent destabilisation of slopes. Abundant water and sediment supply (landsliding, bank erosion), particularly from small catchment basins cut into slaty, schist bedrock, resulted in destructive pulses of debris flow and hyperconcentrated flows. The specific stream power of the Guil and its tributaries was greater than the critical stream power, thus explaining the abundant sediment transport. The Guil discharge was estimated as 180 m³ s⁻¹ at Aiguilles, compared to the annual mean discharge of 6 m³ s⁻¹ and a June mean discharge of 18 m³ s⁻¹. The impacts on the Guil valley floor (flooding, aggradation, generalised bank erosion and changes in the river pattern) were widespread and locally influenced by variations in the floodplain slope and/or channel geometry. The stream partially reoccupied former channels abandoned or modified in their geometry by various structures built during the last four decades, as exemplified by the Aiguilles case study, where the worst damage took place. A comparative study of the geomorphic consequences of both the 1957 and 2000 floods shows that, despite their poor maintenance, the flood control structures built after the 1957 event were relatively efficient, in contrast to unprotected places. The comparison also demonstrates the role of land-use changes (conversion from traditional agro-pastoral life to a ski/hiking-based economy, construction of various structures) in reducing the Guil channel capacity and, more generally, in increasing the vulnerability of the human installations. The efficiency of the measures taken after the 2000 flood (narrowing and digging out of the channel) is also assessed. Final evaluation suggests that, in such high mountainous environments, there is a need to keep most of the 1957 flooded zone clear of buildings and other structures (aside from the existing villages and structures of particular economic interest), in order to enable the river to migrate freely and to adjust to exceptional hydro-geomorphic conditions without causing major damage.     

Regional relief characteristics and denudation pattern of the western Southern Alps, New Zealand

The Southern Alps of New Zealand are the topographic expression of active oblique continental convergence of the Australian and Pacific plates. Despite inferred high rates of tectonic and climatic forcing, the pattern of differential uplift and erosion remains uncertain. We use a 25-m DEM to conduct a regional-scale relief analysis of a 250-km long strip of the western Southern Alps (WSA). We present a preliminary map of regional erosion and denudation by overlaying mean basin relief, a modelled stream-power erosion index, river incision rates, historic landslide denudation rates, and landslide density. The interplay between strong tectonic and climatic forcing has led to relief production that locally attains 2 km in major catchments, with mean values of 0.65–0.68 km. Interpolation between elevations of major catchment divides indicates potential removal of l0¹–10³ km³, or a mean basin relief of 0.51–0.85 km in the larger catchments. Local relief and inferred river incision rates into bedrock are highest about 50–67% of the distance between the Alpine fault and the main divide. The mean regional relief variability is ± 0.5 km.Local relief, valley cross-sectional area, and catchment width correlate moderately with catchment area, and also reach maximum values between the range front and the divide. Hypsometric integrals show scale dependence, and together with hypsometric curves, are insufficient to clearly differentiate between glacial and fluvial dominated basins. Mean slope angle in the WSA (ψ = 30°) is lower where major longitudinal valleys and extensive ice cover occur, and may be an insensitive measure of regional relief. Modal slope angle is strikingly uniform throughout the WSA (φ = 38–40°), and may record adjustment to runoff and landsliding. Both ψ and φ show non-linear relationships with elevation, which we attribute to dominant geomorphic process domains, such as fluvial processes in low-altitude valley trains, surface runoff and frequent landsliding on montane hillslopes, “relief dampening” by glaciers, and rock fall/avalanching on steep main-divide slopes.     

山洪泥石流风险评估与风险管理理论与方法

山洪泥石流是中国常见的自然灾害,充分认识其形成机制与潜在风险是防灾减灾的关键。本文阐述了山洪泥石流形成机理,以及风险分析与管理的方法和内容,系统认识了地表产流流量激增、土体破坏物质供给激增、沟道堵塞体级联溃决流量放大和动床侵蚀规模增大等4个山洪泥石流的形成过程,介绍了基于动力过程的山洪泥石流风险评估方法和承灾体易损性评估方法,构建了基于灾害动力过程的风险评估与风险制图方法。进而,基于风险评估结果,提出可用于具体灾害点减灾的风险管理内容和风险调控技术、灾害防治的工程与非工程措施与制技术方案。最后,重点讨论了包括灾害风险预测、临灾预案、灾害防治工程方案等内容的风险处置对策,并形成一套基于山洪泥石流动力过程的风险评估与风险管理理论与方法体系。     

Predictive landslide susceptibility assessment using the weight-of-evidence method in north-central Garhwal Himalaya,India

Landslides are frequent natural disasters in mountainous regions, particularly in the Himalayas in India during the southwest monsoon season. Although scientific study of landslides has been in progress for years, no significant achievement has been made to preclude landsliding and allay disasters. This research was undertaken to understand the areal distribution of landslides based on geological formations and geomorphological processes, and to provide more precise information regarding slope instability and mechanisms of failure. After completing a landslide inventory, prepared through field investigation and satellite image analysis, 493 landslides, comprising 131 investigated in the field and 362 identified from satellite imagery, were identified and mapped. The areal distribution of these landslides shows that sites more prone to landsliding have moderate to steep slopes, the lithology of the Lesser Himalayan formations, and excavations for road corridors. Landslide susceptibility zones were delineated for the area using the weight-of-evidence method on the basis of the frequency and distribution of landslides. Weights of selected variables were computed on the basis of severity of triggering factors. The accuracy of landslide susceptibility zones, calculated statistically (R² = .93), suggests high accuracy of the model for predicting landsliding in the area.     

Aspects of landslide activity in the Mercantour Massif and the French Riviera, southeastern France

Landslides pose serious hazards in the Mercantour Massif and the French Riviera in southeastern France. The context for landslide development is a particularly favourable one, both in terms of the geomorphic and structural setting of this Alpine region, and of the climatic, hydrologic and seismic factors that trigger such failures. High mountain relief and steep slopes constitute a very favourable setting for failures affecting massive basement rocks and a very heterogeneous sedimentary cover whose resistance has been weakened by weathering, tectonic stresses, and cambering due to gravity. Among trigger factors, the important appears to be the precipitation regime. Rainfalls are commonly concentrated into short high-intensity downpours or into bursts of sustained falls over periods of several days, leading to soil saturation and lubrication of potential failure planes. Snowmelt also contributes to these lubrication processes. Earthquakes affecting this region are also a potentially important landslide trigger. However, while a lot of work has been done on the relationship between extreme climatic events and landslide activity, much less is known of the trigger effects of earthquakes.Both the background factors that promote landslide development and the factors that trigger such failures are discussed within a time frame of landslide development. Progressive changes in soil strength due to weathering, rock cambering and shattering processes lead to long-term reduction in resistance. Superimposed on these progressive changes are episodic triggerings related to rainfall and snowmelt episodes or earthquakes. Landslides may occur as shallow, low-volume “one-time” events or may be part of a progressive long-term failure. The former generally affect unconsolidated or clay-rich sedimentary rocks, especially on the coastal hillslopes of the French Riviera. A notable exception of a major, voluminous “one-time” event was the submarine landslide of the Var Delta in 1979. This landslide, like numerous other smaller subaerial landslides onland, was largely a result of human activities. This landslide occurred following extensive modification of the Var Delta and, notably, reclamation of the steep, fine-grained delta front. Deforestation, quarrying, urbanisation and road network developments are various ways in which human activity has destabilized the coastal hillslopes, favouring the development of numerous shallow landslides following each episode of heavy rainfall.Progressive landslides on the upper hillslopes of the Mercantour Massif have developed over long time spans (order of 10¹ to 10⁵ yrs) and have involved more complex interactions between lithological controls, slope characteristics and trigger factors. The Collelongue and Bois de Malbosc landslides have evolved into now stable failures buttressed by resistant migmatitic diorites or amphibolites. The more voluminous and well monitored Clapière landslide is a relatively simple rotational landslide of the toe-failure type. This active landslide poses a serious to inhabitants and infrastructure in the Tinée Valley. The importance of continued field monitoring, modelling and mapping of landslides and their hazards is emphasised.     

The geographic distribution of small dams in Oregon using ecoregion and landform classification

Abstract

Small dams fragment river landscapes, disrupting channel connectivity and impacting water quality and quantity. Although they impound volumetrically less total water than large dams, the ubiquity of small dams suggests their cumulative impacts could be significant. Documenting the distribution and characteristics of small dams is necessary to understanding and mitigating their impacts. In this study, we compare datasets of small dams in Oregon, compile a new comprehensive dataset, and document geographic variations in small dam distributions between different ecoregions. We used Oregon Water Resources Department dam and Oregon Department of Fish and Wildlife fish passage barrier datasets to compare dam heights and contributing drainage areas between different ecoregions. Over 61% of Oregon’s land area is above one or more small dam. We highlight the location of Oregon’s small dams at valley margins, transition zones between flat plains and mountains, and areas of high population density. Given the hidden nature of small dams, evaluation of small dam impacts using public imagery is not effective. However, knowledge of small dam distributions given their association with landforms can aid in finding unrecorded dams, assessing approaches for evaluating their geomorphic impacts, and informing their management.     

Geomorphological study of the Fadalto landslide, Venetian Prealps, Italy

This research deals with the Fadalto landslide (Lapisina Valley, Venetian Prealps), which took place in the Lateglacial and has continued its activity until today. Our aim is to recognize how the landslide failed, the causes of such failure and the activity of this landslide. The study of this landslide is important not only to understand the geomorphological history of this alpine area, and why the Piave River modified its course in the Late Pleistocene, but also the links with human activities, and specifically with the road and rail network.The geomorphological study, carried out by the interpretation of aerial photos and by a detailed field survey, has been integrated with a geological survey, geophysical investigations and a morphometric analysis (DTM). The Fadalto landslide is considered to be a rockslide reactivated in various phases, with different dimensions and with different characters (slides, slumps and flows). The landslides have been provoked by natural causes, both external and internal; the fundamental external causes are the retreat of the Würmian glacier and tectonic activity; the internal factors that decrease the shear resistance are the bedding planes and joints of the bedrock, the attitude of the rocks dipping towards the valley bottom and, as regards more recent failures, the presence of glacial deposits underlying the landslide debris. Besides, in recent times, we must also consider human activity as a cause of slope instability.As to the activity, the Fadalto landslide is defined “dormant”. This means that in this area there is a geomorphological risk connected with the important road and rail network of the Lapisina Valley.