Landslide Science and Practice

A proposal to supplement existing landslide classifications with a more detailed description of the geomechanical characteristics of the materials to include the effects of “structure”, stress history and initial state has been developed as part of the EC sponsored SafeLand project. This approach provides a valuable insight into and a rational basis for the modelling of the physical mechanisms that govern landslide triggering and subsequent development. The classification, behaviour and modelling of saturated clays and sands is summarized here. The use of advanced strain hardening plasticity models which, where necessary, include progressive damage to the “structure” of the material helps to replicate mechanical weathering, delayed failure and the triggering of flows or slides, depending on the type of material. A numerical example is presented, illustrating the different response of sensitive clays and mechanically overconsolidated clays to rapid erosion or excavation. More details of the proposed geotechnical classification and other numerical examples may be found in Deliverable 1.1 of the SafeLand project.

Mountain deformation and fracture as a result of earthquakes is a complicated evolution process. We need to take advantage of geomechanical simulation and shock effects to reproduce the mountain deformation process based on the understanding and conceptual model through geological analysis, in order to verify and disclose the facts. The author has chosen a certain typical landslide-type geomechanical model of consequent rock slope under earthquake effect, and carried out the simulation testing study for the geomechanical mechanism under the vibration conditions. This paper introduces the methods of model preparation, the test plan design and testing methods, and studies the testing results, from the vibration trace and deformation and fracture evolution process, deformation and fracture evolutionary process and vibration acceleration time-history changes, vibration intensity and deformation and fracture evolution of different geological structures, so some useful results and new knowledge are reported.

Landslides process is one of the most widespread and dangerous processes in the urbanized territories. In Moscow the landslips occupy about 3% of the most valuable territory of the city. In Russia many towns are located near rivers on high coastal sides. There are many churches and historical buildings on high costs of Volga River and Moscow River. The organization of monitoring is necessary for maintenance of normal functioning of city infrastructure in a coastal zone and duly realization of effective protective actions. Last years the landslide process activization took place in Moscow.

Studies assessing landslide hazards have become increasingly common in recent years. They provide information that guides decision-making and the adoption of preventive and mitigation measures for control and protection. The aim of this study was to develop a set of procedures for the prediction of landslides from the results of geotechnical mapping at scales larger than 1:10,000. A methodology based on geotechnical parameters using the calculation of three-dimensional (3D) safety factor combined with an infiltration model it was developed. This set of procedures was applied to pre-defined areas in the city of Campos do Jordão, SP, Brazil. These areas were chosen based on geomorphic variability of geological materials and size criteria. The results from these areas are very promising and reflect the geological, geotechnical and hydrogeological conditions in each area.

Lantau Island is located at the south-western part of Hong Kong. Over 75 % of its areas are natural hillsides and approximately one third of the hillsides are sloping at 30° or above. The geology comprises tuff and granite. On 7 June 2008, a severe rainfall with a return period of 1,100 years occurred over the western part of Lantau Island. This triggered over 2,400 landslides, about 900 of which were debris flows. Some of these were sizeable and mobile, involving active debris volumes and travel distances in excess of 10,000 m

3

and 1,700 m respectively. This paper examines the runout characteristics of selected mobile debris flows. Two-dimensional and three-dimensional dynamic models developed by the Geotechnical Engineering Office (GEO), Civil Engineering and Development Department were used to back analyse the long-runout debris flows, taking into account the complicated geometry of the runout paths. The findings and their engineering implications are presented in the paper.

The deterministic model SliDisp

+

calculates the potential detachment zones of shallow landslides. It is a grid-based model using an infinite slope analysis to calculate the safety factors F (ratio of retaining and driving forces) for each cell.

The input data consists of the slope topography, soil strength parameters, depths and shapes of potential shear planes, and the hydraulic behavior. The variables are derived from a digital elevation model (DEM), geological, geotechnical, and pedological documents, or field investigations. From this data the soil is classified over large areas. For each cell, the critical slope angle as well as the soil cohesion is determined.

Studies in several test areas showed that pedological aspects as well as joint water-input from underlying rock must be taken into account. Combined with the run-out model SliDepot, SliDisp

+

calculates the extent of potential landslides over large areas and thus can be applied for spatial planning and optimized positioning of protection measures.

The Grohovo Landslide, situated on the north-eastern slope in the central part of the Rječina River valley, is the largest active landslide along the Croatian part of the Adriatic Sea coast. The Grohovo Landslide is also a key pilot area for implementation of the monitoring activities of the Japanese-Croatian joint research project on “Risk identification and land-use planning for disaster mitigation of landslides and floods in Croatia”. This complex retrogressive landslide was reactivated in December 1996. It is considered that the basal failure surface is positioned at the contact between the slope deposits and the flysch bedrock. Based on the data from boreholes, geological mapping, geophysical surveys and the geological cross-section of the slope, the three dimensional shape and the position of the sliding surface were simulated using the inverse distance weighted interpolation. Using 3D extended Janbu’s simplified method, global stability of the Grohovo Landslide and stability of 12 separate landslide bodies were analyzed.

The Sierra Norte de Puebla (Mexico) is a tropical mountain frequently affected by torrential rains induced by depressions and hurricanes. In October 1999 and 2005, intense precipitation triggered hundreds of landslides and caused heavy human losses and material damages. All the landslides surveyed in the field and their traces extracted from the satellite images have been analyzed by using morphometric parameters in order to characterize and to classify them. The spatial distribution of observed landslides types is not random but responds to both regional structural features and material nature involved in the movement. It is particularly important to establish these close relationships in order to assess landslide hazards by using different factors and methods and the approach proposed in this paper seems to be very useful to do that.

In tropical mountainous regions, heavy rainfalls often induce landslides and mudflows. Mexico is continually subjected to natural hazards. In February 2010 the town of Angangueo in the state of Michoacan was devastated by a huge muddy debris flow. Use of morphological and climatic parameters of the affected areas can be incorporated into a model that simulates the extension of this natural disaster. It appears that zones whose slope aspect is against the flow direction represent the most vulnerable zones. The model described herein can easily and rapidly be applied to any other tropical region to reliably prevent such catastrophic phenomena. Because in many areas the colluvial slope deposits register multiple cracks and active scars, an assessment of the risk related to slope movements in Mexico is essential.

The paper presents the kinematics of rock instability of a high promontory, where Tithorea village is situated, in the Central Greece peninsula. The instability phenomena pose a significant threat on the town situated immediately down slope. Rock fall episodes occurred in the past, as it has been noted by local authorities and published reports. A more recent rockfall is investigated in this paper, which caused considerable damages to two village houses. The detailed rock fall path was mapped obtaining more than 15 hit points and rolled on different scree and vegetation material till it stopped and crushed upon the house roofs. The predominant types of kinematic instability are of planar or wedge failure and toppling of large blocks. In order to investigate the existing stability conditions and decide upon the protection measures, rockfall analysis was carried out. Such an analysis can be compared in detail with the actual situation, since the geometrical data of the recent rockfall incident are very accurate. On the other hand, other parameters of rockfall hazard such as the run-out distance of the boulders was examined.

Worldwide, most landslides are generally triggered by rainfalls. In this paper, the hydrological model

CM

SAKe

to forecast the timing of activation of slope movements is described.

Model calibration can be performed through an iterative algorithm (named “Cluster”): in this way, optimal kernels can be refined from an initial tentative solution. At each iteration, shape and base time of the tentative kernels are optimized by means of a discretized, self-adapting approach; the fitness is computed for all the examined solutions, and new individuals are generated starting from the best kernel obtained at the previous step.

The initial shape of the kernel can be selected among a set of standard basic types: (1) rectangular; (2) decreasing triangular; (3) any other geometrical shape which may sound reasonable on the base of the physical knowledge of the phenomenon. Alternatively, the kernel can be automatically optimized based on antecedent rainfalls with respect to a given date of landslide activation.

In this paper, a first example of application of the model to a large debris slide threatening a village in Northern Calabria is discussed, together with the results of a preliminary sensitivity analysis aimed at investigating the role of the main parameters.

The “Costa Viola” mountain ridge (Calabria) is exposed to severe geo-hydrological risk conditions, especially in the sector between Bagnara Calabra and Scilla. This sector has repeatedly been affected by slope instability events in the past, mainly related to debris slides, rock falls and debris flows.

An attempt of shallow-landslide susceptibility mapping has been performed for the mentioned coastal sector through a logistic regression (LR) approach. LR is a multivariate type of analysis that allows estimating the presence/absence of a phenomenon in terms of probability (ranging between 0 and 1), on the basis of linear statistical relationships with a set of independent territorial variables.

The adopted LR procedure consists of four steps: (1) variable parameterization, (2) sampling, (3) fitting, and (4) application. Obtained results can be considered acceptable, as 85.6% of cells are correctly classified.

With the aim of extracting 3D structural information of rock masses from high resolution remote sensing data, a Matlab tool, called DiAna (Discontinuity Analysis) has been compiled.

In particular, the proposed approach is able to semi-automatically retrieve some relevant rock mass parameters, namely orientation, number of sets, spacing/frequency, persistence, block size and scale dependent roughness, by analyzing high resolution point clouds acquired from terrestrial or aerial laser scanners.

The proposed method has been applied to different case studies, and the obtained properties have been compared with the results from traditional geomechanical surveys.

These applications demonstrated DiAna’s ability to investigate rock masses characterized by irregular block shapes, and suggest applications in the field of engineering geology and emergency management, when it is often advisable to minimize survey time in dangerous environments and, in the same time, it is necessary to gather all the required information as fast as possible.

In April 2009, North-West Italy was interested by heavy rainfalls that triggered several landslides, especially of shallow type, and caused relevant rise of water level in many rivers, in some cases even beyond the alert level. Particularly in the hills near Alba (NW Italy), many landslides occurred, most of them belonging to the debris flow or soil slip types. In this area, a small but interesting landslide involved a local road and a high quality, recently planted, vineyard. The present study shows the use of different disciplinary approaches focused to understand the behaviour of this landslide: in particular, besides geological and geomorphologic studies, detailed topographic and geophysical surveys together with an in situ geotechnical/geomechanic characterization were applied. The combined interpretation of the different techniques and of field observations allowed to define a geological and technical model of the landslide, both in surface than in depth, that clarified the triggering mechanism of the landslide and allowed to perform a back analysis on both strength and pore pressure parameters.

Debris flows involving ash-fall pyroclastic soils, which mantle slopes in the mountain ranges around the Somma-Vesuvius, represent a relevant societal risk. In the last years, many methods were applied to evaluate susceptibility and mobility of these landslides, considering at least two fundamental parameters: the slope angle and the thickness of the pyroclastic mantle. Despite this understanding, the assessment of soil thickness along slopes is still a challenging issue due to its high spatial variability and the steep morphological conditions that limit the use of traditional exploration methods. To overcome this problem, heuristic geomorphological methods were mostly applied up to now. In this paper, a regional-scale model of the ash-fall pyroclastic soils distribution along slopes is proposed, considering the isopach maps of each principal eruption. By means of field surveys, the model was validated and an inverse relationship between thickness and slope angle was found.

The Arteara rock avalanche is developed in the Fataga Group which is related to the first volcanic stage in the Gran Canaria Island (8.6–13.3 Ma). The materials of the avalanche are mainly phonolitic ignimbrites and lava flows. We have investigated the geotechnical quality of the five lithotypes distinguished in the area: (a) phonolitic lava, (b) phonolitic agglomerate breccia, (c) reddish non-welded ignimbrite, (d) welded fiamme-bearing ignimbrite, and (e) pumice tuff with lithics. The weak geotechnical properties of the reddish non-welded ignimbrite, the pumice tuff with lithics and the agglomerate breccia, and their location at the middle of the slope suggest that these lithotypes can be regarded as potential sliding surfaces of the avalanche. The geomechanical features of the volcanic rocks found in this study could be very useful for future studies concerning the modelling of the mechanism of failure and run out of the Arteara rock avalanche.

In this study, the effects of parameters in landslide simulation model (LS-RAPID) on the dynamic behaviour of earthquake-induced rapid landslides were examined to demonstrate the importance of each parameter. It was applied to a case study in Suruga Bay, Shizuoka Prefecture in Japan. The topographic data of the slope surface and sliding surface was generated from digital elevation model (DEM). Three different real seismic records were used as the inputs of earthquake i.e. 2008 Iwate-Miyagi Nairiku Inland Earthquake, 2009 Suruga Bay Earthquake, and the latest 2011 Tohoku Earthquake. Results show that landslides can be triggered by strong seismic loading using 2011 Tohoku real seismic records under a certain pore water pressure within LS-RAPID. The key parameters, including shear resistance at the steady state with physical meaning were found to have significant effect on the dynamic behaviour of these earthquake-induced rapid landslides.

In this paper we made a comparison between various methods to enter soil thickness as a spatial variable in a deterministic basin scale slope stability simulator. We used a slope stability model that couples a simplified solution of Richards infiltration equation and an infinite slope model with soil suction effect. Soil thickness was entered in the stability modelling using spatially variable maps obtained with four state-of-art methods: linear correlation with elevation; linear correlation with slope gradient; exponential correlation with slope gradient; a more complex geomorphologically indexed model (GIST model). Soil thickness maps and the derivate Factor of Safety (FS) maps were validated. Results confirmed that FS is very sensitive to soil thickness and showed that the same slope stability model can be highly sensitive or highly specific depending on the input soil thickness data. The uncertainty in the FS calculation can be reduced by applying more precise soil thickness input data: mean error of soil thickness maps is closely related to the sensitivity or specificity of the FS computation, while the overall performance of the stability simulation depends on mean absolute error and skewness of the frequency distribution of the errors of soil thickness maps. Despite the fact that slope-based methods are the most used in literature to derive soil thickness, in our application they returned poor results. Conversely, the use of the GIST model improved the performance of the stability model.

The influence of rainfall in triggering landslides is a widely discussed topic in scientific literature. The slope stability of fractured surface soils is often influenced by the soil suction. Rainfall, infiltrating into soil fractures, causes the decrease in soil suction and shear strength, which can trigger the collapse of surface soil horizons. Water flow through fractured soils can also be affected by soil swelling and by capillary barrier effects in the case of low permeable soil overlying a more permeable one.

These conditions are rarely investigated by the existing models, especially from the point of view of rainfall triggering surface landslides. For this purpose, we have developed a dual-porosity model that simulates water flow through fractured swelling soils overlying a more permeable soil. The model has been applied to a soil profile consisting of a thin layer of fractured loamy soil above a coarse sand layer, in order to investigate the influence of different rainfall intensities on the infiltration process, and on the distribution of the pore pressure that affects slope stability.

Aiming at better understanding the processes involved in perched water tables onset and in their development, the case of a soil slope characterised by gradually decreasing hydraulic conductivity at saturation with depth was numerically investigated. Different anisotropy factors and steepness values were accounted for. The problem was led to a dimensionless form on the basis of the Buckingham

π

-theorem. Coherently with a theoretical solution of the 2D sloping case, the simulations evidenced (a) non-monotonic transverse profiles of the pressure head within the perched water, (b) slightly lower infiltration thresholds for perched water onset and for soil waterlogging, with respect to the 1D case. If the slope is long enough, an almost uniform flux can be observed in a branch of its central part.

To model the evolution with time of the Bindo-Cortenova translational landslide (Italian Prealps, Lombardy, Italy), a 1D pseudo-dynamic visco-plastic approach, based on Perzyna’s theory, has been conceived. The viscous nucleus has been assumed to be bi-linear: in one case, irreversible deformations develop uniquely for positive yield function values; in a more general case, visco-plastic deformations develop even for negative values. The model has been calibrated and subsequently validated on a long temporal series of monitoring data and it seems quite reliable in simulating the in situ data.

Landslide susceptibility maps are useful tools for natural hazards assessments. The present research concentrates on an application of machine learning algorithms for the treatment and understanding of input/feature space for landslide data to identify sliding zones and to formulate suggestions for susceptibility mapping. The whole problem can be formulated as a supervised classification learning task. Support Vector Machines (SVM), a very attractive approach developing nonlinear and robust models in high dimensional data, is adopted for the analysis. Two real data case studies based on Swiss and Chinese data are considered. The differences of complexity and causalities in patterns of different regions are unveiled. The research shows promising results for some regions, denoted by good performances of classification.

Rock fall hazard assessment is a very demanding problem because of the complexity of the involved physics and its intrinsic stochasticity. New powerful tools, developed in the last few years, are characterized by different degrees of completeness, models, and algorithms. In the following, the modelling approach developed through the years within the code Hy_Stone is presented. Some issues related to the sensitivity to 3D topographic descriptions, the effects of stochasticity on model results and the calibration of 3D rockfall simulation are presented.

The toe of landslides is subject to compression stresses induced by the upward sliding masses and often characterized by a co mpression structure. These structures can evolve in passive toe thrust, which bias the geomorphologic evolution of the toe zone, particularly for prevailing longitudinal dimension landslide. This work presents a simple analytic model of the passive thrust at the toe of landslide based on the infinite slope approach to stability analysis. It is based on the analysis of the state of stresses according to Mohr circle representation and can be implemented also into a spreadsheet and making it possible to evaluate the form of failure surface at the toe of the landslide and the shear strength contribution to the factor of safety of a landslide.

The residual shear strength as operative strength along the shear surface is commonly used for analyzing the stability of reactivated landslides. On this base, it is not possible to have brittle failure or progressive failure mechanisms. In fact, these kinds of phenomena claim for a recovery of strength along the shear surface, which is due to ageing effects during the quiescence time. Several tests on clay samples were performed using Bromhead ring shear apparatus to verify the presence of shear strength recovery. Tests showed that recovery of strength are present along the shear surface. Moreover, they proved the stress-stain behavior is characterized by a brittle failure after that the strength, along the shear surface, falls again to the residual value.

It is nowadays evident the remarkable increase of natural events such as landslides and rock falls. This paper describes and shows how the ground reinforced soil embankments represent an efficient solution for the mitigation of the rockfall related damage. The advantages of these types of construction are that they can resist multiple impacts and their downslope deformation is negligible. This means they can safely be placed close to infrastructure and moreover, they are environmental friendly. In the last few years the Polytechnic of Turin and Officine Maccaferri S.p.A. has developed a new approach to design reinforced soil embankments used for rockfall applications. This approach is based on FEM modelling, full-scale tests results and real experiences utilized to validate the numerical analysis and modelling.

Roughness of discontinuities, when clean and unfilled, has a significant impact on rock joint mechanical behaviour and, as a consequence, of rock masses. The need to solve ambitious rock mechanics problems encouraged research to study shear mechanism acting at the joint level and to produce sound models to reproduce it. In this framework, roughness measurement allows to gather information about shear strength of rock joints at an acceptable cost.

A brief summary of published approaches and parameters for quantitative roughness evaluation is presented. Then, measures taken on rock samples extracted in a test site placed in the nearby of Esino Lario, 60 km north of Milan, are shown and discussed. The device used on-site to obtain digital replicas of profiles is a laser-camera profilometer, developed in the Department of Mechanical Engineering of the Politecnico di Milano. This device, combining a digital camera and a laser source, uses the principle of laser triangulation to extract a maximum of 746 points/profile. Numerical coordinates obtained by rock joint profiles are submitted to a Matlab® script which computes geometric parameters.

Statistical analysis of gathered data was used to define a minimum number of measures that have to be used to determine roughness of a joint set.

Deformation data obtained by differential interferometric analysis of radar images acquired by the satellite platform were used as a useful support in the definition of geometry and state of activity of landslides for identifying areas at risk and defining priorities for action. The interferometric data derived from radar images acquired by the ERS, RADARSAT and ENVISAT satellites were processed and used for the implementation of the database of landslides in the Arno basin (Tuscany, Italy). This database identifies some 25,500 landslide sites, of which about 10 % have risk elements, such as buildings, infrastructures and so. Of these, over 90 % present information about the movements that have permitted to perform statistical analysis of the velocity values recorded by satellites, reconstruct the displacement vector in the EWZN plane and obtain important information concerning the activity of landslides. The results were organized into a web-GIS project that represents a very important tool for river basin planning.

Because of the modern exploitation technologies for marble extraction, the open pit and underground quarries can quickly reach impressive sizes. Moreover the selection of exploited material or the geostructural conditions of the site can produce excavation with unusual and irregular shapes. Also for these reasons the appropriate design of the excavations, for quarrying activities, should require the knowledge of mechanical behaviour of the rock mass, of geo-structural conditions and of in situ stress. The main purpose of the quarry project should be to prevent instability of the rock mass, in every step of the foreseen exploitation. The lack of this specific technical evaluation contributed in the past to the development of large rock slide. The paper presents an applied approach for enhancing stability assessment and safety controls of marble quarries.

All around the world, people meet a challenge to find a balance between the risk of natural hazards and the need for spatial developments. Densely populated hillside regions in humid, subtropical or tropical climatic zones are often prone to various types of landslides. The complex flow behaviour of such gravitationally driven mass movements is reflected by inconsistent terminologies and ambiguous definitions of various landslide types in literature (Varnes 1978; Hutchinson 1988; Hungr et al. 2001). In this paper we focus on a discussion of on runout prediction methods of flow like mass movements, particularly on debris flows, where all transported material is generally in suspension and fluid and solid particles of all sizes typically travel with the same velocity. The term runout refers to the depositional part of a landslide or debris-flow event, providing information on the areas potentially covered by the transported solid material.

Open Source software is of great interest to many users and developers. One of the main advantages is that users can develop and adapt it to suit their purposes. This work shows modules developed into GRASS GIS for rockfall analysis. Modules examine both the potential failure detection (rockfall susceptibility) and the area of potential propagation. The study investigate three different mechanisms of failure: planar sliding, wedge sliding and toppling. The modules for rockfall susceptibility are called r.SMR, r.SSPC, r.fsplanar, r.wedgeSMR and r.wedgeSSPC according to the method of analysis adopted, while r.droka is the name of the module developed for the propagation of the landslide. Input data are both numbers and raster maps. GIS modules have been tested with good results in Ossola Valley and, in general, they should be applied in geological settings where the failure mechanism is mainly governed by discontinuity sets.

The GIS-based model SliDepot simulates the runout zones of landslide prone areas. It was developed by GEOTEST AG and applied during the last 10 `years for numerous projects. In combination with the SliDisp+ software (modelling of slope instabilities, cf. Tobler and Krummenacher (Modellierung von Anrissgebieten für flachgründige Rutschungen und Hangmuren. In: Proceedings of the 2nd Swiss geoscience meeting, Lausanne, 2004); Tobler et al. (Modeling potential shallow landslides over large areas with SliDisp+. In: Proceedings of the second World landslide forum, Rome, 2011) SliDepot allows to calculate decisive parameters for the dimensioning and optimized positioning of protection measures.

In contrast to other GIS-based models “Casadei et al. (Earth Surf Process Landf 28:925–950, 2003); Godt et al. (Eng Geol 102(3–4):214–226, 2008)”, SliDepot does not rely on a single-flow approach, which calculates the flow direction by direct neighbourhood relationship. The software is capable of analysing multiple cells in a 20°-sector above a potential runout area up to the extent of four cells. The potential runout cell will only be connected to the runout area if the mentioned 20°-sector contains an instable cell or if the necessary initial volumes of mobilised mass are guaranteed. Furthermore the program also considers geomorphologic phenomena like convex topography. With this approach the runout direction is simulated fairly realistic.

The runout is based on the degradational water content of the sliding mass during its downslope movement which finally leads to the break-off. Results from a case study in Switzerland will be presented.

Large debris avalanches are characterized by extremely rapid, flow-like motion of large masses and they travel extremely long distances showing much greater mobility than could be predicted using frictional models. In order to investigate the mechanisms involved and the reasons for the large propagation of these phenomena a discrete element model (DEM) and a combined finite and discrete element one (FEM/DEM) are used to simulate small-scale laboratory experiments carried out by Manzella “Manzella and Labiouse (Rock Mech Rock Eng 41(1):133–151, 2008, Eng Geol 109(1–2):146–158, 2009, Landslides, 2011 submitted); Manzella (Dry rock avalanche propagation: unconstrained flow experiments with granular materials and blocks at small scale. Ph.D. n 4032, Ecole Polytechnique Fédérale de Lausanne, Lausanne, CH, 2008)”. The combined use of different models produces a more complete study of the phenomena since each model can fill certain gaps of the other; they also help in a better understanding of some mechanisms and factors, which are important in the longitudinal propagation of granular and block flows, such as the progressive failure, the initial block packing and the topographical characteristics of the slope break.

The efficiency of a numerical model depends on both the realism of the assumptions it is based on, and on the way its parameters are assessed. We propose a numerical model based on the discrete element method which makes possible, thanks to the definition of an appropriate contact law, to simulate the mechanisms of energy dissipations by friction and shocks during the propagation of an avalanche of granular material on a slope. The parameters of the contact model are obtained from laboratory experiments of single impacts. A particular attention was paid to the values of the run-out, the morphology of the deposit, the proportions of energy dissipations by impacts or friction, and the kinetic energies of translation and rotation. The results of this numerical study provide valuable information on the relevance of some usual assumptions of granular flow continuous models.

The role played by the presence of erodible material along the path of flow-like landslides is analyzed. The effects of type and physical mechanical properties of materials, thickness and slope geometry on the runout and the deposit geometry are investigated. Fully 2D and 3D numerical simulations have been performed representing small scale laboratory experiments and large scale field examples. The properties adopted for the erodible material strongly control the evolution of the landslide and the type of occurring mechanisms. These aspects have a major influence on the results and on the hazard zonation and should be taken into account.

This paper deals with the theoretical-numerical and experimental analysis of dry rock avalanches moving down a chute.

Depth-averaged field equations of balance of mass and momentum as prescribed by Savage and Hutter (

1991

) are implemented in the RASH

3D

code. They describe the temporal evolution of the depth averaged velocity and the distribution of the avalanche depth. A Coulomb-type mechanical behavior of the mass is assumed. To incorporate the curvature effects of the bed, the centripetal acceleration term has been here implemented in the code.

Carried out experiments consist in the release of granular material on an inclined plane that is connected to a horizontal run-out zone through a sharp transition. Comparison of the experimental findings with the computational results proved that neglecting the centripetal acceleration term can have the effect of leading to errors in the determination of the well fitted friction angle. In particular, an overestimation of the computed dynamic friction angle respect to its measured value is observed.

A comparison between the performances of two different debris flow models has been carried out. In particular, a mono-phase model (FLO-2D) and a two phase model (TRENT2D) have been considered. In order to highlight the differences between the two codes, the alluvial event of October 1, 2009 in Sicily in the Giampilieri village has been simulated. The predicted time variation of several quantities (as the flow depth and the velocity) has been then analyzed in order to investigate the advantages and disadvantages of the two models in reproducing the global dynamic of the event. Both models seem capable of reproducing the depositional pattern on the alluvial fan in a fairly way. Nevertheless, for the FLO-2D model the tuning of the parameters must be done empirically, with no evidence of the physics of the phenomena. On the other hand, for the TRENT2D, which is based on more sophisticated theories, the parameters are physically based and can be estimated from laboratory experiments.

Among all kinds of landslides, rock avalanches are characterized by exceptional mobility and destructiveness. Their mobility is well larger than explained by the real material properties and it is usually expressed by means of an “apparent” friction angle which is a-priori unpredictable. We replicate the motion of historical rock/debris avalanches evolved in glacial and volcanic environments. The modelled events involved variable volumes (ranging from millions of m

3

to km

3

) which are well preserved so that their main features are recognizable from satellite images. Within each class of events, and irrespective of the variety of conditions in which they occurred, the best fitting parameters span in a narrow interval. The bulk basal friction angle ranges within 3° and 7.5° for volcanic debris avalanches, within 6° and 12° for ice-rock avalanches. These values are significantly lower than other rock avalanches which require values as high as 11° to 31°. The consistency of the back-analyzed parameters is encouraging for a possible use of the model in the perspective of hazard mapping while set of calibrated values can help the selection of model input parameter values for prediction and for definition of uncertainty on zonation.

Rock avalanches are studied at the Laboratory for Rock Mechanics of EPF Lausanne by means of unconstrained flow experiments of dry granular material. Tests aim at studying the influence of different parameters on the propagation and deposition mechanisms as well as on the characteristics of the final deposit. The existing measurement system, based on the fringe projection method, was improved in order to increase the quality and the reliability of the volume measurement and the computation of the centre of mass of the final deposit. The latest developments allow using this method to follow the mass during its motion. Additionally, a new technique that allows tracking the mass front was set up and compared to previous results gathered with the commercial software WINAnalyze. Results of experiments carried out with a gravel supposed to have the same characteristics as the one used in a previous research study Manzella (2008 Dry rock avalanche propagation: unconstrained flow experiments with granular materials and blocks at small scale. Ph.D. thesis 4032, Ecole Polytechnique Fédérale de Lausanne, Switzerland) showed some differences, suggesting that the grading of the material has a certain influence on the runout and the position of the centre of mass of the final deposit. A more significant difference is related to the state of the propagation surface. Recent experiments with a curved transition at the bottom of the slope and a larger volume lead to new results. Contrary to previous observations, for a given falling height of the source base, the distance travelled by the centre of mass on the horizontal panel increases with the volume. This leads to discuss further some conclusions previously drawn.

The aim of this study is the development of an integrated approach for definition of scenarios of debris flow deposition. The approach has been applied to basins located in the North-western Alps. Firstly the basins are classified on the basis of the bedrock lithology. Hence the features of debris flows are related to different bedrock lithologies: basin area/fan area ratios, alluvial fan architectures, depositional styles and triggering rainfall characteristics. This method permits to capture the essential features of flow phenomena and to address the choice of proper constitutive laws for the numerical modelling of debris flow propagation and deposition. The numerical model calibration is based on the observed styles of deposition of debris flows and architecture of the alluvial fans (shape, slope, grain-size distribution, etc.). The simulations do not match the deposit of a given past event, but the results are in terms of scenarios and supply a more realistic tool for risk mitigation.

This paper focuses on the quantitative investigations carried out to evaluate stability conditions of the northern flank of the

La Fossa

cone (Vulcano Island) and the consequent runout of the debris avalanche that would be generated by a possible failure. The contribution describes first the geological setting and the typical instability phenomena of the area. Then it illustrates how the global morphology of the flank was reconstructed by combining aerial photogrammetry and high resolution 3D surveying techniques such as terrestrial laser scanner. The resulting digital elevation model (DEM) permitted us to extract and measure structural and morphological elements that drive and influence past and potential instabilities. These elements were used to constrain limit equilibrium analyses (LE) that were used to estimate the volume of soil mass susceptible of failure. Strength parameters at large scale were obtained from laboratory tests, described in previous studies, and back analyses of a past failure whose geometry was reconstructed from the DEM. Finally, results of runout analyses of the debris avalanche are presented and discussed. Analyses were conducted by using DAN-W and DAN-3D codes which allowed an estimate of the maximum runout distance and extension of the accumulation zone with respect to the inhabited area.

The main goals of landslide run-out modeling should be the assessment of future landslide activity with a range of potential scenarios, and the information of the local populations about the hazards in order to enable informed response measures. In recent times, numerical dynamic run-out models have been developed which can assess the velocity and extent of motion of rapid landslides such as debris flows and avalanches, flow slides and rock avalanches. These models are physically-based and solved numerically, simulating the movement of the flow using constitutive laws of fluid mechanics in one or two dimensions. Resistance parameters and release volumes are crucial for a realistic simulation of the landslide behavior, whereas it is generally difficult to measure them directly in the field. Uncertainties in the parameterization of these models yield many uncertainties concerning their frequency values, which must be addressed in a proper risk assessment. Based on the probability density functions of release volumes and friction coefficients of a given landslide model, this work aims to systematically quantify the uncertainties in the run-out modeling. The obtained distributions can be used as an input for a probabilistic methodology where the uncertainties in the release volume and friction coefficients (rheological parameters) inside the dynamic models can be addressed. This will improve the confidence of the dynamic run-out model outputs such as the distribution of deposits in the run-out area, velocities and impact pressures, important components for a risk analysis and regulatory zoning.

This paper deals with a laboratory and numerical research on mudflows performed in with the aim to study the interrelations between the grain-size composition of the mud and its rheological properties at different solid concentrations. Furthermore, the predicting capability of a new numerical model in reproducing the flow of viscous materials is evaluated.

The Monte Rosa east face (Macugnaga, Italian Alps) is one of the highest flanks in the Alps. Steep hanging glaciers and permafrost cover large parts of the wall. Since the end of the Little Ice Age (about 1850) the Monte Rosa east face is undergoing a progressive reduction of its ice cover; moreover new instability phenomena related to permafrost degradation and rapid deglaciation have been occurring since over a decade ago. The progressive destabilization of high-mountain faces is a consequence of many factors, such as topography, geological and structural conditions, intense freeze-thaw activity and oversteepened slopes from glacial erosion.

Two major events, an ice avalanche occurred in August 2005 and a rock avalanche occurred in April 2007 are briefly described in this paper. In both cases, the accumulation area was located on the Belvedere Glacier at the foot of the Monte Rosa east face.

A 3D dynamic model (DAN3D) was applied in order to back analyze the runout of the events, enabling the calibration of the input parameters for the assumed rheological models.

In the last 3 years Brazil was hit by the most important group of accidents related to landslides in the history of the country. The biggest accidents affected the States of Santa Catarina, Rio de Janeiro, Pernambuco, Alagoas and Paraná. We estimated more than 20,000 landslides in these most important disasters and around 2,000 victims dead or disappeared. The effects of these accidents are enormous and in many cases the community will be paying the costs for long time. The group of CENACID-UFPR, Federal University of Paraná, responded to almost all of these accidents and developed a new approach to face the challenge of fast evaluation of risk in these events. This new approach is proposed as a tool to rapid and systematic evaluation of groups of landslides, which we call “landslides systems”. This methodology is called “Relative Rapid Landslides Analysis” and is based in five different indicators of risk.

Within the frame of the ongoing project MoNOE (Development of Methods for Modelling Natural Hazards in the Province of Lower Austria) the generation of an indicative multi-hazard map to be used as a scientific reference for regional policies of land use management is one of the main objectives. The applicability and reliability of two relatively simple GIS-based approaches have been analysed aiming at the identification of areas which are potentially endangered by rockfall in a fast and cost-effective manner.

In view of the acute need to solve a problem related to a landslide hazard at local and regional levels, the system-based tool, which integrates a mathematical modeling and GIS technologies, has been developed. A mathematical model of landslide process was developed based on the theoretical grounds of statics of granular medium and the most frequently used calculation methods of slope stability coefficient by Maslov-Berer and Shakhunyants and value of landslide pressure by Shakhunyants. Presented model underlies design of software for modeling slopes stress condition given engineering-geological changes (LANDSLIP07). This software is used for analysis of landslide processes. These research activities have been carried out in the framework of the Institute of Telecommunications and Global Information Space of National Academy of Sciences of Ukraine, and the IPL-153 Project “Landslide hazard zonation in Kharkov region of Ukraine using GIS”.

Tehran is the country’s most densely populated district which is located on the southern part of Alborz Mountains. Many sensitive infrastructures such as dams, roads, power lines and housing complexes are located within the area. In this paper the potential of landslide hazard for North Tehran is evaluated and a hazard zonation map based on ARC-GIS method is prepared. Several parameters such as lithology, slope angle, slope direction, distance from faults and seismicity (earthquake deduced acceleration) are considered as main factors. The quality parameters of the defined effecting factors are quantified by the above named software where the data layers are divided into smaller classes. In the next stage, specific weights are assigned to each class and the hazard potential values are determined. Finally they are presented as a landslide hazard zonation map of the area. It was found that slope angle of 16–35°, slope direction from 40° to 140°, lithology of tuffacious rocks, distance of fewer 2 km from fault and seismicity of over 0.5 g has great effect on the distribution of landslide in the area.

Landslides starting from unstable slopes threat people, buildings and infrastructures all over the world and are therefore intensively studied. On the one hand, engineers use sophisticated models to identify hazardous slopes, mostly based on longitudinal sections. On the other hand, less sophisticated infinite slope stability models are used in combination with Geographic Information Systems (GIS) in order to cover larger areas. The present paper describes an attempt to combine these two philosophies and to come up with a spatially distributed, three-dimensional model for slope stability going beyond the widely used infinite slope stability concept. Both models are applied to artificial topographies in order to compare the outcomes of different slip surface assumptions and to benchmark the validity of the infinite slope stability model. It was found out that the resulting factor of safety is highly sensitive to the type of model used and to the slip surface geometry. In complex terrain, the performance of the infinite slope stability model strongly depends on the specific situation, particularly on slope curvature and slip surface depth.

Computer models play an increasing role for the understanding of the dynamics of granular flows (rock avalanches, debris flows, snow avalanches etc.). Simple empirical relationships or semi-deterministic models are often applied in GIS-based modelling environments. However, they are only appropriate for rough overviews at the regional scale. In detail, granular flows are highly complex processes and deterministic models are required for a detailed understanding of such phenomena. One of the most advanced theories for understanding and modelling granular flows is the Savage-Hutter model, a system of differential equations based on the conservation of mass and momentum. The equations have been solved for a number of idealized topographies, but not yet satisfactorily for arbitrary terrain. Not many attempts to integrate the model with GIS were known up to now. The work presented is seen as an initiative to integrate a fully deterministic model for the motion of granular flows, based on the Savage-Hutter theory, with GRASS, an Open Source GIS software package. The potentials of the model are highlighted with the Val Pola rock avalanche as test event. The limitations and the most urging needs for further research are discussed.

Landslides are especially damaging in Canada and despite their extensive occurrence, the exact location of instability is not homogeneous across the country and depends on several factors. Based on clear evidence it is known that there is a wide range in the scale and diversity of landslide environments, especially as they pertain to landslide problems.

A proposed 1:6 million scale landslide susceptibility map of Canada is presented in this study as a first approximation for those individuals interested in pursuing more detailed investigations. The final map has been constructed based primarily on GIS, by considering digital layers of relevant national information including: slope angle, aspect, precipitation, permafrost, surficial geology, vegetation, distance to rivers, distance to coast (lakes and oceans) and bedrock lithology. These variables have been categorized into several classes depending on their greater or lesser favourability to influence slope instabilities. The values assigned to each class are not equal for the whole country and vary depending on the region considered (Canadian Shield, Hudson Bay Lowlands, or the remaining area).

The present work aims to assess the landslide susceptibility of a hilly area, using two methods: the weight of evidence statistical method (WOE) that is based on bivariate statistical analysis and the deterministic approach based on the infinite slope model.

To document the efficiency of this quantitative methods, a landslide prone area along Prahova Subcarpathian Valley was chosen, where landslide hazard interacts with human settlement and activities.

The Landslide susceptibility map derived from the WOE method has been compared with that produced from the deterministic method and the results analyzed. The values obtained were also in good agreement with the field observations.

Results show that the main controlling factors in the evolution of landslides are the slope gradient, saturation condition and lithology. Beside these, factors like morphography (slope aspect and flow direction) or human intervention (land use, roads, buildings) can interfere.

Two major movement modes of slope failures: landslides and debris-flows are simulated by two integrated models using GIS to represent the dynamic process across 3D terrains. Landslides occur when earth material moves downhill after failing along a shear zone. Debris flows can be differentiated from landslides by the pervasive, fluid-like deformation of the mobilized material. The formation of debris flows most often occurs as a result of a landslide partially or completely mobilizing into a debris flow. GIS-based revised Hovland’s 3D limit equilibrium model is used to simulate the movement and stoppage of a landslide. The 3D factor of safety will be calculated step by step during the sliding process simulation. Stoppage is defined by the factor of safety much greater than one and the velocity equal zero. GIS-based depth-averaged 2D numerical model is used to predict the runout distance and inundated area of a debris flow. The simulation displays the propagation and deposition and deposition of a debris flow across the complex topography. The two GIS-based integrated models are able to provide effective tools for hazard mapping and risk assessment.

Since 1980s the surveys space could not definitely follow the growth and the density of favelas, and some communities started to trigger the first records of accidents in areas hitherto stable. The mapping made in 2010 is today the largest geological-risk database in the country. Today, all these information are included in the “Habisp”. Habisp is a mapping system of precarious settlements in the city of Sao Paulo, which contains valuable information to face the urban poverty. Poverty which is materialized in Sao Paulo in many ways of informal settlements: slums, irregular settlements, tenement housing, temporary housing and degraded sets. The Habisp stores, organizes, processes and produces high quality geographic information, which serves as support for the technicians of the Housing department in making decisions. The results have been making possible reassessment and adjustment of the low-income intervention projects by the government, prioritizing housing, social-educational infrastructure, and basic-sanitation actions in areas of greatest susceptibility.

This work describes an organizing of landslides data and proposes to develop research in information technology with the aim to improve the analysis and representation of the landslides areas.

Geological Information System (GIS) techniques were used in a ‘test area’, by a play of transparency on the layer of orthophotos and of hillshade, to underline physiographic aspects of the area and to map the landslides. Moreover, it is shown that the use of high-resolution DTM in the mapping of landslide hazard, improves the visual interpretation to give a useful contribution to the analysis and photo interpretation.

Landslides and floods are widespread and recurrent in Italy, where they cause damage and pose a threat to the population. To estimate geo-hydrological risk in Italy, catalogues of landslide and flood events that have caused damage to the population were compiled from a variety of sources. The catalogues covers the 1,943-year period from 68 A.D. to 2010, and list 3,310 landslide events and 2,624 flood events that have resulted in deaths, missing persons, injured people, and homeless. For each event in the catalogue, different types of information were collected and organized in a database. We describe the Spatial Data Infrastructure (SDI) we have implemented to collect, store, analyze, and disseminate the historical information, and results of the analysis of landslide and flood risk to the population.

The geomorphological cartography explained in this work is the result of a process of synthesis resulting from detailed geological and geomorphological and hydrogeological researches and from numerous land surveys carried out during the past ten years. The choice of the study area can be explained by the presence of important and complex phenomena involving the upper course of the Velino river, such as palaeo-landslides, mass-movements and debris flows, subsidence phenomena, etc. In particular, these phenomena involve several towns (Posta, Micigliano, Sigillo, Villa Camponeschi), as well as a fundamental route (State Road 4), following the ancient consular road, named Salaria, roman in age.

The informations deriving from the land surveys has been integrated and controlled by photo-interpretation (related at 1982 and 2000 aero-photographs). The result has been implemented by the 2006 cartography, which was obtained from the ortho-rectified images, subject to a “spreading” over the digital model of the terrain that was obtained from the regional topographical cartography at a scale of 1:10,000, promoted by the Lazio Region. Even if they have well-known limits of representation that derive from the approximate restoration of the basic regional maps that were realised in the 1980s, the latter represent a very up-to-date control instrument, depending on a critical analysis of the information.

All the data have been computerised by exploiting the tools made available by the ESRI platform, through the development of an innovative logical pattern for the geomorphology, since the aforesaid data were treated by considering forms, processes and deposits on the basis of the prevailing morphogenetic agent. In this way, the problems deriving from the contiguity and superimposition of several polygonal forms in logical contrast and topological conflict have been overcome.

Hopefully, several representational limits will be overcome once the set-up valid for the published printing has been made ready, since the instruments currently available have several technical limitations in the raster process and, in spite of the information present is correct, it does not yet respect several qualitative standards typical of Italian graphic representation.

Several graphic improvements typical of the Italian publishing language and cartographic tradition will be the topic of the subsequent phase.

Despite Italian tradition imposes extreme attention on the graphic aspect, it is very important to note that the territorial computer-related systems facilitate an analysis of the coverings, depending on the substrate on which they are imposed, and make it possible to define a model according to which the forms, processes and deposits evolve over time, infact the whole upper valley of the Velino river is annually subjected to deep changes due to the high energy of relief and extreme weather conditions: knowledge and proper analysis of these phenomena can contribute significantly to the right land management, giving due importance to the prevention and the screening to be daily carried out.

The Geological Survey of Norway is currently developing a database for unstable rock slopes in Norway. The database is intended to serve three main purposes: (1) to serve as a national archive for potential unstable slopes for use in hazard and risk analysis, aerial planning and mitigation work as well as research; (2) to serve as a robust and easy to use database during field mapping of unstable rock slopes; (3) provide a public available database accessible through an online web map service. The database structure contains several feature classes storing both raw and processed data, including structures, lineaments, displacement measurements, run-out areas and other observations. All feature classes are linked to one main point which holds a general summary for each unstable slope.

This paper presents the first relevant results of the project BRISEIDE – BRIdging SErvices, Information and Data for Europe. BRISEIDE aims at providing operators with a time-aware extension of data models and value added services for spatio-temporal data management, authoring, processing, analysis and interactive visualization in several emergency-related scenarios including, most notably, landslides. Within this context a number of WPSs devoted to spatial analysis have been developed and integrated within existing open source frameworks. Spatio-temporal processing services are exposed via the web and are made available through compatible WebGIS applications. Through BRISEIDE, operators can access processing services through an interactive web-based 3D GeoBrowser capable to allow management, authoring, interaction, filtering of existing data. The 3D GeoBrowser allows interactive orchestration of spatio-temporal WPSs providing support to chaining of required processing units. This ensures interactive access to datasets and asynchronous processing at the server side. The project mobilizes a value-chain of stakeholders to validate the pilots from a technical, organizational and legal standpoint.

Historic, post-eruptive debris flows of remobilised volcanic ash are rare in Iceland, being restricted to explosive eruptions. Volcanic ash slurry from the southern slopes of the ice-capped Eyjafjallajökull volcano on 19 May 2010 is the first lahar observed in Iceland since the 1947 Hekla eruption. This study focuses on the volume of sediment transported, the size and hydrological behavior of watersheds, and the resulting erosion. The analysis is based on: (1) direct measurements of the 19 May lahar; (2) direct measurements of ash fallout; (3) aerial and ground-based imagery; (4) topographic data from an airborne LIDAR survey; (5) airborne synthetic-aperture radar; and (6) precipitation data. The volume of the lahar in the Svaðbælisá channel was estimated at 200,000 m

3

. This flow originated from crown and flank failures, similar to slab avalanches, with water-saturated, fine-grained ash as the slip surface. Several ash-laden floods occurred in Svaðbælisá and neighboring channels during the summer of 2010. None, however, were as saturated as the 19 May lahar. An increased number of small debris flows were also recorded some blocking roads to farms. Precipitation during the summer of 2010 was not higher than average and therefore does not explain this increased erosion. Large quantities of volcanic ash mantle the lower slopes of the ice-cap. Ash in the ablation zone is expected to be transferred down-slope in the next few years inducing the erosion to the root of the mountain endangering homes and infrastructure. Fieldwork during the summer of 2010 has resulted in a map showing the volume of ash above and below the ablation zone of the main catchments and recorded erosion events. This data was used to assess the hazard and the need for immediate actions.