Vous êtes sur la page 1sur 86

Alpine Mass Movements:

Implications for hazard assessment and mapping

Seite 1

Seite 2

Inhalt
Florian Rudolf-Miklau, Richard Bk, Franz Schmid, Christoph Skolaut: Hazard Mapping for Mass Movements: Strategic Importance and Transnational Development of Standards in the ASP-Project ADAPTALP

Seite 3

Imprint / Disclosure
Federal Ministry of Agriculture, Forestry, Environment and Water Management, Marxergasse 2, 1030 Vienna, Austria. Verein der Diplomingenieure der Wildbach- und Lawinenverbauung, Bergheimerstrasse 57, 5021 Salzburg, Austria Editorial Team: Florian Rudolf-Miklau, Richard Bk, Christoph Skolaut and Franz Schmid Coordination: Barbara Kogelnig-Mayer Layout: Studio Kopfsache, Mondsee Cite as: BMLFUW (2011): Alpine Mass Movements: Implications for hazard assessment and mapping, Special Edition of Journal of Torrent, Avalanche, Landslide and Rock Fall Engineering No. 166. This publication was implemented within the framework of EU-project AdaptAlp, Workpackage 5, and is co-nanced by the European Regional Development Fund (ERDF)

Florian Rudolf-Miklau: Principles of Hazard Assessment and Mapping

Richard Bk, Hugo Raetzo, Karl Mayer, Andreas von Poschinger, Gerlinde Posch-Trzmller: Mapping of Geological Hazards: Methods, Standards and Procedures (State of Development) - Overview Mateja Jemec & Marko Komac: An Overview of Approaches for Hazard Assessment of Slope Mass Movements

BLOCK 1: Key-note papers

Roland Norer: Legal Framework for Assessment and Mapping of Geological Hazards on the International, European and National Levels

Karl Mayer, Bernhard Lochner: Wolfram Bitterlich: Internationally Harmonized Terminology for Wildbachverbauung und kologie Widerspruch oder sinnvolle Ergnzung? Geological Risk: Glossary (Overview)

Michael Mlk, Thomas Sausgruber, Richard Bk, Arben Kociu: Standards and Methods of Hazard Assessment for Rapid Mass Movements (Rock Fall and Landslide) in Austria

Hugo Raetzo, Bernard Loup: Geological Hazard Assessment in Switzerland

Cover picture: Grohangbewegung Rindberg, Gde. Sibratsgfll, Vorarlberg Source: die.wildbach

BLOCK 2

Stefano Campus: Landslide Mapping in Piemonte (Italy): Danger, Hazard & Risk

Seite 102

Seite 94

Seite 82

Seite 70

Seite 64

Seite 48

Seite 24

Seite 14

Seite Seite12 6

Seite 4

Inhalt
Marko Komac, Mateja Jemec: Standards and Methods of Hazard Assessment for Rapid Mass Movements in Slovenia

Seite 5

BLOCK 2: Hazard assessment and mapping of mass-movements in the EU

Karl Mayer, Andreas von Poschinger: Standards and Methods of Hazard Assessment for Geological Dangers (Mass Movements) in Bavaria

Didier Richard: Standards and Methods of Hazard Assessment for Rapid Mass Movements in France

Pere Oller, Marta Gonzlez, Jordi Pinyol, Jordi Marturi, Pere Martnez: Goeohazards Mapping in Catalonia

Claire Foster, Matthew Harrison & Helen J. Reeves: Standards and Methods of Hazard Assessment for Mass Movements in Great Britain

Karl Mayer, Bernhard Lochner: International Comparison: Summary of the Expert Hearing in Bolzano on 17 March 2010

Seite 158

Seite 150

Seite 142

Seite 130

Seite 118

Seite 108

Seite 6

Zusammenfassung: Massenbewegungen (Steinschlag, Rutschungen, Felsgleitungen) bedrohen den alpinen Lebensraum und verursachen zahlreiche Risiken. Durch die intensive Raumnutzung in den Bergtlern besteht ein zunehmender Bedarf an genauen Gefahrenkarten fr diese Gefahrenarten. Aufgrund fehlender Daten und zuverlssiger Methoden fr die Gefahrenbeurteilung wurden bisher keine generellen Standards fr die Gefahrdarstellung von Rutschungen und Steinschlgen entwickelt. Die Unsicherheit in der Beurteilung der Gefahren wird durch den Einuss des Klimawandels noch erhht. Das Projekt ADAPTALP zielt darauf ab, diese Lcke durch die Entwicklung transnationaler Standards fr die Gefahrenzonenplanung fr Massenbewegungen zu schlieen.
snow avalanches. However there are no legal Alpine Space at risk: Importance of hazard maps FLORIAN RUDOLF-MIKLAU, RICHARD BK, FRANZ SCHMID, CHRISTOPH SKOLAUT In the Alpine countries, natural hazards constitute a security risk in many regions. Floods, debris ows, avalanches, landslides and rock falls threaten people, their living environments, their settlements and economic areas, transport routes, supply lines, and other infrastructure. They constitute a major threat to the bases of existence of the population. The increasing settlement pressure and area consumption, the opening up of transport routes in the Alps as well as strong growth rates in tourism have brought about a considerable spatial extension of endangered areas. With the rising demands on welfare and quality of life, the need for safety and protection (technical) standards available for the outline of areas endangered by mass movements (e.g. landslides, rock fall). The assessment of these processes concerning the frequency and intensity of events (disasters) is difcult and demanding due to the lack of measurements and basic data. In addition, the knowledge of geotechnical parameters, physical properties and triggering mechanisms of the displacement processes still are fragmentary, although wide progress were achieved by improved monitoring methods and the detailed analysis of past events. Recently the expansion of settlement areas in Alpine valleys and the growing vulnerability of human facilities have signicantly increased the risk for natural disasters caused by mass movements. The growing demand for hazard maps that cover these risky processes has initiated strong efforts in all mountainous countries in Europe to develop exact methods and appropriate standards that enable the production of hazard maps for mass movements with sufcient accuracy. By bundling these initiatives the ASP (Alpine Space Program/Funding Initiative of the European Commission) project ADAPTALP in cooperation with other projects like SAFELAND, PERMANET or MASSMOVE aims at the development of technical standards and provision of harmonized quality criteria for all member states.

Hazard Mapping for Mass Movements: Strategic Importance and Transnational Development of Standards in the ASP-Project ADAPTALP Gefahrendarstellung von Massenbewegungen: Strategische Bedeutung und lnderbergreifende Entwicklung von Standards im Projekt ADALPTALP
Summary: Mass movements (rock falls, landslides, rock slides) are major threats for the Alpine living space and cause various risks. Due to the intensive land use in the mountain valleys, there is an urgent need for reliable hazard maps for these types of hazards. Missing data and the lack of reliable methods for the assessment of hazards has obstructed the development of general standards in hazard mapping for landslides and rock fall. The uncertainties and inaccuracies of models are increased by the impact of climate change. The project ADAPTALP (within the Alpine Space Program) aims to close this gap by creating transnational standards for hazard mapping concerning geological risks (mass movements).

of the population increased as well. Hazard maps that show areas at risk by natural hazards are of paramount importance for the development of Alpine regions. The maps count among the active planning measures in natural hazard management and serve to the safety of existing settlements and their inhabitants as well as to the steering of land-use only outside of endangered areas. Since the beginning of 1970s, these maps have been established in several countries (Switzerland, Austria, France) for the hazards ood, debris ow and

Seite 7

Seite 8

Mass movements: Hazard processes on slopes A variety of processes exist by which materials can be moved through the slope system. These processes are generically known as mass movement or mass wasting. Mass movements per denition are movements of bodies of soil, sediments such as residual soil and bed rock which usually occur along steep-sided slopes and mountains. Mass movements can be classied due to the rate of movement (rapid or slow), the type of movement (falling, sliding or owing) and to the type of material involved (soil, sediments or rock debris).

damming up bodies of water. Expenses related to landslides include actual damages to structures or property, as well as loss of tax revenues on devalued properties, reduced real estate values in landslide prone areas, loss of productivity of agricultural lands affected by landslides, and loss of industrial productivity because of interruption of transportation systems by landslides. Not only rapid types of mass movements are harmful. Slow movement of creep does more long term economic damage to roads, railroads, building structure and underground pipes. The operation of mass movement processes relies upon the development of instability in the slope system. The predominant source of stress is the gravitational force. Other factors that affect mass movements are the steepness of slopes, the lithological property of the slope materials, and the amount of water in the material. The two most important parameters in mass movement is the angle of friction and the cohesion. The magnitude of the gravitational force is related to the angle of the slope and the weight of slope sediments and rock. The following equation models this relationship: F = W sin

of the particles. Slopes consisting of silt and clay particles obtain it from particle cohesion, which is controlled by the availability of moisture in the soil. Rock slopes generally have the greatest internal strength due to the crystalline structures. Instability is not always caused by an increase in stress. In some cases, the internal strength of the materials can be reduced resulting in the triggering of a mass movement. Failure of the slope material can occur over a range of time scales. Some types of mass movement involve rather rapid, spontaneous events. Sudden failures tend to occur when the stresses exerted on the slope materials greatly exceed their strength for short periods of time. Mass movement can also be a less continuous process that occurs over long periods of time. Slow failures often occur when the applied stresses only just exceed the internal strength of the slope system. Many factors can act as triggers for slope failure. One of the most common is prolonged or heavy rainfall. Rainfall can lead to mass movement through three different mechanisms. Often these mechanisms do not act alone. The saturation of soil materials with water increases the weight of slope materials which then leads to greater gravitational force. Saturation of soil materials can also reduce the cohesive bonds between individual soil particles resulting in the reduction of the internal strength of the slope. Lastly, the presence of bedding planes in the slope material can cause material above a particular plane below ground level to slide along a surface lubricated by percolating moisture. Additionally, a large variety of other trigger mechanism for mass movement other than the gravitational are known, such as: Earthquake down. Human modication of the land or shocks cause sections of mountains and hills to break off and slide

weathering and erosion help loosen large chunks of earth and start them sliding downhill. Vibrations from machinery, trafc, weight loading from accumulation of snow, stockpiling of rock, from waste piles and from buildings and other structures. In the Alps, mass movements occur in a wide range of processes consisting of bedrock and soil or a mixture of both. Mass movement on hard rock slopes is often dramatic and quick. They involve the downward movement of small rock fragments pried loose by gravitational stress, the enlargement of joints during weathering and/or freeze-thaw processes (rock fall). Larger scale, down slope movement of rock can also occur along welldened joints or bedding planes. This type of movement is called rock slide. Rock slides often occur when a fracture plane develops causing overlying materials to slide down slope. Slopes formed from clays and silt display somewhat unique mass sediments

movement processes. Two common types of mass movements in these cohesive materials are rotational slips (slumps) and mudows. Both of these processes occur over very short time periods. Rotational slips or slumps occur along clearly dened planes of weakness which generally have a concave form beneath the earth's surface. These processes can be caused by a variety of factors. The most common mechanical reason for them to occur is erosion at the base of the slope which reduces the support for overlying sediments. Mudows occur when slope materials become so saturated that the cohesive bonds between particles is lost. In a mudow there is enough water to allow the mixture to ow easily, as a viscous stream. Mudows can occur on very low slope angles because internal particle frictional resistance and cohesion is negligible.

Fig. 1: Land slide in cohesive soil resulting from slope instabilities and saturation of material by water. Abb. 1: Rutschung in bindigem Boden resultierend aus Hanginstabilitten und Wassersttigung des Bodens.

and

where F is gravitational force, W is the weight of the material occurring at some point on the slope, and is the angle of the slope. The stability of a slope depends on the

Mass

movements

have

direct

indirect impact on a number of human activities. The steepness and structural stability of slopes determines their suitability for agriculture, forestry, and human settlement. Instable slopes can also become a hazard to humans if their materials move rapidly through the process of mass wasting. Landslides can suddenly rush down a steep slope causing great destruction across a wide area of habitable land and sometimes also oods by

relationship between the stresses applied to the materials that make up the slope and their internal strength. Mass movement occurs when the stresses exceed the internal strength. Slopes composed of loose materials, such as sand and gravel, derive their internal strength from frictional resistance, which depends on the size, shape, and arrangement

Seite 9

Seite 10

An earth ow is slower moving than a mudow and involves a mass of material that retains rather distinct boundaries as it moves. Debris ow is a term used generally for rapid mass movements consisting of water and residual soil. The term implies a heterogeneous mixture of materials including a considerable fraction of particles that are coarser than the particles in mud. Debris ows occur on slopes as well as in laterally conned channels. Engineering soil predominantly coarse Fall Topple Slide Spread Flow Rock fall Rock avalanche Rock topple Rock slide Rock spread (Rock ow) (Debris fall) (Debris topple) Debris slide (Debris spread) Debris ow (in channels) ne (Earth fall) (Earth topple) Earth slide (Earth spread) Earth ow

to the increasing temperatures. The uncertainties and the increase of natural hazards due to the impacts of climate change require concerted management in the Alpine Space. It must be managed on a transnational, national, regional and local scale to effectively save human life, settlements and infrastructure. Nevertheless, there is still a lack of precise data taking climate change into account. The result is an insufcient accuracy of available models and inaccurate prediction of natural hazard and menacing catastrophic events. The impact of climate change increases these uncertainties. Harmonized cross-sectoral hazard assessment and hazard mapping must be balanced on a transnational level. The ADAPTALP project (www.adaptalp.org) focuses on the harmonization of the various national approaches and methods for the assessment of hazards related to mass movements. Along with the harmonization Hazard zones are designated areas threatened by natural risks such as avalanches, landslides or ooding. The formulation of these hazard zones is an important aspect of spatial planning. The basis for hazard maps is a comprehensive assessment of geological and hydro(geo)logical framework conditions, slope instabilities, relevant triggering mechanisms, properties of displacement processes, potential risks and the vulnerability of endangered areas (objects). Consequently it is essential to distinguish the three aspects of mass movement assessment and mapping: Dangers (susceptibilities): Assessment and characterization of threat (typology, morphology, inventory of mass movements). Hazards: Spatial and temporal probability, intensity and forecasting of evolution (scenarios) are needed. Risks: Interaction between a threat having particular hazard and human activities. In principle, these theoretical concepts are well of terminology, an important issue tackled by ADAPTALP is the provision of reliable data and models for this kind of processes. The more reliable the information basis, the more efciently adaptation strategies on local and regional level can be implemented. The project is based on an integrated transnational approach. That means that a comprehensive comparison of all available standards and methods is carried out covering all countries in the Alpine region (Austria, Germany, Italy, France, Switzerland, Slovenia) and other European states with a considerable share of mountain regions (Great Britain, Spain, Norway). The transnational exchange of knowledge and the international harmonization in method and procedure will raise the quality of hazard assessment considerably. A general state-of-the-art for hazard mapping concerning mass movements seems to be within reach. Hazard maps for mass movements
Fig. 2: Transnational standards in hazard mapping are of major importance for the prevention of catastrophic events according to land use in endangered areas. Abb. 2: Die Entwicklung von lnderbergreifenden Standards in der Gefahrendarstellung ist bei der Prvention von Katastrophenereignissen von groer Bedeutung, da gefhrdete Gebiete immer strker genutzt werden.

known by experts but may cause problems in practice when applied in a legal framework. It is not unusual for unsuitable types of hazard maps to be applied for the wrong purposes. For example it is often to nd landslide or risk maps. When geological mapping hazards inventory maps used as hazard

Type

Bedrock

(mass movements) in principle we have to

distinguish between two situations: 1. Scientic studies on mass movements with no legal implications (e.g. on land use planning): Typical cases are studies carried out by universities (research institutes). The aim of these studies is to understand the mechanical features of instability or to study different ways of evolution of the phenomenon (scenarios) in order to assess the susceptibility of investigated areas. Landslide inventories can be made by means of a historical or morphological approach. 2. Susceptibility/Hazard index/Hazard maps that have direct (obligatory) consequences for land use planning and building trade at different scale: The scale used to present the results of the hazard assessment depends on the desired product (susceptibility map, hazard index map, hazard zone map) and must be balanced with the precision requirements according to the spatial level of application (supra-regional, regional, local). The legal signicance of these maps requires technical standards and a stateof-the-art concerning formal requirements (e.g. investigation methods, documentation),

Tab. 1: Types of mass movements (classication) after Raetzo. Tab. 1: Typen von Massenbewegungen (Klassikation)

ASP-project ADAPTALP: Adaptation of natural hazard management to climate change Climate change is, to a large extent, constituted by increasing temperatures and changed precipitation patterns. Any change of these critical factors has implications on the frequency and extent of natural hazards including mass movements. A major impact on the intensity of mass movements at high altitudes (above 2300 m in the Alps) has thaw of permafrost and the retreat of glaciers due

Seite 11

Seite 12

hazard assessment and procedures of the check and approval of the maps. ADAPTALP (in Work Package 5) will harmonize and improve different evaluate,

a climate change adaptation strategy. The results will be summarized in a synthesis report. These elds of research within the

Anschrift der Verfasser / Authors addresses: DI Dr. Florian Rudolf-Miklau Bundesministerium fr Land- und Forstwirtschaft, Umwelt und Wasserwirtschaft, Abteilung IV/5, Wildbach- und Lawinenverbauung Federal Ministry for Agriculture, Forestry, Enviroment and Water Management, Department IV/5, Torrent and Avalanche Control 1030 Wien, Marxergasse 2 Tel.: (+43 1) 71 100 - 7333 FAX: (+43 1) 71 100- 7399 Mail: orian.rudolf-miklau@lebensministerium.at Homepage: http://www.lebensministerium.at/forst Dr. Richard Bk Amt der Krntner Landesregierung, Abt. 15 Umwelt Unterabteilung Geologie und Bodenschutz, A 9020 Klagenfurt, Flatschacher Strae 70 Tel: +43 - (0) 50536 - 31510 Fax: +43 - (0) 50536 - 41500 Mob. +43 - (0) 664 - 8053631510 Mail: richard.baek@ktn.gv.at DI Franz Schmid Bundesministerium fr Land- und Forstwirtschaft, Umwelt und Wasserwirtschaft, Abteilung IV/5, Wildbach- und Lawinenverbauung Federal Ministry for Agriculture, Forestry, Enviroment and Water Management, Department IV/5, Torrent and Avalanche Control 1030 Wien, Marxergasse 2 Tel.: (+43 1) 71 100 - 7338 FAX: (+43 1) 71 100- 7399 Mail: franz.schmid@lebensministerium.at Homepage: http://www.lebensministerium.at/forst

DI Christoph Skolaut Wildbach- und Lawinenverbauung, Sektion Salzburg Torrent and Avalanche Control, District Salzburg 5020 Salzburg, Bergheimerstrae 57 Tel.: (+43 662) 871853 303 FAX: (+43 662) 870215 Mail: christoph.skolaut@die-wildbach.at Homepage: http://www.lebensministerium.at/forst

project contain the topics to work out the minimum standards (minimal requirements) for the creation of danger (susceptibility) and hazard maps for landslides. The rst step is the evaluation of the state of the art in hazard mapping in each involved country. Two main questions will be answered by the project: What kinds of danger (susceptibility), hazard and risk maps are ofcially applied in each country? Which standards are these maps based on? The second step will be the harmonization of the different methods, which are used in several countries. Therefore similarities should be worked out and the least common denominator in the methods of hazard mapping should be found. The nal step will be the creation of guidelines and recommendation, which include the results of this harmonization. They will include minimum requirements for the creation of danger (susceptibility), hazard and risk maps. Other important results developed in cooperation with other projects as MASSMOVE will be: Denition of minimal requirements for the collection of the relevant data of endangered areas and cartographic representation of slides and rock falls. Specication of minimal requirements for the spatial description of the dangers. Development of minimal requirements for the determination of the hazard potential of slides and rock falls. Development of tools for the reduction of the risk potential by consideration of the

methods of hazard mapping applied in the Alpine area. A main emphasis will be on a comparison of methods for mapping geological hazards in the individual countries. A glossary will facilitate interdisciplinary and multilingual cooperation as well as support the harmonization of the various methods. In selected model regions methods to adapt risk analysis to the impact of climate change will be tested. This should support the development of hazard zone planning towards

Literatur / References:
BATES A. L., JACKSON J. A.: Glossary of Geology. American Geological Institute, 3rd Edition, 1987. CAMPUS S., BABERO S., BOVO S., FORLATI F. (EDS.): Evaluation and prevention of natural risks. Taylor and Francis/Balkema, 2007. GLADE T., ANDERSON M., CROZIER M. J. (HRG.): Landslide Hazards and Risk. John Wiley & Sons, Chichester, 2005. GRUNER U., WYSS R.: Anleitung zur Analyse von Rutschungen. Swiss Bull. angew. Geol., Vol. 14/1+2, 2009. RAETZO, H. , RICKLI, C.: Rutschungen. In: Bezzola G.R, & Hegg, C. (Hrsg.) 2007: Ereignisanalyse Hochwasser 2005, Teil 1 Prozesse, Schden und erste Einordnung. Bundesamt fr Umwelt BAFU, Eidgenssische Forschungsanstalt WSL. Umwelt-Wissen Nr. 0707, 2007. RUFF, M.: GIS-gesttzte Risikonanalyse fr Rutschungen und Felsstrze in den Ostalpen (Vorarlberg, sterreich). Georisikokarte Vorarlberg. Diss. Univ. Karlsruhe, 2005. SIDLE R. C., OCHIAI H.: Landslides processes, prediction and land use. American Geographical Union, water resources monograph 18, Springer Verlag, 2006.

Fig. 3: Example for a susceptibility map of the Arlberg region (Vorarlberg/Austria) after Ruff Abb. 3: Beispiel einer Suszeptibilittskarte der Arlbergregion (Vorarlberg/sterreich) nach Ruff

hazards during land use planning by the local administrations and during the land use as well as for the planning of preventive measures.

Seite 13

Key-note papers

Seite 14

Basic concept of hazard assessment Effective prevention against natural hazards requires a better understanding of the processes occurring in nature. The primary aim of hazard assessment is to gain a deep and comprehensive knowledge of these processes in order to provide accurate prognosis of the expected magnitude of hazardous events and the corresponding damaging effects. (RUDOLF-MIKLAU in SUDA ET. AL., 2011 [18.]) Another important demand is the prediction of the time of occurrence and duration of a catastrophic event (predictability FLORIAN RUDOLF-MIKLAU and advanced warning time; Fig. 1) (RUDOLFMIKLAU, 2009 [14.]). The initial purpose of hazard assessment is the provision of basic knowledge for the planning of protection measures (e.g. ood control, avalanche control), which requires quantitative information about the order and magnitude of catastrophic events and their probable damaging consequences on human health, economic activities, environment, and cultural heritage.

According to the well-established basic concept of hazard assessment, the procedure can be divided in three distinct steps (HBL ET AL., 2007 [9.]: The survey of basic information (data) The analysis of hazards (and risks) The valuation of hazards (and risks) As a rule, the survey of information related to natural hazards focuses on the acquisition of basic data on relevant factors in nature. The survey includes geo-data (topography, geology, and soil), meteo-data (climate, weather), hydrodata (precipitation, run-off, and groundwater) and eco-data (environmental parameters). In addition, data on past (historic) events represent a major source of information. (RUDOLF-MIKLAU, 2009 [14.]) For the purpose of risk assessment, data for natural processes must be combined with data related to human activities. These sources of information include demographic and economic statistics, data on land use and agriculture, and records of damages caused by past events (BRNDL ET AL., 2009 [5.]).

Principles of Hazard Assessment and Mapping Grundlagen der Analyse und Bewertung von Naturgefahren
Summary: The article summarizes the general principles for the assessment of natural hazards. The main emphasis lies on the basic approaches and methods of hazard assessment with special attention to the frequency-intensity-concept (including the decits of this approach). The strategic importance of preventive planning with regards to the use and development of endangered areas in mountain areas is discussed. In addition, a summary of the most important standards and categories of hazard (risk) mapping is provided. Zusammenfassung: Der Beitrag fasst die generellen Grundlagen der Analyse und Bewertung von Naturgefahren zusammen. Der Schwerpunkt liegt im Bereich der grundlegenden Anstze und Methoden fr die Gefahrenbewertung, wobei das Hugkeits-Intensitts-Konzept besondere Beachtung ndet (einschlielich der Dezite dieses Ansatzes). Weiters wird auf die strategische Bedeutung der prventiven Planung hinsichtlich der Nutzung und Entwicklung von gefhrdeten Gebieten im Gebirge eingegangen. Abschlieend erfolgt eine zusammenfassende Darstellung der wichtigsten Standards und Kategorien der kartographischen Darstellung von Naturgefahren.

Predictability

Drought Floods Debris ow Earthquake Rockfall Avalanches Landslides Storm Wildre Volcanism Deceases

seconds

minutes

hours

days

weeks

Advanced warning time(T)


Fig. 1: Predictability of natural hazards (RUDOLF-MIKLAU, 2009 [14.]). Abb. 1: Vorhersagbarkeit von Naturgefahren (RUDOLF-MIKLAU, 2009 [14.]).

Seite 15

Key-note papers

Seite 16

HAZARDS
Survey

RISKS
Risk analysis

regionally measurements and data from documented events ahead of application. The application of physical models is not only performed for one single data set but for a frequency range of the input values.
Fig. 2: System of hazard and risk management (RUDOLFMIKLAU/SAUERMOSER, 2011 [16.]). Abb. 2: System des Gefahrenund Risikomanagements (RUDOLF-MIKLAU/ SAUERMOSER, 2011 [16.]).

The frequency-intensity-concept is based

on extreme value statistics and is appropriate for answering two basic questions: How often does an extreme event of dened intensity occur statistically? What is the expected extreme value for a dened time period? The two established methods to analyse extreme events are the block-maxima-method and the peak-over-threshold-method (KLEEMAYR in RUDOLF-MIKLAU & SAUERMOSER, 2011 [16.]). For the statistic analysis, random and representative samples (data sets) are needed (e.g. time series of extreme precipitation). By means of statistical methods, it is attempted to conclude from properties of the sample to the rules of the total population. In technical terms, an unknown stochastic distribution function (e.g. Gumbel, Frchet, Weibull) is derived from an empirical distribution of measured values. The most common eld of application of the extreme value statistics is the prediction of weather extremes, extreme discharge in rivers and torrents of the extreme run-out distance of falls, slides or falls (mass movements or avalanches). The key problem of the method is the limited availability of measurements (data sets) that cover a sufciently long period of time. In most cases the available data represents either a too short observation (measuring) period, or is fragmentary or both. Besides this major disadvantage, the method of extreme value statistics shows other considerable short comings. Especially for torrential processes, the frequencyintensity-function shows an emergent behavior implying a limited predictability of discharge from extrapolations of measurement data when a certain threshold value is exceeded. The event disposition of a catchment or risk area, dened

Hazard analysis

Localization and topography Triggering mechanism Displacement processes/scenarios Frequency/intensitiy

Analysis of damages: direct/indirect damage Damage potential Damage scenarios

Scenarios are checked concerning their plausibility. Approaches to hazard assessment: The frequencymagnitude-concept for design events (DE) According to ONR 24800:2008 [13.] an event represents the entirety of all processes occurring in a temporal, areal and causal relationship and corresponds to a specic probability of recurrence and intensity. The extreme event represents the maximum magnitude observed in the concerning catchment or risk area. The design event (DE) is applied as reference value (criteria) for the planning of protection measures and hazard maps and represents the striven level of safety (acceptable risk). (RUDOLF-MIKLAU, 2009 [14.]) The underlying concept of intensity and frequency was originally established by WOLMAN & MILLER (1960) [19.]. Intensity in colloquial use refers to strength or magnitude of a process or event. Intensity of natural events (hazards) can be expressed by physical criteria like discharge, ow depth, pressure (process energy) or area (mass) of deposited debris. (GEBUDEVERSICHERUNG GRAUBNDEN, 2004 [7.]) In general the frequency represents the period of recurrence between two events with comparable magnitude. Frequency is often expressed as return period, which is equal to the reciprocal of the exceedance probability of extreme precipitation or discharge values. As a rule the DE is determined according to a dened return period (e.g. ood with return period of 100 years). Frequency and intensity are functionally correlated. (RUDOLF-MIKLAU in BOLLSCHWEILER ET AL., 2011 [3.])

Validation

Hazard assessment

Levels of hazard (risk) Classication of intensity Intensity criteria: e.g. pressure

Risk assessment

Validation of risks Risk acceptance (aversion)

Presentation

Process-/Suszeptibility maps Hazard (information) maps Hazard zone maps Risk map
Cartographical presentation of risks

Risk management
Management

Denition of protection goals Creation of protection concepts Management plans Protection measures Effectiveness / Efciancy

The analysis of hazards is subdivided into several tasks: the survey and localization of hazard sources, the identication of triggering factors, the description of the triggering and displacement process and the potential effects (impact) on objects. The results of the hazard analysis are usually mapped in specic types of hazard maps (e.g. susceptibility maps, intensity maps). The analysis of natural hazards provides a comprehensive image of the processes, their causes and effects, but requires additional information concerning the order of magnitude of the relevant event. (RUDOLF-MIKLAU in BOLLSCHWEILER ET AL., 2011 [3.]) Consequently, the valuation of hazards aims at the description of magnitude in a graded manner. Hazards scales, physical intensity criteria or intensity classications count among the established methods to present the magnitude of events. Usually the intensity of a hazardous process is functionally related to the frequency of its occurrence. In practice this frequencyintensity-concept is the preferentially applied

method for most natural hazards in order to value their effects (see below). (HBL, 2010 [8.]) Natural hazards in the Alpine environment are a complex system consisting of process chains with multiple interactions and dependencies. Thus the assessment of a hazard is not a mono-causal procedure but must take into account a large variety of more or less probable courses. (RUDOLF-MIKLAU in BOLLSCHWEILER ET AL., 2011 [3.]) The scenario analysis was established in risk management as an appropriate method to solve the complexity of comprehensive hazard assessment. Scenarios implicate that not only a single process but all relevant developments of an event within a dened period of recurrence are taken into account. (MAZZORANA ET AL., 2009 [12.]) In practice this means: Several applied. Models have to be calibrated with assessment historic, methods stochastic) (e.g. are morphologic,

Seite 17

Key-note papers

Seite 18

as the entirety of all conditions essential for the emergence of hazardous processes, consists of the basic disposition (susceptibility) comprising all factors immutable over a long range of time (e.g. geology, soils) and the variable disposition, which is the sum of all factors subject to a short-term or seasonal change (e.g. precipitation, saturation of soil with water, land use). If the variable disposition of a catchment or risk area is altered in the course of an event (e.g. exceedance of the water storage capacity of soil), the debris potential increases erratically, resulting in a possible transition of the predominant displacement process and a nonlinear increase of discharge. (HBL, 2010 [8.]) The practical procedure of specication of a design event can be lucidly explained by the example of a design ood (RUDOLF-MIKLAU & SEREINIG, 2010 [15.]): Generally, a design ood [discharge in m/s] with a return period of 100 years represents the striven level of safety for ood (torrent) control measures in European countries. Expected values for a rainfall and ood events of a dened return period (including a corresponding condence interval) can be derived from the hydraulic extreme value statistics. Flood statistics are based on the assumption that the observation period is representative for the long-term runoff behavior of the watershed. However, extreme ood events are qualied as statistical outliers that are not represented by the measured data collection (due to limited observation periods), but nevertheless contribute valuable information on hydrological extremes. Consequently, the statistically deduced design criterion should be supported by additional information of temporal, spatial or causal reference. Especially the dating of historic ood events from chronicles or traces in nature (ood marks, silent witnesses) can provide precious additional information on return periods, levels of ooding, or peak ood discharge. By dating historic events, extreme

oods can approximately be related to a certain return period. A causal supplement of information is gained if observed oods are analyzed with respect to their emergence regarding the weather conditions, the behavior of precipitation, and the disposition of the catchment area. In a rst step, the determination procedure of the design ood requires the specication of the expected value of discharge by means of ood statistics and additional hydrological methods. From this basic design discharge, the design ood can be derived by taking into account solid transport, transient ow conditions and inuences of stream morphology. The applicability of the frequencyintensity-concept is strongly limited for all types of hazards for which measurements or observation data of extreme events are insufciently or generally not available. In addition, it has to be taken into account that the period of recurrence of a triggering event can signicantly differ from the frequency of the impact (damage) event. Recently, alternative concepts for the assessment of magnitude of events are sought that could replace the frequency-intensity-concept. This holds especially true for the assessment of extreme mass movements and avalanches where frequency hardly can be determined with sufcient accuracy. Methods of hazard assessment The aim of hazard assessment is the determination of relevant scenarios and the related return period for the purpose of providing a prognosis of the substantial process, the extension and intensity of an event as well as for the magnitude of hazard (BRNDL ET AL., 2009 [5.]). Normally neither the physical properties of hazard processes are completely claried, nor is sufcient data on extreme events available. Consequently, the most important principle of

hazard assessment is the compliance of a high redundancy in the procedures and methods applied (KIENHOLZ, 2005 [10.]). Two principle approaches are eligible for hazard assessment (Fig. 3): The analysis of past events (retrospective indication). The prognosis of future events (foresighted indication).

Morphological Method: This method

is based on the identication of triggering/ displacement processes and the spatial distribution by means of silent witnesses (AULITZKY, 1992 [1.]) in the morphology (deposition area) and at the vegetation (e.g. trees). Dendromorphology counts among these methods, which (besides other dating methods (BOLLSCHWEILER ET. AL., 2011 [3.])) provides

Historical Method
chronicles, witnesses

Retrospective Indication
is based on the assumption, that an occured event will reoccur with comparable course and effects.

comprehensive Statistical the and means of

time

series of past events. Method: of This method includes analysis measurements

silent witnesses, dendromophology

Morphological Method

extreme value statistics, triggering mechanism

Statistical Method

Foresightes Indication
is based on the identication and analysis of factors and processes, which represent evidence for existing hazards according to gained experiences. The method presupposes knowledge about the triggering mechanism, the displacement process and the effect (impact) and includes the investigation of probability of recurrence (return period).

observation stochastic statistics). the trends prognoses a sufcient

(monitoring) data by methods (e.g. extreme value Nevertheless, (signicant) and requires

Physical/Mathematical Method
Numerical/empirical models

derivation of reliable

Expert opinion (estimation)

Pragmatic Method

Fig. 3: Principle approaches to hazard assessment (after KIENHOLZ, 2005 [10.]; modied). Abb. 3: Grundlegende Vorgehensweisen bei der Gefhrdungsanalyse (nach KIENHOLZ, 2005 [10.]; gendert).

quantity of data for a time representative (observation)

According to these principles, the following procedures can be chosen and should be applied corresponding to the rule of redundancy (HBL et al., 2007 [9.]): Historical Method: The method is based on the (qualitative and quantitative) analysis of reports, testimonies and chronicles of past events (catastrophes). This data provides evidences for the frequency of events, the triggering mechanism and the extension of the process as well as the damages occurred. As a rule, historic sources tend to be fragmentary and distorted due to subjective perception.

period. (KLEEMAYR in RUDOLF-MIKLAU & SAUERMOSER, 2011 [16.]) Physical/Mathematical Method: These methods are mainly based on numerical or empirical models, which provide information (physical criteria) for the intensity of an event for a dened return period. In practice models are the preferred tool for the determination of design events in natural hazard engineering. Due to the limited accuracy of numerical models, the application always presupposes a calibration of regional measurements (data) and the validation of the results with expert opinions. In addition,

Seite 19

Key-note papers

Seite 20

models should not only be applied for a single data set but for a range of scenarios as well as for a distribution of input parameters. A comprehensive summary of available models for torrential processes is given in BERGMEISTER ET AL. (2009) [2.], for avalanches in RUDOLF-MIKLAU & SAUERMOSER (2011) [16.]. Pragmatic Method: This method is based on the expert opinion of experiences practitioners and local experts. The pragmatic method is applied if other methods are not applicable or do not meet the goal of satisfying hazard (risk) assessment. In addition, this method serves as a redundancy and is used for the validation of results of exact assessment methods (mentioned above). Hazard assessment methods always suffer from major restrictions concerning their meaningfulness and accuracy. For the interpretation and validation of results, it is essential to know the sources of uncertainties and methodical short-comings. Some of these deciencies are summarized below (KIENHOLZ, 2005 [10.]): Limited availability of data Limited observation (measuring) period Lack of direct measurements (e.g. velocity of mass propagation during events; impact pressure) Incomplete or false documentation of past events Inconsistent quality of information and data due to variable measuring (observation, monitoring, documentation) standards Uncertainties in the selection of relevant scenarios Misjudgement of the effeminacy and condition (usability) of existing protection measures Misjudgement concerning the residual risk

Preventive planning: principles and function Prevention by planning today is qualied as the most effective measure in natural hazard management. Planning in relation to natural hazards and risks can also unfold active as passive protection effects. Planning procedures concerning natural hazards are not limited to the cartographic outline of endangered areas (areas at risk), but also provide the passivity to reduce hazards/risk by keeping endangered areas free from buildings or limiting the use of these zones (e.g. inundation areas). Thus preventive planning is the basis for the protection strategy prevention by area. (RUDOLF-MIKLAU, 2009 [14.]) In addition, the cartographic depiction of hazard zones provides the essential information (process intensity, magnitude of impact forces) for the technical protection of existing buildings. Also the suitability of planned building sites concerning the risk by natural hazards can be efciently judged on the basis of hazard maps. In development planning, the localization of new settlements can be steered away from impending hazards. (BUWAL/BRP/BWW, 1997 [6.]) In principle, in the Alpine environment the usability of land for building purposes is limited according to the expansion of hazards. In mountainous regions, the total avoidance of hazard zones for spatial development is not possible. Consequently, preventive planning denes limits (border lines) for areas that are appropriate for building. Within these limits, hazard maps provide bases for standards and regulations for a hazard-adapted construction practice. Logically, the main emphasis of preventive planning lies in the sector of hazards spatially delimited in action, such as oods, avalanches, mass movements. For natural hazards that do not allow an exact delimitation (e.g. earthquake,

storm, forest re, snow load), preventive planning is limited to rough-scale maps showing a general gradation of risks. (RUDOLF-MIKLAU, 2009 [14.]) The environmental planning is of major preventive planning can be importance for the application of hazard maps. Consequently, understood as a part of development planning. In order to regulate the use and development of endangered areas, the intervention of the state is essential. The primary goal of development planning concerning natural hazards is to keep the endangered areas free from buildings (passive protection function). The active protection function of preventive planning lies in the reservation (provision) of areas for the spreading of hazardous processes (e.g. inundation areas) or in the provision of standards (limits) for the use of endangered areas in order to reduce the risk potential. Mapping hazards in Alpine environment The cartographic outline of endangered areas according to KIENHOLZ (2005) [10.] includes the elaboration of scientic and technical bases and the depiction in hazard (indication) maps. In a second step, the geographic information provided on triggering disposition and impact intensity of hazardous processes is used for the provision of hazard zone maps and their implementation in the process of development planning. As a rule, hazard maps have no legal liability but are dened as spatial expert opinions with prognosis character, while the hazard zones become legally binding only by incorporating them into development planning documents (land use maps). Thus legal liability of hazard zones may arise on the local level depending on the national legal framework. Consequently, it is essential to adapt the standards of hazard mapping to the requirements and goal of development planning on the regional

and local level. In the Alpine countries in general the following categories of maps for the outline of hazards and risks can be distinguished: Process maps (susceptibility, intensity) Hazard (indication) maps Hazard zone maps Risk maps The following denitions are valid only with restrictions since terminology of hazard mapping substantially differs between countries and scientic branches. A hazard (indication) map roughly indicates in which areas natural hazard have to be taken into account in land use and development activities. The character of the map is only demonstrative, while no concrete information about the magnitude of the danger is provided. In many countries hazard zone maps are not available, leaving hazard indication maps as the only source of spatial information. spatial Process maps show hazards by the distribution of physical parameters

(criteria) describing the triggering, displacement and impact processes. These maps are most often the result of numerical or empirical modeling. In some countries, process maps are transformed into intensity maps showing the process criteria graded according to the levels of impact intensity (e.g. Switzerland: frequency-intensity-matrix; LOAT, 2005 [11.]). Susceptibility is dened as the extent to which an area suffers from the risk of emergence of a hazardous process if exposed to a triggering factor, without regard to the likelihood of exposure. Analogously, susceptibility maps show the disposition of an area for these events, but does not provide information about the frequency and expected intensity. Hazard zone maps show the impact of processes according to its magnitude (intensity, frequency) on the scale of the local cadastre (1.2000 1.5000). Consequently, these

Seite 21

Key-note papers

Seite 22

Overlaying this information makes feasible a comprehensive assessment of risks for human health, economic acidities, environment and cultural heritage. As shown in this article, the methods for the assessment of natural hazards still suffer from major short-comings and signicant sources of inaccuracy. In addition, a comprehensive understanding of the triggering and displacement processes of Alpine natural hazards is still
Fig. 6: Hazard zone map for torrents (including indication of landslide areas) (Austria). Fig. 4: Hazard indication map for mass movements (Bavaria, Germany). Abb. 4: Gefahrenhinweiskarte fr Massenbewegungen (Bayern, Deutschland). Abb. 5: Gefahrenzonenplan Wildbche (einschlielich des Hinweises von Rutschgebieten) (sterreich).

Literatur / References:
[1.] AULITZKY H. (1992): Die Sprache der "Stummen Zeugen". Tagungsband der Internationalen Konferenz Interpraevent 1992, S. 139-174. [2.] BERGMEISTER K., SUDA J., HBL J., RUDOLF-MIKLAU F. (2009): Schutzbauwerke der Wildbachverbauung. Verlag Ernst und Sohn Berlin (Wiley VCH). [3.] BOLLSCHWEILER M., STOFFEL M., RUDOLF-MIKLAU F. (2011): Tracking torrential processes on fans and cones. Springer Dortrecht (in preparation). [4.] BORTER P. (1999): Risikoanalyse bei gravitativen Naturgefahren. Bern: Bundesamt fr Umwelt, Wald und Landschaft BUWAL. Umwelt-Materialien 107/I+II. [5.] BRNDL M., ROMANG H., HOLTHAUSEN N., MERZ H., BISCHOF N. (2009): Risikokonzept fr Naturgefahren Leitfaden; Teil A: Allgemeine Darstellung des Risikokonzepts. Bern: Nationale Plattform Naturgefahren PLANAT (vorluge Fassung). [6.] BUNDESAMT FR UMWELT, WALD UND LANDSCHAFT BUWAL, BUNDESAMT FR RAUMPLANUNG BRP, BUNDESAMT FR WASSERWIRTSCHAFT BWW (1997): Bercksichtigung von Hochwassergefahren bei der raumwirksamen Ttigkeit, Biel. [7.] GEBUDEVERSICHERUNG GRAUBNDEN (2004): Vorschriften fr bauliche Manahmen an Bauten in der blauen Lawinenzone. [8.] HBL J. (2010): Hochwsser in Wildbacheinzugsgebieten. Wiener Mitteilungen (in press). [9.] HBL J., FUCHS S., AGNER P. (2007): Optimierung der Gefahrenzonenplanung. Weiterentwicklung der Methoden der Gefahrenzonenplanung. IAN-Report 90. Wien: Universitt fr Bodenkultur (unverffentlicht). [10.] KIENHOLZ H. (2005): Gefahrenzonenplanung im Alpenraum Ansprche und Grenzen, Imst: Imst: Wildbach- und Lawinenverbau (Zeitschrift fr Wildbach-, Erosionsund Steinschlagschutz), Nr. 152, 135-151. [11.] LOAT R. (2005): Die Gefahrenzonenplanung in der Schweiz. Imst: Wildbach- und Lawinenverbau (Zeitschrift fr Wildbach-, Erosions- und Steinschlagschutz), Nr. 152, 77-92. [12.] MAZZORANA B., FUCHS S., HBL J. (2009): Improving risk assessment by dening consistent and reliable system scenarios, Nat. Hazards Earth Syst. Sci., 9: 145159. [13.] ONR 24800: 2008, Schutzbauwerke der Wildbachverbauung Begriffe und ihre Denition sowie Klassizierung. Austrian Standards Institute, Vienna. [14.] RUDOLF-MIKLAU F. (2009): Naturgefahren-Management in sterreich. Verlag Lexis-Nexis Orac . [15.] RUDOLF-MIKLAU F., SEREINIG N. (2009): Festlegung des Bemessungshochwassers: Prozessorientierte Harmonisierung fr Flsse und Wildbche, WAW 7-8: 29 32. [16.] RUDOLF-MIKLAU F., SAUERMOSER S. (Hrsg.) (2011): Technischer Lawinenschutz. Verlag Ernst und Sohn/Wiley Berlin (in preparation). [17.] SCHROTT L., GLADE T. (2008): Frequenz und Magnitude natrlicher Prozesse; in Flegentreff, Glade (Eds.): Naturrisiken und Sozialkatastrophen. Spektrum Akademischer Verlag Springer: 134 150. [18.] SUDA J., RUDOLF-MIKLAU F., HBL J., KANONIER A. (Hrsg.) (2011): Gebudeschutz vor Naturgefahren. Verlag Spring Wien (in preparation). [19.] WOLMAN M. G., MILLER J. P. (1960): Magnitude and frequency of forces on geomorphic processes. Journal of Geology 68 (1): 54 74.

missing due to the limited availability of direct measurements and observation. Although hazard maps have gained a key role in the process of preventive planning, the information provided by these maps should still be treated with care and only be interpreted by experts. This reservation especially holds true for hazard maps devoted to mass movements. As the standards of hazard mapping in this eld are still under development, preventive planning concerning rock fall and landslides (unlike ood and avalanche hazards) is still in situ nascendi. This delay justies the strong efforts within the Alpine space to establish and harmonize general standards for the assessment and mapping of hazards caused by mass movements. Anschrift des Verfassers / Authors address: DI Dr. Florian Rudolf-Miklau

event (period of recurrence) for the assessment of the relevant hazards. (HBL ET AL., 2007 [9.]) The elaboration of risk maps is based on the depiction of objects at risk (risk potentials) within endangered areas. In principle there are two types of risk maps available (BORTER ET AL., 1999 [4.]): Risk maps only showing risk potential without assessing (value) them. Risk maps based on a graded, qualitative or quantitative assessment of risks (levels of risk; e.g. low medium - high). These maps are elaborated by combining the impact intensity with the damage potential (value), the vulnerability and the exposition of objects/persons in the endangered area. Closing remarks Hazard (risk) assessment and mapping count among the most important tasks (measures) in natural hazard management. The maps provide the key information for most of the other mitigation measures in order to reduce risk to an acceptable level. GIS technology provides a powerful tool to

maps provide specic information about the usability of certain plots for building or other development purposes. Hazard zone maps are regularly produced for the hazard types oods, avalanches and debris ow, and only in few countries (Switzerland, France, and Italy) for mass movements as well. In most countries, hazard zone maps are regulated by legal and technical standards concerning their content, formal requirements, approval procedure and implementation in the development planning. Some countries have also dened a specic design

Bundesministerium fr Land- und Forstwirtschaft, Umwelt und Wasserwirtschaft, Abteilung IV/5, Wildbach- und Lawinenverbauung Federal Ministry for Agriculture, Forestry, Enviroment and Water Management, Department IV/5, Torrent and Avalanche Control 1030 Wien, Marxergasse 2 Tel.: (+43 1) 71 100 - 7333 FAX: (+43 1) 71 100- 7399 Mail: orian.rudolf-miklau@lebensministerium.at Homepage: http://www.lebensministerium.at/forst

Fig. 5: Hazard map for falls (rock fall) (Switzerland). Abb. 5: Gefahrenzonenplan Felssturz (Steinschlag) (Schweiz).

combine spatial information on natural hazards with other cartographic information concerning human activities and development actions.

Seite 23

Key-note papers

Seite 24

RICHARD BK, HUGO RAETZO, KARL MAYER, ANDREAS VON POSCHINGER, GERLINDE POSCH-TRZMLLER

Mapping of Geological Hazards: Methods, Standards and Procedures (State of Development) - Overview Geologische Gefahrenkartierung: Methoden, Standards und Verfahren (derzeitiger Status) ein berblick
Summary: In spite of different methods used, geological hazard mapping is accepted as a tool for hazard prevention in Europe. Scientic characterization of mass movements is based on similar methods with mostly comparable results. However, the implementation in spatial planning and risk management differs considerably due to different regional legal acts, ordinances, responsibilities and pecularities. Whereas in Italy and Switzerland there are technical guidelines and legal acts regarding landslides and rock fall, in Austria only hazard mapping concerning oods and avalanches is regulated. In Germany a recommendation on how to create a susceptibility map was published. Because of a lack of regulations in European Alpine states inventory maps, susceptibility and hazard maps are created in different scales with different contents and quality. This, as well as different dentions of terms such as susceptibility, danger and hazard, makes comparison of hazard assessment products difcult. Consequently a multilingual glossary, landslide inventories at regional authorities and minimal requirements as to how to create hazard maps (requirements concerning input data and purpose of assessment) are necessary. In the AdaptAlp project (Interreg IV B, Alpine Space) the Alpine regions elaborate the common principles.

Zusammenfassung: Die geologische Gefahrenkartierung ist in Europa trotz unterschiedlicher Methoden eine anerkannte Notwendigkeit fr die Prvention. Die wissenschaftliche Charakterisierung der Massenbewegungen basiert oft auf hnlichen Methoden und ist deshalb eher vergleichbar. Hingegen ist die Umsetzung in die Raumplanung und in das Risikomanagement auf europischer Ebene sehr unterschiedlich. Der Grund liegt primr in unterschiedlichen Gesetzen, Verordnungen und Verantwortlichkeiten, bzw. in sozio-konomischen Eigenheiten der Lnder. Whrend in Italien und in der Schweiz technische Richtlinien bzw. gesetzliche Regelungen zur Erstellung von Gefahrenkarten bestehen, gibt es in sterreich nur fr Hochwasser bzw. Lawinen Regelungen zur Ausweisung von Gefahrenzonen. In Deutschland wurde eine Empfehlung fr die Erstellung von Gefahrenhinweiskarten publiziert. Aufgrund fehlender Regelungen in den alpinen Staaten Europas werden Ereigniskarten, Indexkarten, Gefahrenhinweiskarten und Gefahrenkarten als Grundlagen fr die Gefahrenbeurteilung in verschiedenen Mastben mit unterschiedlichem Inhalt erarbeitet. Dies und unterschiedliche Denitionen erschweren den Vergleich. Ein multilinguales Glossar, die Einrichtung von Ereigniskatastern bei der Verwaltung und die Festlegung von Mindestanforderungen zur Erstellung von Grundlagen und Gefahrenkarten (Anforderungen hinsichtlich Eingangsdaten und Zweck) sollten daher ein primres Ziel sein. Im Projekt AdaptAlp (Interreg IV B, Alpine Space) arbeiten die Alpenlnder an gemeinsamen Grundstzen.
countries varies in its quality and quantity: In Introduction In Alpine regions, slopes of different some regions, detailed landslide inventories exist and are the basis for susceptibility and hazard assessment. Different approaches to hazard mapping are in practice. This fact and dissimilar meanings for terms like susceptibility, danger and hazard make a comparison of the regional approaches difcult. Using various input data also handicaps the comparison of hazard assessment. Within the INTERREG IV B project Adaptation to Climate Change in the Alpine Space (acronym AdaptAlp), work package 5.1 Hazard Mapping - Geological Hazards is focusing on the transnational harmonization of standards (minimal requirements in the eld of hazard assessment and mapping) by exchanging experiences in the partner regions. This issue provides an overview of methods, standards and procedures without a pretense of completeness. The denitions of terms used regarding morphological and geological conditions are prone to landslides. Taking into consideration one of the geological principles for landslide hazard assessment the past is the key to the future future slope failures will probably occur in areas with similar geological, morphological and hydrological situations that have led to past failures. Some triggering mechanisms happen sporadically and are not readily obvious. Because of the lack of memories of past landslide events, the susceptibility to mass movements is not considered accurate in land use. But the effects of mass movements (damages) necessitate new strategies on how to manage the future potential of natural (geological) hazards in alpine regions. Information about landslides in alpine

Seite 25

Key-note papers

Seite 26

landslides sometimes differ contradictorily in literature and in practice. For this reason the second goal of the work package 5.1 named above is the elaboration of a multilingual glossary. Landslide inventories Landslide inventories are the basis for all scientic and planning activities. They contain the basic data of natural hazard processes and should mainly include the facts. Therefore all partner countries in the AdaptAlp Interreg project are working on landslide inventories. [11] Guzzetti 2005 wrote about landslide inventories: Despite the ease with which they are prepared and their immediateness, landslide inventories are not yet very common. Inventory maps are available for only a few countries and mostly for limited areas. This is surprising because inventory maps provide fundamental information on the location and size of landslides that is necessary in the assessment of slope stability at any scale, and in any physiographical environment. Nevertheless, all of the countries considered for the literature survey have landslide inventories and maps, even if contents, scales and the state of completeness vary. In order to predict landslide hazard in an area, the morphological, geological, and hydrological conditions and processes have to be identied. Their inuence on the stability of the slopes has to be estimated. Different methods of data acquirement are used to establish databases to assess hazards: Landslide inventories as an important tool for the assessment of the susceptibility of slopes to mass movements are created nowadays more and more using digital technology. A general indication of landslide susceptibility can be obtained based on landslide inventories, geological, soil and geomor-

phological maps. Using digital DTM data in a GIS allows the production of hillshades with several geometries to detect typical landslide forms. Modern methods for modelling processes are designed for the GIS environment. Slope stability and rock fall trajectories can be computed over large areas to get indications of the hazards. Analysis of aerial photographs is also a classical and valuable technique to identify landslide features. More subtle signs of slope movement cannot be identied on the maps mentioned above. Field observation by experts is necessary for accurate assessment. The requirements for acquired data are raised by the main goal: The accurateness and detail of input data and scale depends on the aim of the product susceptibility map, hazard assessment or risk analyses. For hazard assessment, information about possible scenarios is needed. For this reason it is important that landslide inventories are induced to sustain landslide knowledge over time. In most regions of the Alps, inventories have been established by authorities and are to some extent available to the public. Tab.1 gives information about what kind of data is stored in different landslide event inventories, and what questions are asked on the landslide reporting form. For the comparison, information from the countries Austria (Geological survey of Austria, of Lower Austria, of Carinthia, project MASSMOVE, project DIS-ALP), Germany, Switzerland, Slovenia, Italy, France, Slovakia, Australia and the USA (Oregon, Washington, Utah) was taken into account. if The rst section of table 1 shows inventories exist. The second section

give evidence, if e.g. information on the activity, geometry and slope position of a landslide is recorded. Recorded geological information (fourth section) is sometimes specied in detail, sometimes only the information is given that geological information is being stored. In many cases additional information such as data on vegetation (land cover), hydrogeological or hydrological conditions, as well as specic data such as the shadow angle are stored in the databases. Most inventories provide information on the causes or triggers of landslides. In some cases the damages due to landslides are listed in the inventory, sometimes even the monetary value of the damage and the costs of remediation measures. Most inventory forms also provide information about how the listed data was gathered (e.g. eld survey), some provide a rating about the reliability of the degree of precision of the information. In most databases additional reports, documentation and bibliography are included or mentioned. In Austria the Geological survey of Austria, in cooperation with the Geological Survey of Carinthia, has created not just one inventory map but a level of information (Fig. 1): Process index maps (map of phenomena Prozesshinweiskarte, Karte der Phnomene) can have different scales (1:50,000 and bigger) and can be of varying quality; it contains information about process areas and phenomena of mass movements that have already happened. The event inventory (Ereigniskataster) records only those processes for which an event date is known (5W-questions); it is independent of a scale. In Carinthia, a digital landslide inventory was created with historical events of the last 50 years ([7] Bk et al. 2005). The inventory map/event map (Ereigniskarte) contains only information about processes for which an event

date is known. The thematic inventory map contains only information related to a type of process, categorized according to the quality of the data. In Switzerland, the generation of a map of phenomena is mandatory ([30] Raetzo 2002). As with the Austrian map of phenomena, it shows the geologic-geomorphologic features. An extensive manual with a digital GIS-legend was published on a DVD by BWG ([8]BWG 2002, [14] Kienholz & Krummenacher 1995). The scale used depends on the purpose the map is used for, ranging from 1:2,000 (or even more) for a detailed study to 1:50,000 as an indicative map ([32] Raetzo & Loup 2009). On the other hand, the Federal Ofce for the Environment (FOEN) manages a database with all the events where damages were recorded. This national database is called StorMe and contains data on every natural hazard process: landslides, debris ows, snow avalanches and oods. In Italy, a country with a particularly high landslide risk owing to its landform conguration and its lithological and structural characteristics, the need for a complete and homogeneous overview of the distribution of landslides was recognized after the disastrous event at Sarno. The aim of the IFFI Project (Inventario dei Fenomeni Franosi in Italia Italian Landslide Inventory) implemented by ISPRA (formerly: APAT, the Italian Environment Protection and Technical Services Agency) and by the regions and selfgoverning provinces was to identify and map the landslides in accordance with standardized and shared methods. The work method included the collection of historical and archive data, aerial photo interpretation, eld surveys, and detailed mapping. A Landslide Data Sheet was prepared for collecting the landslide information, subdivided into three levels of progressively

deals with the basic data, mainly with the 5W-questions: What happened where, when and why, and who reported it (or made the database entry). The landslide conditions in the third section

Seite 27

Key-note papers

Seite 28

increasing detail (from: [13] ISPRA, 2008): First level: contains the basic information (location, type of movement, state of activity) and is mandatory for every landslide. Second level: contains the geometrical, geological, and lithological parameters, land use, causes and activation date. Third level: provides detailed information on the damage, investigations and remedial measures. A scale of 1:10,000 is used for surveying and mapping the landslides throughout most of Italy, only in high mountainous areas or in lower populated areas is a scale of 1:25,000 used. As with many regions, the region of South Tyrol (Autonome Provinz Bozen Sdtirol, [27] Nssing 2009) also has a landslide database that resulted from the IFFI Project. The type of movement, the litho-logical unit, the volume of the moving masses, the internal cause and the external trigger, as well as the induced damage are noted for each event. The extensive landslide database, GEORISK of landslides Bavaria, is an essential step to been documented in the

that, since the sources for the inventory map of Slovenia are quite different from each other, the scales vary but landslides were always mapped at a quite detailed scale. In France a database for mass movements is accessible on the internet. The processes taken into account are landslides, rock fall, debris ows, subsidence and bank erosion. For each mass movement, the following detailed information can be retrieved: type of movement, detailed geographical data, information about the quality, the precision and the origin of the data, detailed information about the mass movement (size, activity), the damage caused, the causes for the movement and geological information as well as information about the survey of the phenomenon. A prototype landslide database has been established by Geoscience Australia in collaboration with the University of Wollongong and Mineral Resources Tasmania, displaying the location of the landslides on a map and providing information regarding the type of landslide, date of occurrence (if known), a brief summary of the event, its cause and damage. In England after the Aberfan disaster the UK government funded a number of research projects to look at the UKs geohazards ([33] Reeves 2010). Now in the UK the BGS investigates geohazards by looking at primary geohazards such as earthquakes, volcanic eruptions and secondary geohazards such as landslides, swelling/shrinking etc. Topics of consideration are the cause of events, return periods determined by analysis of past events, affected regions, inuence of regional geology. An inventory is the rst step in building an understanding of the occurrence of geohazards. Currently BGS maintains two main shallow geohazard databases: the National Landslide and the Karst Database. These inventories provide the basis for analysing the spatial distribution

of the geohazards and their causal factors. This understanding can be used to assess susceptibility. In the USA the Landslide Inventory Steering Committee, composed of members of USGS and State Geological Surveys and other state agencies, are working on the Landslide Inventory Pilot Project. The purpose of this project is to provide a framework and tools for displaying and analyzing landslide inventory data collected in a spatially aware digital format from individual states. To get information about further landslides, the Oregon Department of Geology and Mineral Industries, among others, has prepared an inventory form. Besides information about the exact location (coordinates) of a landslide, the following specications should be listed: date of slide, activity, estimated dimension (length, width, depth, volume, estimated dimensions from: aerial photos, eld evaluation), predominant type of material (rock, debris, earth, ll), predominant type of movement (fall/topple, ow, translational slide, rotational slide, spread), approximate original slope (e.g.: 30 +/- 5, estimated from e.g. 1:24K USGS topo map), land use where slide occurred (forested area, harvested area, rural area, urban area, agriculture), cause of slide (road construction, road cut, road ll, earthquake, preexisting slide, steep natural slope, natural drainage, human built drainage, other), damage caused by slide and additional comments. In California the landslide inventory maps are available at a scale of 1:24,000. The inventory was prepared primarily by geomorphological analysis, interpretation of aerial photographs and also by eld reconnaissance, interpretation of topographic map contours, and review of geological and landslide mapping. Also, each landslide was classied according to its activity: active or historic, dormant-young, dormant-mature, dormant-old. The landslide

material (rock, soil, earth, debris) and type of movement (slide, ow, fall, topple, spread) are also classied. Furthermore, each landslide is classied according to a condence (denite, probable, questionable) assigned by the geological interpreter. It can be regarded as a measure of likelihood that the landslide actually exists. Susceptibility/hazard assessment in Alpine regions A literature study regarding susceptibility/hazard mapping ([29] Posch-Trzmller 2010) shows the different approaches to hazard assessment in alpine regions. For the assessment of natural hazards (hazard maps) mainly heuristic methods are in practice. In this case scientic reports, geological and morphological mapping are the basis for weighting methods. Statistical analysis (bivariante or multivariate) are used for the weighting. The weight of evidence method is based on a statistical Bayesian bivariate approach. Originally developed for ore exploration, this probabilistic method is now commonly used for the statistical assessment of landslides. It is based on the assumption that future landslides would be triggered or inuenced by the same or similar controlling factors as previously registered landslides ([15] Klingseisen & Leopold 2006, [16] Klingseisen et al. 2006). In Germany a recommendation on how to create a susceptibility map is given by the Geohazards team of engineering geologists of German federal governmental departments of geology ([37] SGD 2007). Basic minimal requirements for inventory records are dened, such as spatial positioning and technical data of mass movements. Digital modelling (rock fall, shallow landslides) can be used to identify the susceptibility of areas to mass movements, veried by landslide inventories or evaluation through

creating susceptibility maps. Until now 2,800 have database, with information about the type of movement, the extension, age and status of the landslides. The following landslide processes are recorded: ow ("Hangkriechen", "Schuttstrme"), slide ("Rutschungen", "Hanganbrche"), fall/rock fall ("Steinschlge", "Felsstrze", "Bergstrze"), Karst, subsidence ("Erdflle", "Dolinen", "Senken", "Schwinden",..). Based on the inventory, maps were created, showing existing landslides and their activity (Karten der Aktivittsbereiche). The Slovenian landslide inventory map is shown as a small inlet on the susceptibility map of Slovenia at a scale of 1:250,000. Personal information from M. Komac (Geo ZS) revealed

Seite 29

Key-note papers

Seite 30

eld work. Indications of active/inactive landslides can be found by using registers, mapping and/or remote sensing (DTM) methods. Potential landslide areas (where landslides have not yet taken place) are determined by empirical methods in account of geological and morphological situation and land use. Alternatively areas prone to landsliding can be derived semi-automatically by a cross-over between DTM and a geological entity. Regarding rock fall processes, source areas of rock fall are derived in a rst step from landslide inventories and/or remote sensing (DTM). Usually Alpine areas with an inclination > 45 are potential rock fall escarpments. In the second step, the runout zone is depicted by empiric angle methods (shadow angle, geometric slope angle) or physical deterministic methods. The guidelines also include ow processes, subrosion, subsidence and uplift. For the whole Bavarian Alps (about 4.300 km) ([23] Mayer 2007), an extended danger map at a scale of 1:25,000 has already been presented or is being completed. That means that, in contrast to the susceptibility map (without information on intensity and probability), it includes a qualitative statement about the probability through a predened design event. The legend for the rock fall danger map discerns between indication of danger, yes or no, the legend for the danger map of supercial landslides discerns 3 entries (source area, accumulation zone, none), the deep-seated landslides danger map also discerns 3 entries (indication, indication in extreme case, none). The Swiss indicative map (Gefahrenhinweiskarte) is generated at a scale of 1:10,000 to 1:50,000. The legend gives only the information indication of hazard - yes or no, without specication of classes. It indicates the potential process areas of rock falls, landslides and debris ows. It doesnt include information about intensity or probability. The creation of an

indicative map is not obligatory in Switzerland, since the law refers to the standardized hazard map ([32] Raetzo & Loup 2009). Detailed information on hazard maps in Switzerland is given by Raetzo & Loup in this issue [31]. Because of the lack of a regulatory or technical norm concerning framework

statistics (landslides) and cost analysis (rock falls), working with a 25x25m grid. The inventory map is included in the susceptibility map. Also, the local department of the Austrian Service for Torrent and Avalanche Control (WLV) creates hazard maps within the hazard zonation plan. In Upper Austria, Lower Austria and Burgenland, different approaches have been chosen to develop susceptibility maps (different scales, processes) derived from existing data sets and maps ([29] Posch-Trzmller 2010): The main focus in Burgenland is concentrated on shallow landslides with an annual movement rate of 1-2cm. For the prediction of landslide susceptibility based on morphological and geological factors, the method called Weights of Evidence was chosen ([16] Klingseisen et al. 2006). In Lower Austria susceptibility maps have been created until now using a heuristic approach based on geological expertise, historical data and interpretation of DTM and aerial photos ([36] Schweigl & Hervas 2009). To provide the municipalities with assistance in spatial planning, landslide susceptibility maps were generated for the main settled areas in Upper Austria (O). The priority, which is a susceptibility class, was evaluated on the basis of the in-tensity and the probability of an event for each type of mass movement ([19] Kolmer 2009). As these maps include the intensity and the frequency of mass movements, they can be called hazard maps by denition. Nevertheless it has to be taken into account that the method of generating these maps did not include either eld work or remote sensing techniques. The method of assessment is based solely on geological expertise. The national project of Italy, IFFI, also represents an important tool for landslide risk assessment, land use planning and mitigation measures. By using the information contained in the database of the IFFI Project and the Corine Land Cover Project 2000, it was possible to carry

out an initial evaluation of the level of attention on a municipal basis. The level of attention was for example rated very high, when the landslide points, polygons and lines intersected with urban, industrial or commercial areas ([13] ISPRA 2008). The regions in Italy also have programs in cooperation with the IFFI Project (IFFI started as a national project and is continued by the separate regions), as well as with the PAI Project. For example, the region of Friuli Venezia Giulia has a landslide inventory that originated within these two studies, collecting data from several different regional ofces (in particular: Protezione Civile della Regione and the Direzione Centrale Risorse Agricole, Naturali, Forestali e Montagna) as well as from other public subjects that work on the territory. It homogenizes the information according to national standards and surveys new data. The program is used for the evaluation of the hydrogeological hazard and risk and also to give a clear and updated view of the interventions made in the region to preserve vulnerable areas. The data is recorded in an ofcial GIS structure called Sitgeo (Geological Service Information System). The main focus lies on hazard assessment at the scale of a slope. Slovenia generated a susceptibility map of the whole country at a scale of 1:250,000 using statistical analyses ([20] Komac & Ribicic 2008). In 2002, BGS (England) developed a nationwide susceptibility assessment of deterministic geohazards such as landslides, skrink-swell, etc. called GeoSure ([33] Reeves 2010). It was developed from the 50K digital geology polygons (DiGMap50), published information, expert judgement knowledge, national landslide database, database national and geotechnical DTM. information Probabilistic modied

landslides and rock fall in Austria - only the course of actions concerning oods, avalanches and debris ows are regulated by law (ordinance of hazard zone mapping, [34] Rudolf-Miklau & Schmidt 2004) - the federal states all follow a different course of action. a At the Geological Survey of Austria, database-system about for the documenting different mass of

movements in Austria (GEORIOS) containing information types processes, geological, hydrological, geometric and geographical data, information on studies or tests carried out as well as mitigation measures and the source of information (archives, eld work) is in use. Susceptibility maps in different scales and with different methods (heuristic approach, neural network analysis) have already been generated. Using the digital geological map (1:50,000), the inventory map, map of phenomena and a lithological map, susceptibility maps for Carinthia were generated in collaboration with the Geological Survey of Austria (GBA) and the Geological Survey of Carinthia at a scale of 1:200,000 ([17] Kociu et al., 2006). These are, of course, still lacking information about intensity and recurrence period or probability of occurrence. For a small study area in Styria, the Geological Survey of Austria generated a susceptibility map at a scale of 1:50,000 using neural network analysis ([38] Tilch 2009). In Vorarlberg risk maps (susceptibility map, vulnerability map, risk map) were produced in the course of a university dissertation ([35] Ruff 2005). For modelling, he used bivariate

methods are used for hazard management by primary geohazards, deterministic methods by secondary geohazards.

Seite 31

Key-note papers

Seite 32

A number of guidelines have been published in Australia by the Australian Geomechanics Society concerning mass movements and landslide risk management, as well as slope management and maintenance. These guidelines are tools that were made to be introduced into the legislative framework of Australian governments at national, state and local levels, and they are also useful for land use planning. Regional susceptibility mapping of areas prone to landsliding is not yet commonly undertaken in Australia: Because of a lack of good inventory maps and validated inventory databases, landslide hazard mapping is very limited. Determining temporal probability is often not possible because of the lack of historical information ([25] Middleman 2007). Landslide mapping is generally done on a site-specic scale and is performed by geotechnical consultants for the purpose of zoning, building infrastructure and applying for development approvals ([25] Middleman 2007). Mineral Resources Tasmania (MRT, Department of Infrastructure, Energy and Resources, State Government of Tasmania) is the only state government agency in Australia to undertake several activities with respect to landslides, including regional mapping, administration of declared landslide areas and monitoring of a small number of problematic landslides. Mazengarb ([24], 2005) describes in detail the methodology of creating the Tasmanian landslide hazard map series that started with a pilot area coinciding with the Hobart municipality. The following basic information was used to create the individual landslide hazard maps (note: In the report the maps are called hazard maps, but on the homepage, where the maps are accessible via the internet, the individual maps are called susceptibility maps, but, nonetheless, giving hazard zones in the legends.): geological

mapping (1:25,000), geomorphological mapping and analysis (1:5,000), landslide and engineering data compilation, construction of digital elevation models (10x10m). For example, a threshold slope value of 42 was chosen for modelling rock fall source areas. It does not imply that rock fall will not occur on lower slopes, but it becomes steadily less likely with reduced slope angles. A simple modelling approach was developed for modelling the rock fall runout area using the direction of maximum downhill slope dened by an aspect raster and calculating with a travel angle of 30. In southwestern California, soil-slip susceptibility maps have been produced. These show the relative susceptibility of hill slopes to the initiation of rainfall triggered soil slip-debris ows. They do not attempt to show the extent of runout of the resultant debris ows. The susceptibility maps were created in an iterative process from two kinds of information: locations of sites of past soil slips and aerial photographs taken during six rainy seasons that produced abundant soil slips. These were used as the basis for a soil slip-debris ow inventory. Also, digital elevation models (DTM) of the areas were used to analyze the spatial characteristics of soil slip locations. Slope and aspect values used in the susceptibility analysis were 10 metre DTM cells at a scale of 1:24,000. For convenience, the soil-slip susceptibility values are assembled on 1:100,000 scale bases ([26] Morton et al. 2003). Comparison of hazard assessment methods Methods of hazard assessment used in Switzerland, Italy (Friuli Venezia Giulia), Australia, France and USA are considered in this section. First the Swiss and the Italian methods are compared, as these dene intensity and probability parameters. The

Australian method of hazard assessment, which is quite different from the rst ones, as well as the method applied in the state of Washington (USA), is also looked into (Tab. 2). Tab. 3 gives an overview about hazard maps generated in the considered countries. Comparison of hazard assessment methods in Switzerland and Friuli Venezia Giulia (Italy) The hazard maps in Switzerland are compared especially to Friuli Venezia Giulia. More detailed information on the Swiss method is given by Raetzo & Loup in this issue [31]. The Swiss method ([30] Raetzo 2002) and the method used in Italy ([21] Kranitz & Bensi 2009) are based on an intensity-probability matrix. They differ from each other in determining the intensity and the probability of a landslide event. In Switzerland, 5 degrees of hazard are used. In Italy the hazard is rated in 4 classes (from very high [P4] to moderate [P1]). Concepts of hazard assessment in Switzerland In Switzerland the method to establish the hazard map was simplied as much as possible due to the objective of facilitating its integration into land use (planning). In order to have simple construction regulations, only 5 degrees of hazard were dened: high, medium, low, residual and neglectable hazard. The degree of hazard is dened in a hazard matrix based on intensity and probability criteria ([32] Raetzo & Loup 2009). For the planning of protection measures, more detailed investigations and calculations are done (e.g. all energy classes). In general the methods used are related to the product, scales and the risk in order to respect economic criteria. Applying this concept, low efforts were used for the swiss

indicative map (level 1). Important efforts are taken when a hazard map is established or reviewed (level 2). Hazard maps are an accurate delineation of zones on scales from 1:2,000 to 1:10,000. Detailed analyses and engineering calculations are foreseen for the planning of countermeasures or for expertises (level 3). It is planned to apply this concept of increased efforts for geological investigations when the assessment takes place on the second or third level. These investigations include analyses, geologic mapping, geomorphologic numerical monitoring, geophysics,

modelling and other methods. Assessment of the intensity (Switzerland/ Friuli Venezia Giulia) Intensities are assessed through a classication that is represented in table 2. The assessment of intensities in Switzerland is different for each process, also for oods and snow avalanches ([30] Raetzo 2002). For continuous landslide processes, the only criterion is the intensity. For spontaneous processes the intensity and the probability both ranging from high to low in three classes (high medium low) are needed: For rock falls, the intensity is dened by the energy. High intensity is dened as e300kJ, which is approximately the limit of resistance of massive armored walls. For slides, the mean long-term velocity, the variation of the velocity (dv, or acceleration), the differential movement (D), and the depth of the slide (T) are used to determine the intensity ([32] Raetzo & Loup 2009). For owing processes like earth ows, the potential thickness and the possible depth of the depo-sition determine the intensity.

Seite 33

Key-note papers

Seite 34

For landslides and rock falls the Swiss evaluation is normally based on intensity maps where 3 or more classes can be chosen. (e.g. 10-20,000 kJ for rock falls). In Italy, different methods of assessment are used. For example, the regional method of Friuli Venezia Giulia ([21] Kranitz & Bensi 2009) for rock fall: The intensities are classied by different methods using several tables. For fall processes, a table with denition of classes of the geometry is determined (after [12] Heinimann et al. 1998). The classication takes into account the block size of the rocks ([21] Kranitz & Bensi 2009). Another table determines the velocity factor (v), also ranging from 1- 3, using the denitions from Cruden & Varnes ([9], 1996). The intensity class, ranging from 1- 9, is then determined with the geometry-velocity matrix. Comparison between the Swiss and the Italian intensity classication: The differences in determining the intensity between the Swiss ([32] Raetzo & Loup 2009) and the Friuli method ([21] Kranitz & Bensi 2009) are: For fall processes in the Italian method, the energy does not need to be calculated, only the block sizes and the velocity need to be determined, while in Switzerland the energy is calculated. The Italian method does not differentiate for continuous processes. Switzerland uses the mean long-term velocity for these continuous landslides. The Swiss method determines 3 intensity classes to apply within the hazard matrix for the land use planning. If protection measures are planned in Switzerland, all the energy values are taken into account. The Italian method determines 9 intensity classes.

Assessment of the probability (Switzerland/ Friuli Venezia Giulia) Swiss method ([32] Raetzo & Loup 2009): The probability assessment of the Swiss method denes the probability in analogy to the recurrence periods used in ood and avalanche protection (30, 100, 300 years return period), which corresponds to yearly probabilities of 0.03, 0.01 and 0.003. An event with a return period higher than 300 years is normally also considered for the assessment (risk analysis, residual risk,). It corresponds mainly to the ood prevention strategy. The probability of an event has to be calculated or estimated: Big events (Bergsturz, >1mio m3) do not recur. For smaller events the probability is dened by the elements at risk. For continuous slides the probability is 1 (or 100%), meaning that the event is happening already. Scenarios are dened when sudden landslide failure or acceleration can take place. When fast moving landslides (debris or earth slides according to Varnes) have long run-out distances, the process is moving into a ow. In this case the Swiss method takes into account the change from the rst to the second move and criteria of the ow processes are applied (see below). The probability for debris and earth ows is determined through eld work and based on inventory data. Numerical modelling of ow processes is also used and the importance of these results is rising. Method of Friuli Venezia Giulia ([21] Kranitz & Bensi 2009): The possible frequency or occurrence probability is determined through the records of historical events. If there is a lack of sufcient historical data

for the statistical evaluation of the return period, the values will be assigned by a typological approach based on bibliographical data inherent to the characteristics of temporal return of the various typologies of landslides. This will be calibrated on geomorphologic observations, analyses of historical photos, and aerial pictures (which is also the case in the Swiss method) from the year 1954 up to now, and historical data from local sources. The probability is then classied in 4 classes: high: 1-30 years (active landslides, continuous and/or intermittent landslides, quiescent episodic with high frequency) medium: 30-100 years (quiescent episodic landslides with medium frequency) low: 100-300 years (quiescent episodic landslides with low frequency) >300 years (ancient landslides or palaeolandslides). Other approaches to hazard assessment

Australia In the Australian guidelines for landslide susceptibility, hazard and risk zoning for land use planning, the number of events per length of source area per year (rock fall) or per square kilometer of source area per year (slides) is used for describing the hazard of small landslides. For large landslides, the annual probability of active sliding or the annual probability that movement will exceed a dened distance or the annual probability that cracking within a slide exceeds a dened length is used to describe the hazard. The description of the hazard should include the classication and the volume or the area of the landslides. Whether landslide intensity is required for hazard zoning is to be determined on a caseby-case basis. For rock fall hazard zoning, it is likely to be required. Therefore the frequency assessment is much more important for hazard zonation than the intensity according to AGS. Intensity assessment in Australia:

France Malet et al. ([22] 2007) describes the French methodology for landslide risk zoning (Plan de Prvention des Risques), where 3 classes of risk (R1, R2, R3) with specic rules for land use regulations and urbanism can be represented in a matrix depicting hazards and potential consequences. This qualitative method is based on the expert opinion of the scientist. No specic investigation is necessary, available data and reports are sufcient. The scale of work is specied as 1:10,000. The hazard map is an interpretation of the type of processes, activity, age and magnitude of the processes; the hazard map is an interpretation of the type of processes, activity, magnitude and frequency. The risk map is the crossing of the hazard map and the inventory map of major stakes ([22] Malet et al. 2007).

The landslide intensity is assessed as a spatial distribution of: the velocity of sliding coupled with slide volume or the kinetic energy (e.g. rock falls, rock avalanches), or the total displacement or the differential displacement or the peak discharge per unit width (m3/m/ sec., e.g. debris ows) For basic and intermediate level assessments of intensity, only the velocity and volume might be assessed. But for the advanced assessments of rock fall or debris ow hazard, the energy should be determined. In AGS ([3] 2007b) it is noted that there is no unique denition for intensity. Those carrying out the zoning will have to decide which denition is most appropriate for the study.

Seite 35

Key-note papers

Seite 36

Frequency assessment in Australia: In AGS ([3], 2007b), the assessment of the frequency of a landslide event for the generation of hazard maps is usually determined from the assessment of the recurrence intervals (the average time between events of the same magnitude) of the landslides. If the variation of recurrence interval is plotted against magnitude of the event, a magnitude-frequency curve is obtained. The methods listed for determining the frequency include: historical records; sequences of aerial photographs and/or satellite images; silent witnesses; correlation with landslide triggering events (rain storms, earthquakes); proxy data (e.g. pollen deposition, lichen colonization, fauna assemblages in ponds generated by a landslide,); geomorphologic features (ground cracks, fresh scarps,); subjective assessment. It is further noted that landslides of different types and sizes do not normally have the same frequency (annual probability) of occurrence. Small landslide events often occur more frequently than large ones. Different landslide types and mechanics of sliding have different triggers (e.g. rainfalls of different intensity, duration and antecedent conditions; earthquakes of different magnitude and peak ground acceleration) with different recurrence periods. Because of this, to quantify hazard, an appropriate magnitude-frequency relationship should in principle be established for every landslide type in the study area. In practice, the data available is often limited and this can only be done approximately. A row of useful references on frequency assessment are listed in AGS ([3], 2007b). In AGS ([1], 2000) it is noted that even if extensive investigation is carried out, assessing the probability of landsliding (particularly for an unfailed natural slope) is difcult and involves much uncertainty and judgement. In recognition

of this uncertainty, it has been common practice to report the likelihood of landsliding using qualitative terms such as likely, possible or unlikely. Procedures of hazard mapping in the considered regions Tab. 3 gives an overview of hazard maps generated in the considered countries. In Germany a recommendation on how to create a susceptibility map is given by the Geohazards team of engineering geologists of German federal governmental departments of geology ([37] SGD 2007). In 2007, the LfU completed the Landslide susceptibility map of Oberallgu (Bavaria). For this map, the processes of rock falls, supercial landslides and deep seated landslides were treated separately. The susceptibility maps for rock falls and supercial landslides were created using modelling, whereas the susceptibility map for deep seated landslides was created empirically, assuming that deep seated landslides tend to occur in areas already affected by landslides in the past, but taking into consideration that process areas can expand during reactivation of a landslide. The basic data used for the investigations contained the following: topographic map 1:25,000, raster format; geological map 1:25,000 or 1:50,000 and also maps in smaller scales where the detailed maps were not available, vector format; DTM, 10m raster data; aerial photographs 1:18,000 and orthophotos; data on forests; GEORISK data (BISBY); data on catchment areas; historical data. In Austria only the Austrian Service for Torrent and Avalanche Control (WLV) generates hazard maps, called Gefahrenzonenkarte or hazard zone maps for oods, avalanches and debris ows within the Hazard zonation plan (Gefahrenzonenplan). This is regulated by law (Forest Act BGBL. 440/1975). The implementation

is regulated by a decree (Verordnung des Bundesministeriums fr Land- und Forstwirtschaft, 1976, BGBl. Nr. 436/1976). The scale usually ranges between 1:2,000 and 1:5,000, it must not be smaller than 1:50,000. The map gives information about the determined effects in the relevant area of catchment areas (torrent buffer areas) in red and yellow hazard zones. The design event is determined by a return period of 150 years. In the red hazard zone, infrastructures cannot be maintained or can only be maintained with a very high effort due to the high intensity or a high recurrence of avalanches or torrential events. The yellow hazard zone includes all other areas affected by avalanches and torrents. The constant use of these areas by infrastructures is affected due to these hazards. The hazard zone map also delineates blue areas (for the implementation of technical or forestal measures as well as protective measures), as well as brown and violet reference areas. The brown reference areas are areas presumably affected by other hazards than torrents or avalanches, like rock fall or landslides. The violet reference areas are areas, where soil and terrain have to be protected in order to keep up their protective function. In Switzerland, the Federal Ofce for the Environment FOEN (Bundesamt fr Umwelt, BAFU) is responsible for creating guidelines concerning protection against natural hazards (oods, mass movements, snow avalanches). The concepts are similar for these processes to reach a certain level of protection. Protection against natural hazards takes place on the principle of integral risk management, taking into account: Prevention of an event Conict management during an event Regeneration an event. and reconstruction after

The Swiss regulations are described in more detail by Raetzo in this issue [31]. In some regions of Italy the hazard is assessed using the Swiss method ([30] Raetzo 2002). This method is similar to the method planned by the Italian legislative body for hydrogeological risk assessment. Appropriate changes have been introduced in order to standardize these aspects and contextualize the method for territorial jurisdiction ([21] Kranitz & Bensi 2009). Four classes of hazards are distinguished, ranging from very high (P4 molto elevata), high (P3 elevata), medium (P2 media), to moderate (P1 moderata). The French hazard map, PPR, Plan de prevention des risques, is made by the local authorities (mayors), but with support by national agencies like CEMAGREF or agencies of the departments. It was introduced in 1995. Made by the municipalities at a scale of 1:10,000 -1:25,000, the plans need to be authorized by the prefects in collaboration with the local authorities and the civil society, such as insurance companies. The PPR gives information about the identication of danger zones; 3 classes of risk with specic rules for land use regulations and urbanism can be represented. The method is a qualitative method based on the expert judgment of the scientist. There are PPRs for oods, mass movements, avalanches and wood res. Nonobservance of the PPR has legal consequences. In Spain the Geological Institute of Catalonia (IGC) is responsible to study and assess geological hazards, including avalanches, to propose measures to develop hazard forecast, prevention and mitigation and to give support to other agencies competent in land and urban planning, and in emergency management ([28] Oller et al. 2010). Therefore, the IGC is charged with making ofcial hazard maps with such nality. These maps comply with the Catalan

Seite 37

Key-note papers

Seite 38

Urban Law (1/2005), which indicates that in those places where a risk exists, building is not allowed. For hazard mapping, the work is done on two scales: land planning scale (1:25,000), and urban scale (1:5,000 or more detailed). These scales imply different approaches and methods to obtain hazard parameters. The maps are generated in the framework of a mapping plan or as the nal product of a specic hazard report. The Australian AGS guidelines ([1] AGS, 2000, [2]- [6] AGS 2007a-e) provide for a hazard zonation at a local (1:5,000 -1:25,000) and a site specic (>1:5,000, typically 1:5,000 -1:1,000) scale with 5 hazard descriptors: very high high moderate low very low. The state of Washington (USA) generated

hazard zonation maps at a scale of 1:12,000. The hazard assessment included evaluating a landslide frequency rate (LFR) and a landslide area rate for delivery (LAR). The LFR is obtained by taking the number of delivering landslides per landform, divided by the total area of that landform, and normalized to the period of study. The LAR is the area of delivering landslides normalized to the period of study and the area of each landform. The resulting values are multiplied by one million for easier interpretation. In California soil-slip susceptibility maps were produced at a scale of 1:24,000 delineating the susceptibility in 3 classes: low, moderate and high. They give information about the relative susceptibility of hill slopes to the initiation sites of
Prozesshinweiskarte (Karte der Phnomene)

rainfall-triggered soil-slip debris ows ([26] Morton et al., 2003). The state of Utah prepared a landslide susceptibility map for the whole state at a scale of 1:500,000 for deep seated landslides, based on existing landslides and slope angle thresholds for different geologic units. The susceptibility is delineated in 4 classes: high moderate low very low ([10] Giraud & Shaw, 2007). Conclusion and recommendations Guzzetti ([11], 2005) discusses hazard assessment in his thesis: Despite the time [since the denition of landslide hazard given by Varnes and the IAEG Commission on Landslides and other Mass Movements ([39], 1984)] and the extensive list of published papers most of which, in spite of the title or the intention of the authors, deal with landslide susceptibility and not with landslide hazard, landslide hazard assessment at the basin scale is sparse. And further: This is largely due to difculties associated with the quantitative determination of landslide hazard. In carrying out the literature survey, this unfortunately proved to be true and contributed to the confusion existing with denitions ([29] Posch-Trzmller 2010). The differences call rst for a national

through time and represent the main resource for susceptibility/hazard assessment. The evidence identied in the eld are the facts dealing with natural hazards. Inventories are the essential base for accurate hazard/risk assessment and have therefore to be established by authorities. The variability of phenomena of mass makes regulations concerning movements

methods of hazard assessment difcult. Guidelines regarding hazard assessment should declare the minimal requirements taking into account the nal objective and the scale of product.

Informationsebene

Ereigniskataster

Ereigniskarte

Thematische Inventarkarte Standortparameter und -verhltnisse

Gefahrenhinweiskarte qualitativ / semiquantitativ Dispostionskarte Grunddispositionskarte


Erweiterte Dispositionskarte

Interpretationsebene / Bewertungsebene

harmonization and second for international comparable methods (minimal requirements). To assess landslide hazards, the geological, morphological, hydrogeological and hydrological conditions must be known and analysed: The differences regarding acquisition of information and assessment of the susceptibility/ hazard of slopes to landslides and rock fall shown in the chapter above call for a harmonization of the different methods (e.g. parameters, assessment minimal requirements). Hazard

Gefahrenpotentialkarte (Karte der potentiellen Wirkungsbereiche)

quantitativ

Gefahrenkarte

Risikokarte

Fig. 1: Workow of hazard mapping. ([18] Kociu et al. 2010) Abb. 1: Flussdiagramm zum Prozess Gefahrenkartierung. ([18] Kociu et al. 2010)

needs information about possible scenarios. Landslide inventories sustain landslide knowledge

Seite 39

Seite 40

Countries Austria GBA Inventory Basic information when what why who reported when Landslide conditions geometry slope position approx. original slope x site description depth to bedrock depth to failure plane slope aspect slope Geology in general Geology, specied x x x x x x lithology/ stratigraphy x bedding attitude weathering geotechnical properties geologic/ tectonic unit x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x activity x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x where x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x N K MM S By CH SLO IT F AUS O W U D CH SLO IT F AUS USA

Key-note papers

geotechnical parameters rock mass structure joints/ joint spacing discontinuities structural contributions Land cover/ use Hydrogeology Relationship to rainfall Classication of mass movements Classication type rate of movement material water content Causes, Trigger Precursory signs Silent witnesses Damage "Hazard" to infrastructure Remedial measures Costs of measures and investigation Methods used Degree of precision info/ reliability Reports etc.
Tab. 1: Comparison of information collected for different inventories Tab. 1: Vergleich der Informationen in Ereigniskatastern

x x x x x x x x x x x x x x x x x x x x x x x x x x

x x x x x x x

x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x

Seite 41

Seite 42

Switzerland rock fall


E=energy

low intensity E<30kJ 30kJ<E<300kJ E>300kJ

moderate intensity

high intensity

continous slides dv, D, T dv, D, T M>2m; h>1m dv, D, T

v2cm/year

2cm/year<v<10cm/year

v>10cm/year, (or 1m dislocation per event)

v=velocity dv=variation of v, acceleration D=differential movement T=thickness M=thickness of potentially displaced mass h=thickness of accumulation of shallow slide or ow

spontaneous slides

M<0.5m

0.5m<M<2m; h<1m

ow (earth ow) v2cm/year dv, D, T dolines potentially existing or soluble rocks (gypsum, etc.) presence of dolines veried intensity 3 intensity 4 intensity 1 intensity 2 dv, D, T 2cm/year<v<10cm/year

M<0.5m

0.5m<M<2m; h<1m

M>2m; h>1m

creep (+Permafrost)

v>10cm/year, dislocation per event >1m dv, D, T

Key-note papers

subsidence Italy

dolines and danger of collapsing intensity 6 intensity 9

Raetzo & Loup ([32], 2009)

rock fall

SG=1 and v=1 (meaning: block diam.<0.5m extremely slow, <16mm/ year) or v=2 (meaning: very slow (16mm/ year to rapid [1.8m/hour]) SG=3 (d>2m), v=1 (<16mm/ year) or: SG=1 (<0.5m), v=3 (very high to extremely rapid: 3m/min to 5m/sec.)

SG=2 (meaning: block diameter 0.5-2m), v=1 (meaning extremely slow, <16mm/ year), or: SG=1 (<0.5m) and v=2 (16mm/year to 1,8m/hour)

SG=2 (d=0.52m), v=2 (16mm/year to 1,8m/hour)

SG=2 (d=0.52m), v=3 (3m/ min. to 5m/ sec.)

SG=3 (d>2m), v=3 (3m/min. to 5m/sec.)

SG=geometry factor, v=velocity factor

slide

SG=2 (depth: 2-15m), v=1 SG=1 (<0.5m), (<16mm/year), v=1 (16mm/ or: SG=1 year to 1.8m/ (<2m), v=2 hour) (16mm/year to 1,8m/hour)

SG=3 (depth>15m), v=1 (<16mm/ year) or: SG=1 (depth<2m), v=3 (very high to extremely rapid: 3m/min to 5m/sec.)

SG=2 (depth: 2-15m), v=2 (16mm/year to 1,8m/hour)

SG=2 (depth: 2-15m), v=3 (3m/min. to 5m/sec.)

SG=3 (depth>15m), v=3 (3m/min. to 5m/sec.)

Kranitz & Bensi ([21], 2009)

Australia rock fall, rock avalanche slide ow

Whether landslide intensity is required for hazard zoning is to be determined on a case-by-case basis. For rock fall hazard zoning it is likely to be required. The landslide intensity is assessed as a spatial distribution of: The kinetic energy or or the total displacement or the differential displacement, or The velocity of sliding coupled with slide volume or the total displacement or the differential displacement, or The peak discharge per unit width (m3/m/sec., e.g. debris ows) For basic and intermediate level assessments of intensity only the velocity and volume might be assessed, but for advanced assessments of rock fall or debris ow hazard the energy should be assessed.
Tab. 2: Comparison of the intensity-assessment in Switzerland, Italy and Australia Tab. 2: Vergleich Intensitt Gefahrenabschtzung in der Schweiz, in Italien und in Australien source: AGS ([2], 2007a)

Seite 43

Key-note papers

Seite 44

Anschrift der Verfasser / Authors addresses: USA: Washington statistical 1:12,000 Richard Bk Amt der Krntner Landesregierung Abt. 15 Umwelt Unterabteilung Geologie und Bodenschutz statistic and empirical Australia: AGS 1:5,0001:25,000 Flatschacher Strae 70, A 9020 Klagenfurt Karl Mayer Bayerisches Landesamt fr Umwelt Abt. 6 Wasserbau, Hochwasserschutz, qualitative 1:10,000 (urban), -1:25,000 (rural) France: PPR Gewsserschutz 2 (3) Ref. 61 Hochwasserschutz und alpine Naturgefahren Lazarettstrae 67 empirical, probabilistic national is possible, regional D 80636 Mnchen Gerlinde Posch-Trzmller 5 Geologische Bundesanstalt
Tab. 3: Vergleich von verschiedenen Gefahrenkarten, Mastben und Legenden (Grad der Gefahren)

Literatur / References:
[1] AGS - AUSTRALIAN GEOMECHANICS SOCIETY, SUB-COMMITTEE ON LANDSLIDE RISK MANAGEMENT (2000): Landslide Risk Management Concepts and Guidelines. Australian Geomechanics, Vol 35, No 1, March 2000. [2] AGS (2007a). Guideline for Landslide Susceptibility, Hazard and Risk Zoning for Land Use Planning. Australian Geomechanics Society. Australian Geomechanics, Vol 42, No 1, March 2007. [3] AGS (2007b). Commentary on Guideline for Landslide Susceptibility, Hazard and Risk Zoning for Land Use Planning. Australian Geomechanics Society. Australian Geomechanics, Vol 42, No 1, March 2007. [4] AGS (2007c). Practice Note Guidelines for Landslide Risk Management. Australian Geomechanics Society. Australian Geomechanics, Vol 42, No 1, March 2007. [5] AGS (2007d). Commentary on Practice Note Guidelines for Landslide Risk Management 2007. Australian Geomechanics Society. Australian Geomechanics, Vol 42, No 1, March 2007. [6] AGS (2007e). The Australian GeoGuides for slope management and maintenance. Australian Geomechanics Society. Australian Geomechanics, Vol 42, No 1, March 2007. [7] BK, EBERHART, GOLDSCHMIDT, KOCIU, LETOUZE-ZEZULA & LIPIARSKI: Ereigniskataster und Karte der Phnomene als Werkzeug zur Darstellung geogener Naturgefahren (Massenbewegungen), Arb. Tagg. Geol. B.-A., Gmnd 2005. [8] BWG - BUNDESAMT FR WASSER UND GEOLOGIE: Naturgefahren, Symbolbaukasten zur Kartierung der Phnomene, 2002 [9] CRUDEN D.M. UND VARNES D.J.: Landslide types and processes. In: A. Keith Turner & Robert L. Schuster (eds): Landslide investigation and mitigation: 36-75. Transportation Research Board, special report 247. Washington: National Academy Press, 1996. [10] GIRAUD, R.E., SHAW, L.M.: Landslide Susceptibility Map of Utah. MAP 228DM, Utah Geological Survey, Utah Department of Natural Resources, Salt Lake City 2007. [11] GUZZETTI, F.: Landslide hazard and risk assessment. Diss. Math.-Naturwiss. Fak. Univ. Bonn, Bonn 2005. [12] HEINIMANN, H.R., VISSER, R.J.M., STAMPFER, K.: Harvester-cable yarder system evaluation on slopes: A Central European study in thinning operations. In: Schiess, P. and Krogstad, F. (Eds.): COFE Proceedings Harvesting logistic: from woods to markets, 41-46. Portland, OR, 20-23 July, 1998. [13] ISPRA INSTITUTE FOR ENVIRONMENTAL PROTECTION AND RESEARCH: Landslides in Italy. Special report 2008. 83/2008, Rome 2008. [14] KIENHOLZ, H., KRUMMENACHER, B.: Empfehlungen Symbolbaukasten zur Kartierung der Phnomene Ausgabe 1995, Mitteilungen des Bundesamtes fr Wasser und Geologie Nr. 6, 41 S., Reihe Vollzug Umwelt VU-7502-D, Bern 1995.

Countries/ projects

Italy: Guzzetti

Italy: Friuli, Veneto

quantitative, statistical (incl. eld investigation)

Fachabteilung Rohstoffgeologie Neulinggasse 38, A-1030 Wien Andreas von Poschinger Bayerisches Landesamt fr Umwelt Abt. 10 Geologischer Dienst Ref.106 Ingenieurgeologie, Georisiken, Lazarettstrae 67, D 80636 Mnchen Hugo Raetzo Federal Ofce for the Environment FOEN Bundesamt fr Umwelt BAFU CH - 3003 Bern, Schweiz

30 years 100 years 300 years >300 years

detail

Basic data: inventory

Return periods considered for land use (probability)

Basic data: susceptibility map

Legend: Levels of hazard

Comparison of hazard maps

Method (assessment, modelling)

Scale

Tab. 3: Comparison of different hazard maps, their scales and legends (levels of hazard)

30 years 100 years 300 years (Residual risk zones for RP>300y)

quantitative, statistic, qualitative (incl. eld investigation)

1:2,000- 1:10,000

Switzerland: FOEN/BAFU

eventually

2 (for torrent and debris ow), indication for landslides and rock fall

quantitative, statistic, empirical

150 years

1:2,0001:5,000

Austria: WLV

Seite 45

Seite 46

[15] KLINGSEISEN, B., LEOPOLD, PH.: Landslide Hazard Mapping in Austria.-GIM International 20 (12): 41-43, 2006. [16] KLINGSEISEN, B., LEOPOLD, PH., TSCHACH, M.: Mapping Landslide Hazards in Austria: GIS Aids Regional Planning in NonAlpine Regions. ArcNews 28 (3): 16, 2006. [17] KOCIU, A., LETOUZE-ZEZULA, G., TILCH, N., GRSEL, K.: Georisiko-Potenzial Krnten; Entwicklung einer GIS-basierten Gefahrenhinweiskarte betreffend Massenbewegungen auf Grundlage einer digitalen geologischen Karte (1:50,000) und eines georeferenzierten Ereigniskatasters. Endbericht, Gefhrdungskarte, Ausweisung von Bereichen unterschiedlicher Suszeptibilitt fr verschiedene Typengruppen der Massenbewegung. Bund/Bundeslnderkooperation KC-29, Bibl. Geol. B.-A., Wiss. Archiv, Wien, 2006 [18] KOCIU, A., TILCH N., SCHWARZ L,. HABERLER A., MELZNER S.: GEORIOS - Jahresbericht 2009; Geol.B.-A. Wien 2010. [19] KOLMER, CH.: Geogenes Baugrundrisiko bersterreich. Vortrag im Rahmen des Landesgeologentages 2009, 26.2.2009, St. Plten, 2009. [20] KOMAC, M.; RIBICIC, M.: Landslide Susceptibility Map of Slovenia 1:250,000. Geological Survey of Slovenia, Ljubljana 2008. [21] KRANITZ, F., BENSI, S.: The BUWAL method. In: Posch-Trzmller, G. (Ed.): Second Scientic Report to the INTERREG IV A project MASSMOVE - Minimal standards for compilation of danger maps like landslides and rock fall as a tool for disaster prevention. Attachment 4 to the second progress report, Geological Survey of Austria, Wien, 2009. [22] Malet, J.-P.; Thiery, Y.; Maquaire, O.; Sterlacchini, S.; van Beek, L.P.H.; van Asch, Th.W.J.; Puissant, A.; Remaitre, A.: Landslide risk zoning: What can be expected from model simulations? JRC Expert Meetings on Guidelines for Mapping Areas at Risk of Landslides in Europe 23-24 October 2007, JRC, Ispra EU, 2007. [23] MAYER, K.: Manahme 3.2a Schaffung geologischer und hydrologischer Informationsgrundlagen. Vorhaben Gefahrenhinweiskarte Oberallgu. Bayerisches Landesamt fr Umwelt, Mnchen 2007. [24] MAZENGARB, C.: The Tasmanian Landslide Hazard Map Series: Methodology. Tasmanian Geological Survey Record 2005/04, Mineral Resources Tasmania, 2005. [25] MIDDELMANN, M. H. (ED.): Natural Hazards in Australia: Identifying Risk Analysis Requirements. Geoscience Australia, Canberra 2007. [26] MORTON, D.M., ALVAREZ, R.M., CAMPBELL, R.H.: Preliminary soil-slip susceptibility maps, southwestern California. USGS Open-File Report OF 03-17, Riverside, 2003.

[27] NSSING, L.: Gefahrenzonenplanung in Sdtirol. Vortrag im Landesgeologentages 2009, 26.2.2009, St. Plten 2009.

Rahmen

des

[28] OLLER, P., GONZALEZ, M., PINYOL, J., MARTINEZ, P.: Hazard mapping in Catalonia. Vortrag Workshop AdaptAlp, 17.3.2010, Bozen 2010. [29] POSCH-TRZMLLER, G.: AdaptAlp WP 5.1 Hazard Mapping - Geological Hazards. Literature Survey regarding methods of hazard mapping and evaluation of danger by landslides and rock fall. Final Report, Geologische Bundesanstalt, Wien, 2010 [30] RAETZO, H.: Hazard assessment in Switzerland codes of practice for mass movements, International Association of Engineering Geology IAEG Bulletin, 2002. [31] REATZO, H. & LOUP, B.: Geological hazard assessment in Switzerland (this issue) [32] RAETZO, H. & LOUP, B. ET AL.; BAFU: Schutz vor Massenbewegungen. Technische Richtlinie als Vollzugshilfe. Entwurf 9. Sept. 2009. [33] REEVES, H.: Geohazards: The UK perspective. Vortrag Workshop AdaptAlp, 17.3.2010, Bozen 2010. [34] RUDOLF-MIKLAU F. & SCHMIDT F.: Implementation, application and enforcement of hazard zone maps for torrent and avalanches control in Austria, Forstliche Schriftenreihe, Universitt fr Bodenkultur Wien, Bd. 18, p. 83-107, 2004. [35] RUFF, M.: GIS-gesttzte Risikonanalyse fr Rutschungen und Felsstrze in den Ostalpen (Vorarlberg, sterreich). Georisikokarte Vorarlberg. Diss. Univ. Karlsruhe, 2005. [36] SCHWEIGL, J.; HERVAS, J.: Landslide Mapping in Austria. JRC Scientic and Technical Report EUR 23785 EN, Ofce for Ofcial Publications of the European Communities, 61 pp. ISBN 978-92-79-11776-3, Luxembourg, 2009. [37] SGD, PERSONENKREIS GEOGEFAHREN: Geogene Naturgefahren in Deutschland- Empfehlungen der Staatlichen Geologischen Dienste (SGD) zur Erstellung von Gefahrenhinweiskarten., 2007. [38] TILCH, N.: Datenmanagementsystem GEORIOS (Geogene Risiken sterreich). Vortrag im Rahmen des Landesgeologentages 2009, 26.2.2009, St. Plten 2009. [39] VARNES, D.J. AND IAEG COMMISSION ON LANDSLIDES AND OTHER MASS-MOVEMENTS: Landslide hazard zonation: a review of principles and practice. The UNESCO Press, Paris, 1984.

Seite 47

Key-note papers

Seite 48

Zusammenfassung: In den Bergregionen treten an Steilhngen verschiedene Arten von Massenbewegungen auf, die Wasser und Sedimente mit sich fhren: Muren, Bergsturz und Steinschlag. Das Ziel dieser Abhandlung ist es, einen kurzen berblick ber die vergangenen Analysen der Gefahren von Hangmassenbewegungen zu geben. Obwohl der Schwerpunkt auf Bergstrzen liegt, knnen die prsentierten Anstze auch zur Gefahrenbeurteilung von Muren und Steinschlag verwendet werden. Insbesondere Bergstrze und Muren sind sehr hug miteinander verochten. Im Folgenden wird Bergsturz im weiteren Sinn als ein Begriff verwendet, der nicht nur auf einen Erdrutsch zu beziehen ist, sondern auch auf andere Hangmassenbewegungen. Schlsselwrter: Bergstrze, Muren, Felssturz, numerische Anstze, Bergsturzgefahrenanalyse
movements on slopes, including rock-fall, topples 1. The Early Ages MATEJA JEMEC, MARKO KOMAC The rst extensive papers on the use of spatial information in a digital context for landslide susceptibility mapping date back to the late seventies and early eighties of the last century. Among the pioneers in this eld were Carrara et al. (1977) in Italy and Brabb et al. (1978) in California. Nowadays, practically all research on landslide susceptibility and hazard mapping makes use of digital tools for handling spatial data such as GIS, GPS and Remote Sensing. These tools also have dened, to a large extent, the type of analysis that can be carried out. It can be stated that to a certain degree the capability of GIS tools and the accuracy of the in-situ and remote sensing data have determined the current state of the art in landslide hazard and risk assessment. Many publications about landslides and some worldwide landslide research problems can be found in the literature of Einstein (1988), Fell (1994), Dai et al. (2002) and Glade et al. (2005). 2. Terminology The term landslide was dened by Varnes and IAEG (1984) as almost all varieties of mass and debris ow, that involve little or no true sliding. Cruden (1991) moderated the accepted denition as the movement of a mass of rock, earth or debris down a slope. Later different working groups were established to support a specic level of standardisation in elds related to landslides (UNESCO, IUGS, ISSMGE, ISRM and IAEG) and created the JTC (Joint Technical Committee on Landslides and Engineered Slopes), which continues to work for the standardisation and promotion of research on landslides among the different disciplines. A large set of denitions was later presented by ISSMGE TC32 (Technical Committee on Risk Assessment and Management, 2004) where international terms recognized for hazard, vulnerability, risk and disaster can also be found. Since these denitions were published, many approaches have been implemented (Einstein, 1988; Fell, 1994; Soeters and van Westen, 1996; Wu et al., 1996; Cruden and Fell, 1997; van Westen et al., 2003; Lee and Jones, 2004; Glade et al., 2005) allowing one to conclude that nowadays denitions regarding landslides risk assessment are generally accepted. The latest information of guidelines for landslide susceptibility, hazard and risk zoning are published by JTC-1 (2008) and van Westen et al. (2008).

An Overview of Approaches for Hazard Assessment of Slope Mass Movements Ein berblick ber die Anstze zur Gefahrenbeurteilung von Massenbewegung
Summary: In mountainous areas, various types of mass movements occur on steep slopes involving water and sediment: debris ows, landslides and rockfalls. The aim of this paper is to gather a short overview of the past analyses that dealt with the hazard assessment of slope mass movements. Although the main focus is on landslides, the approaches presented can be used to assess debris ows and rockfall hazards. In particular, landslides and debris ow are very often interlaced between each other. In the following text, the term landslide will be used as a term that might not always be strictly connected to only landslides but also to other slope mass movements. In a way it has a broader meaning. Keywords: landslides, debris-ows, rockfall, numerical approaches, landslide hazard assessment

Seite 49

Key-note papers

Seite 50

Data layer and types 1. Landslide occurrence


Landslides Terrain mapping units Geomorphological units Digital elevation model (DEM) Slope map Aspect map Slope length Slope shape Internal relief Drainage density Lithologies Soils and material sequences Structural geological map Vertical movements Land use map Drainage Catchment areas Water table

Accompanying data in tables


Type, activity, depth, dimensions, etc Units description Geomorphological description Altitude classes Slope angle classes Slope direction classes Slope length classes Concavity/convexity Altitude/area classes Longitude/area classes Lithology, rock strength, weathering process Soils types, materials, depth, grain size, distribution, bulk density Fault type, length, dip, dip direction, fold axis Vertical movements, velocities Land use type, tree density root depth Type, order and length Order, size Depth of water table in time Precipitation in time Earthquakes database and maximum sesismic acceleration Number, sex, age, etc. Roads and railroad types, facilities types Types of lifeline network and capacity of fascilities Type of structure and occupation Industry production and type Number and type of health, educational, cultural and sport facilities Type of touristy facilities Area without natural resources combined

Used methods for data collecting


Fieldwork, orthophoto, satellite images In-situ survey (eldwork), satellite images Ortophoto, eldwork, high resolution DEM SRTM DEM data, topographic map With GIS form DEM With GIS form DEM With GIS form DEM With GIS form DEM With GIS form DEM With GIS form DEM Fieldwork and laboratory tests, archives, orthophoto Modelling form lithological map, geomorphological map and slope map, eldwork and laboratory analysis Fieldwork, satellite images, orthofoto Geodetic data, satellite data Satellite images, orthofoto, eldwork Orthophoto, topographic map Orthophoto, topographic map Hydraulic stations Meteorological stations and modelling Seismic data, engineering geological data and modelling Statistics information Atlas, topographic map, local information Atlas, topographic map, local information Topographic map, Housing information Atlas, topographic map, local information Atlas, topographic map, local information Atlas, topographic map, local information Atlas, topographic map, local information

Landslide related data can be grouped into four main sets, Table 1 (Soeters and van Westen, 1996). have Debris several ows are processes and that sub-categories different

ow are very often interlaced between each other (Fig.1). In many cases, heavy precipitation is recognised as the main cause, and thresholds under different climatic conditions have been empirically evaluated (Caine, 1980; Canuti et al., 1985; Fleming et al., 1989; Mainali and Rajaratnam, 1994; Anderson, 1995; Cruden and Varnes, 1996; Finlay et al., 1997; Crosta, 1998; Crozier, 1999; Dai et al., 1999; Glade, 2000; Alcantara-Ayala, 2004; Fiorillo and Wilson, 2004; Lan et al., 2004; Malet et al., 2005; Wen and Aydin, 2005). Landslides may mobilise to form debris ows by three processes: (a) widespread Coulomb failure within a sloping soil, rock, or sediment mass, (b) partial or complete liquefaction of the mass by high pore-uid pressure, and (c) conversion of landslide translational energy to internal vibrational energy (Iverson et al., 1997).

2. Environmental (preparatory) factors

characteristics. Debris ows are gravity-induced mass movements, intermediate between land sliding and water ooding, with mechanical characteristics different from either of these processes (Johnson, 1970). According to Varnes (1978), debris ow is a form of rapid mass movement of rocks and soils in a body of granular solid, water, and air, analogous to the movement of liquids. In the landslide classication of Cruden and Varnes (1996), debris ows are ow-like landslides with less than 80% of sand and ner particles. Velocities vary between very rapid and extremely rapid with typical velocities of 3 m/min and 5 m/sec, respectively. Landslides and debris

3. Triggering factors
Rainfall and maximum probabilities Earthquakes and seismic acceleration

4. Elements at risk
Population Transportation system and facilities Lifeline utility system Building Industry Services facilities Tourism facilities Natural resources

Fig. 1: Classication of slope mass movements as a ratio of solid fraction and material type. Modied after Coussot and Meunier (1996). Abb. 1: Klassikation von Massenbewegungen als Verhltnis von Geschiebefraktion und Materialart. Modiziert nach Coussot und Meunier (1996).

Tab. 1: Summary of data needed for landslide hazard and risk assessment. Adapted from Soeters and van Westen (1996). Tab. 1: Zusammenfassung der Daten fr Erdrutsch-Gefhrdungs- und Risikoanalyse. Adaptiert von Soeters und van Westen (1996).

Seite 51

Key-note papers

Seite 52

Rockfall is one of the most common mass movement processes in mountain regions and is dened as the free falling, bouncing or rolling of individual or a few rocks and boulders, with volumes involved generally being < 5 m3 (Berger et al., 2002). Numerous studies exist concerning various aspects of rockfall, such as the dynamic behaviour (Ritchie, 1963; Erismann, 1986; Azzoni et al., 1995), boulder reaction during ground contact (Bozzolo et al., 1986; Hungr and Evans, 1988; Evans and Hungr, 1993), or runout distances of falling rocks (Kirkby and Statham, 1975; Statham and Francis, 1986; Okura et al., 2000). Much research was also done on the possible triggers of rockfall, such as freeze-thaw cycles (Gardner, 1983; Matsuoka and Sakai, 1999; Matsuoka, 2006), changes in the rock-moisture level (Sass, 2005), the thawing of permafrost (Gruber et al., 2004), the increase of mean annual temperatures (Davies et al., 2001), tectonic folding (Coe and Harp, 2007) or the occurrence of earthquakes (Harp and Wilson, 1995; Marzorati et al., 2002). In addition, several studies exist on the long-term accretion rates of rockfall (Luckman and Fiske, 1995; McCarroll et al., 1998). Furthermore, since the late 1980s, the eld of numeric modelling has become a major topic in the eld of rockfall research (Zinggerle, 1989; Guzzetti et al., 2002; Dorren et al., 2006; Stoffel et al., 2006). 3. Numerical approaches to landslide hazard assessment According to Van Westen (1993), the landslide hazard assessment methods have been divided into four groups of analysis. Weve added an additional group Articial Neural Networks. The selection of one method over another depends on several factors (the data costs and availability, the scale, the output requirements, the geological and geomorphological conditions, the tectonogenetic

and morphogenetic behaviour of the landslides, and computing capabilities of software and hardware tools). Firstly, inventory analysis, which are based on the analysis of the spatial and temporal distribution of landslide attributes and such inventories are the basis of most susceptibility mapping techniques. On detailed landslide inventory maps, the basic information for evaluating and reducing landslide hazards on a regional or local level may be provided. Such maps include the state of activity, certainty of identication, dominant type of slope movement, primary direction, and estimated thickness of material involved in landslides, and the dates of known activity for each landslide (Wieczorek, 1984). Secondly, the popular heuristic analysis (Castellanos and van Westen, 2003; R2 Resource Consultants, 2005; Ruff and Czurda, 2007; Firdaini, 2008) based on expert criteria with different assessment methods. The landslide inventory map is accompanied with preparatory factors to be the main input for determining landslide hazard zoning. Experts then dene the weighting value for each factor. Many researchers utilize statistical analysis (Neuland, 1976; Carrara, 1983; Pike, 1988; Carrarra et al., 1991; van Westen, 1993; Chung & Fabbri, 1999; Gorsevski et al., 2000; Dhakal et al., 2000; Zhou et al., 2003; Saha et al., 2005; Guinau et al., 2007; Komac and Ribii, 2008; Magliulo et al., 2008; Miller and Burnett, 2008; Pozzoni et al., 2009; Komac et al., 2010), where several parameter maps are surveyed to apply bivariate and multivariate analysis. The key of this method is the landslide inventory map when the past landslide occurrences are needed to forecast future landslide areas. The next approach is deterministic analysis (van Westen, 1994; Terlien et al., 1995;

van Westen and Terlien, 1996; Soeters and Westen, 1996; van Asch et al., 1999; Zaitchik et al., 2003; Mazengarb, 2004; Schmidt and Dikau, 2004; Mayer et al., 2010), which is based on hydrological and slope instability models to evaluate the safety factor. Montgomery et al. (1994, 1998 and 2000) have attributed a great importance to precipitation and many other investigations have also been carried out about the relationship between rainfall and landslides (Crozier, 1999; Lida, 1999; Dai and Lee, 2001; Guzzetti et al., 2007). For rainfall induced failures, these models couple shallow subsurface ow caused by rainfalls of various return periods, predicted soil thickness and soil mantle landslides. Numerous studies have used rainfall characteristics, such as duration, intensity, maximum and antecedent rainfall during a particular period, to identify the threshold value for landslide initiation. Many authors (Caine, 1980; Caine and Mool, 1982; Brabb, 1984; Cannon and Ellen, 1985; Jakob and Weatherly, 2003) applied the rainfall intensity duration equation to estimate the threshold. With regard to specic rainfall characteristics, Wieczorek and Sarmiento (1983) used total rainfall duration before specic rainfall intensity occurs; Govi et al. (1985) applied total rainfall during a specic period after rainfall starts; and Crozier (1986) utilized the ratio of total rainfall to antecedent rainfall. Guzzetti et al. (2004) identied the local rainfall threshold on the basis of local rainfall and landslide record and concluded that landslide activity in Northern Italy initiates 8-10 hours after the beginning of a storm. However, many other investigations have been published about the relationship between rainfall and landslides and attribute a large impact to precipitation for the time duration of landslides (Carrara, 1991; Mongomery et al., 1994, 1998; Terlien et al., 1995; Crozier, 1999; Laprade et al., 2000; Alcantara-Ayala, 2004; Coe et al., 2004; Fiorillo and Wilson, 2004; Lan et al.,

2004; Wen and Aydin, 2005; Zezere et al., 2005; Giannecchini, 2006; Jakob et at., 2006). While some of them deal with specic cases, others are more concerned with the statistical relationship for creating correlations models and even produce forecasting models based on rainfall threshold values. One of the relatively new methods applied to landslide hazard and susceptibility assessment are articial neural network (ANN) tools. ANN is a useful approach for problems such as regression and classication, since it has the capability of analyzing complex data at varied scales such as continuous, categorical and binary data. The concept of ANN is based on learning form data with known characteristics to derive a set of weighting parameters which are used subsequently to recognize the unseen data (Horton, 1945). Lee et al. (2003b) developed landslide susceptibility analysis techniques using a multilayered perception (MLP) network. The results were veried by ranking the susceptibility index in classes of equal area and showed satisfactory agreement between the susceptibility map and the landslide location data. Lee et al. (2003a) obtained landslide susceptibility by using neural network models and compared neural models with probabilistic and statistical ones. They also show a combination of ANN for determination of weights used spatial probabilities to create a landslide susceptibility index map (Lee et al., 2004). Rainfall and earthquake scenarios as triggering factors for landslides have been used in hazard assessment with ANNs (Lee and Evangelista, 2006; Wang and Sassa, 2006). Several studies recognize ANN as a promising tool for these applications and most of them use a Multi layer Perceptron (MLP) network and a back propagation algorithm for training the network (Rumelhart et al., 1986; Arora et al., 2004; Ercanoglu, 2005; Ermini et al., 2005;

Seite 53

Key-note papers

Seite 54

Numerical approach Inventory analysis

Basic description of approach Analysis of the spatial and temporal distribution of landslide attributes Based on expert criteria with different assessment methods

References Wieczorek (1984) Castellanos and van Westen (2003); R2 Resource Consultants (2005); Ruff and Czurda (2007); Firdaini (2008) Neuland (1976); Carrara (1983); Pike (1988); Carrarra et al. (1991); van Westen (1993); Chung and Fabbri (1999); Gorsevski et al. (2000); Dhakal et al. (2000); Zhou et al. (2003); Saha et al. (2005); Guinau et al. (2007); Komac and Ribii (2008); Magliulo et al. (2008); Miller and Burnett (2008); Pozzoni et al. (2009); Komac et al. (2010) van Westen (1994); Terlien et al. (1995); van Westen and Terlien (1996); Soeters and Westen (1996); van Asch et al. (1999); Zaitchik et al. (2003); Mazengarb (2004); Schmidt and Dikau (2004); Mayer et al. (2010) Caine (1980); Caine and Mool (1982); Wieczorek and Sarmiento (1983); Brabb (1984); Cannon and Ellen (1985); Govi et al. (1985); Crozier (1986); Carrara (1991); Terlien et al. (1995); Montgomery et al. (1994, 1998 and 2000); Crozier (1999); Lida (1999); Laprade et al. (2000); Dai and Lee (2001); Jakob and Weatherly (2003); Alcantara-Ayala (2004); Coe et al. (2004); Fiorillo and Wilson (2004); Guzzetti et al. (2004); Lan et al. (2004); Zezere et al. (2005); Wen and Aydin (2005); Giannecchini (2006); Jakob et al. (2006); Guzzetti et al. (2007)

Gomez and Kavzoglu, 2005; Wang et al., 2005; Pradhan and Lee, 2007, 2009a, 2009b, 2009c; Pradhan et al., 2009; Youssef et al., 2009). Ermini et al. (2005) and Catani et al. (2005) used unique conditions units for the terrain unit denition in ANNs analysis. More critical analyses compare ANN techniques with other methods such as logistic regression, fuzzy weighing and other statistical methods (Ercanoglu and Gokceoglu, 2002; Lu, 2003; Neaupane and Achet, 2004; Miska and Jan, 2005; Yesilnacar and Topal, 2005; Kanungo et al., 2006; Lee, 2007). In the neural network method, Nefeslioglu et al. (2008) showed that ANNs give a more optimistic evaluation of landslide susceptibility than logistic regression analysis. Melchiorre et al. (2006) did further research on the behaviour of a network with respect to errors in the conditioning factors by performing a robustness analysis and Melchiorre et al. (2008) improved the predictive capability and robustness of ANNs by introducing a cluster analysis. Neaupane and Achet (2004) used ANN for monitoring the movement. Moreover, Kanungo et al. (2006) showed that a landslide susceptibility map derived from combined neural and fuzzy weighting procedure is the best amongst the other weighting techniques. Lui et al. (2006) assessed the landslide hazard using ANNs for a specic landslide typology (debris ow), considering among the triggering factors frequency of ooding, covariance of monthly precipitation, and days with rainfall higher than a critical threshold. 4. Approaches to landslide hazard assessment The landslide susceptibility assessment is a particular step in the landslide hazard assessment and is usually based on the comparison of the previously surveyed landslides and the

conditional or preparatory causal factors. With this combination a GIS is obtained in a landslide susceptibility map. In susceptibility analyses, triggering causal factors are often not considered. Some research has been done specically related to the landslide susceptibility assessment (Lee et al., 2003; Sirangelo and Braca, 2004; Guzzetti et al., 2006). Several countries have published national landslide susceptibility maps that are based on their national landslide inventory (Brabb et al., 1999; Guzzetti, 2000; Komac and Ribii, 2008). One of the proven techniques for landslide susceptibility assessment is the weights of evidence (WofE) modelling. Many landslide susceptibility have been carried out using this method (van Westen, 1993; Fernandez, 2003; van Westen et al., 2003; Lee and Choi, 2004; Suzen and Doyuran, 2004; Neuhauser and Terhorst, 2007; Magliulo et al., 2008). Essentially, the WofE method is a bivariate statistical technique that calculates the spatial probability and odds of landslides given a certain variable. Many investigations have included landslide runout in the analyses for landslide hazard assessment. With research on landslide runout or travel distance started in mid Nineties of the last century (Hungr, 1995; Finlay et al., 1999; Chen and Lee, 2000; Okura et al., 2000; Fannin and Wise, 2001; Wang et al., 2002; Crosta et al., 2003; Hunter and Fell, 2003; Bertolo and Wieczorek, 2005; Hungr et al., 2005; Malet et al., 2005; Crosta et al., 2006; van Asch et al., 2006; Pirulli et al., 2007; van Asch et al., 2007a; van Asch, et al., 2007b) where authors use three types of approaches for runout analysis. These are the empirical approach from previous landslides and geomorphological analysis, the deterministic approach from the geotechnical parameters and the dynamic approach from numerical modelling of runout.

Heuristic analysis

Statistical analysis

Several parameter maps are surveyed to apply bivariate and multivariate analysis

Deterministic analysis

Apply hydrological and slope instability models to evaluate the safety factor

rainfall

Use rainfall characteristic to identify the threshold value for landslide initiation

Articial neural network (ANN)

Horton (1945); Rumelhart et al. (1986); Ercanoglu and Gokceoglu (2002); Lee et al. (2003a); Lee et al. (2003b); Lu (2003); Arora et al. (2004); Lee et al. (2004); Neaupane and Achet (2004); Catani et al. (2005); Ercanoglu Learning from data with known characteristics to derive (2005); Ermini et al. (2005); Gomez and (2005); Miska and Jan (2005); Wang a set of weighting parameters, Kavzoglu et al. (2005); Yesilnacar and Topal (2005); which are used subsequently Kanungo et al. (2006); Lee and Evangelista to recognize the unseen data (2006); Lui et al. (2006); Melchiorre et al. (2006, 2008); Wang and Sassa (2006); Lee (2007); Pradhan and Lee (2007,2009a, 2009b, 2009c); Nefeslioglu et al. (2008); Pradhan et al. (2009); Youssef et al. (2009)

Tab. 2: Review of numerical approaches to landslide hazard assessment with short description of approach and references. Tab. 2: berprfung von numerischen Anstzen zur Gefahrenabschtzung von Rutschungen mit einer kurzen Darstellung des Ansatzes und Referenzen.

Seite 55

Key-note papers

Seite 56

Landslide fundamental

vulnerability component

assessment in the

is

and the risk areas are categorized generally in three or ve classes as very high, high, moderate, low and very low. This method is applicable for spatial analysis using GIS and usually applied at national or regional levels. This approach were found in literature from Lateltin (1997), AGS (2000), Budetta (2004), Cascini (2004), Ko Ko et al. (2004), IADB (2005), Nadim et al. (2006). With the semi-qualitative landslide risk assessment approach, weights are assigned under certain criteria, which provide numbers as outcome, instead of qualitative classes (0-1, 0-10 or 0-100). It could be applicable to any scale, but more reasonably used at medium scale. Semi-quantitative approach efciently uses spatial multi-criteria techniques implemented in GIS that facilitate standardization, weighting and data integration in a single set of tools. More details about the weighting system are published by Brand (1988), Koirala and Watkins (1988), Chowdhury and Flentje (2003), Blochl and Braun (2005), Castellanos Abella and van Westen (2005) and Saldivar-Sail and Einstein (2007). When implementing the semi-quantitative model, usually the multi-criteria evaluation is used (see references below). The input is a set of maps that are the spatial representation on the criteria, which are grouped, standardised and weighted in a criteria tree. Meanwhile the output is one or more composite index maps indicating the completion of the model used. The theoretical background for the multicriteria evaluation is based on the Analytical Hierarchical Process (AHP) developed by Saaty (1977). The AHP has been extensively applied on decision making problems (Saaty and Vargas, 2001). Recently some research has been carried out to apply AHP to landslide susceptibility assessment (Barredo et al., 2000; Mwasi, 2001; Nie et al., 2001, Wu and Chen, 2009).

Komac (2006) designed multivariate statistical processing techniques in order to obtain several landslide susceptibility models with data at scale 1:50,000 and 1:100,000. Based on the statistical results, several landslides susceptibility maps were created. Quantitative landslide risk assessment has been used for specic slopes or very small areas using probabilistic methods or percentage of losses expected (Whitman, 1984; Chowdhury, 1988). Probabilistic values (0-1) are obtained at the expense of a certain amount of monetary or human loss. Quantitative risk analysis and consequent assessment uses information about hazard probability, values of elements at risk and their vulnerability. Among the quantitative approaches found in literature there are some basic similarities but also some differences between the approaches. They include either estimation of hazard or estimation of vulnerability and consequences (Morgan, 1992; Einstein, 1988, 1997; Fell, 1994; Fell et al., 2005; Anderson et al., 1996; Ragozin, 1996; Ragozin and Tikhvinsky, 2000; Lee and Jones, 2004; AGS, 2000). 5. Landslide risk management At the end of the assessment process when landslide susceptibility and risk assessment have been identied, results and measures obtained should or may be included into the landslide risk management process governed by decision makers to mitigate landslide risk of the community or, at this level, several further approaches are possible. The strategies may be grouped into planning control, engineering solution, acceptance, and monitoring or warning systems. The risk assessed can be compared with the acceptance criteria to decide upon the landslide mitigation measures required.

Landslide (or any natural hazard for that matter) assessment process is just one of several steps in the (Landslide) Risk Management Cycle (RMC), which doesnt end at the stage where results of assessment process are included in the RMC. RMC is a live system where each measure/provision results in a consequence(s) that inuence(s) further development in and steps of this cycle. In a way we could dene it as a spiral rather than as a circular process since the same position is never reached again. 6. Conclusion In this paper, different approaches for the evaluation of slope mass processes are reviewed. In general, all analyses are based on the assumption that historical landslides and their causal relationships can be used to predict future ones (past is a key to the future). However, we can see that many researchers use different approaches to evaluate landslides, debris ow or rockfall hazard risk assessment, which mainly depend on data availability. In developing countries, usually the lack of nancial support to produce risk assessment maps for dangerous areas results in emphasis on remediation measures. Whereas in countries with high standards, the approach to the topic is focused into prevention and into remediation if disasters occur. In any event the obstacles related to the availability of data are smaller each day due to low-cost satellite information, the use of SRTM, ASTER and Google Earth, which ease the creation of landslide inventory databases, a basis for any further hazard assessments. The landslide inventory map is probably the most important data set to work on for producing a reliable prediction map of spatial and temporal probability for landslides or other slope mass movements and a necessity for any type of analyses.

evaluation

of landslide risk (Leone et al., 1996). Most publications about vulnerability are related to hazard and risk assessment (Mejia-Navarro et al., 1994; Leone et al., 1996; Ragozin and Tikhvinsky, 2000; van Westen, 2002; Hollenstein, 2005). The main object of these investigations determined the elements of risk which have impact on structures on its surface and estimate the cost. The vulnerability maps are expressed with values between 0 and 1, where 0 means no damage and 1 means total loss. Generally, the vulnerability to landslides may depend on runout distance; volume and velocity of sliding; elements at risk (buildings and other structures), their nature and their proximity to the slide; and the elements at risk (person), their proximity to the slide, the nature of the building/road that they are in, and where they are in the building, on the road, etc (Finlay, 1996). The aim of landslide hazard and risk assessment studies is to protect the population, the economy and environment against potential damage caused by landslides (Crozier and Glade, 2005). Risk in this context, is seen as a disaster that could happen in the future. The total risk map could be obtained by combining hazard and vulnerability and made directly or specic risk or consequence maps can be created and analyzed in order to achieve some preliminary conclusions. The classication of the landslide risk assessment is still in progress. At the moment the classication is based on the level of quantication dividing the landslide risk assessment methods in qualitative, semi-qualitative and quantitative (AGS, 2000; Powell, 2000; Walker 2000; Chowdhury and Flentje, 2003). The qualitative landslide risk assessment approach is based on the experience of the experts

Seite 57

Key-note papers

Seite 58

Anschrift der Verfasser / Authors addresses: Mateja Jemec Dimieva ulica 14 SI 1000 Ljubljana, Slovenia mateja.jemec@geo-zs.si Marko Komac Dimieva ulica 14 SI 1000 Ljubljana, Slovenia marko.komac@geo-zs.si

BOZZOLO, D., PAMINI, R. AND HUTTER, K.,1986. Rockfall analysis a mathematical model and its test with eld data. Proceedings of the 5th International Symposium on Landslides in Lausanne. Balkema, Rotterdam, The Netherlands, pp. 555560. BRABB, E.E., PAMPEYAN, E.H., BONILLA, M.G., 1978. Landslide susceptibility in San Mateo County, California. US Geological Survey Miscellaneous Field Studies Map, MF-360, Map at 1: 62,500 scale. BRABB, E.E., 1984. Inovative approaches to landslide hazard and risk mapping. Canadian Geotechnical Society I, 403-324. BRABB, E., COLGAN, J.P. AND BEST, T.C., 1999. Map showing inventory and regional susceptibility for Holocene Debris Flows and related fast-moving landslides in the conterminous United States. MAP MF-2329. USGS. BRAND, E. W., 1988. Landslide risk assessment in Hong Kong. In: Proceedings of the 5th International Symposium on Landslides (Lausanne, Switzerland), vol. 2, 1059-1074. BUDETTA P., 2004. Assessment of rockfall risk along roads. Department of Geotechnical Engineering, Section of Applied Geology, University of Naples Federico II piazzale Tecchio, 80, Italy.

CHEN, H. AND LEE, C. F., 2000. Numerical simulation of debris ows, Canadian Geotechnical Journal 37, 146160. CHOWDHURY, R.N., 1988. Special lecture: Analysis methods of assessing landslide risk - recent developments. In: C. Bonnard (Editor), Proceedings of the 5th International Symposium on Landslides, Rotterdam, 515-524. CHOWDHURY, R. AND FLENTJE, P., 2003. Role of slope reliability analysis in landslide risk management, Bull. Eng. Geol. Env. 62, 4146. CHUNG, CJ. F. AND FABBRI, A. G., 1999. Probabilistic Prediction Models for Landslide Hazard Mapping.Photogrammetric engineering and remote sensing, Vol. 65(12), American Society of Photogrammetry and Remote Sensing, p. 13891399, Falls Church. COE, J.A., HARP, E.L., 2007. Inuence of tectonic folding on rockfall susceptibility. American Fork Canyon, Utah, USA. Natural Hazards and Earth System Sci. 7, 114. COE, J.A., MICHAEL, J.A., CROVELLI, R.A., SAVAGE, W.Z., LAPRADE, W.T. AND NASHEM, W.D., 2004. Probabilistic assessment of precipitation triggered landslide using historical records of landslide occurence, Seattle, Whashington. Environmental & Engineering Geoscience X (2), 103-122. COUSSOT, P., MEUNIER, M., 1996. Recognition, classication and mechanical description of debris ows. Earth-Sci. Rev. 40. 209-227. CROSTA, G., 1998. Regionalization of rainfall thresholds: an aid to landslide hazard evaluation. Environmental Geology, 35, 131-145. CROSTA, G.B., CUCCHIARO, S. AND FRATTINI, P., 2003. Validation of semi-empirical relationships for the denition of debris-ow behaviour in granular materials. In: D. Rickenmann and C. Chen, Editors, Proc. 3rd International Conference on Debris Flows Hazards Mitigation: Mechanics, Prediction and Assessment, Davos, Switzerland, Millpress, Rotterda. 821831. CROSTA, G., Imposimato, S. and Roddeman, D., 2006. In: S.G. Evans, G. Scarascia Mugnozza, A. Strom and R.L. Hermanns, Editors, Proceedings of the NATO Advanced Research Workshop on Massive Rock Slope Failure: New Models for Hazard Assessment, Celano, Italy, 1621 June 2002, NATO Science Series. Series IV, Earth and Environmental Sciences vol. 49. CROZIER, M., 1986. Landslides: causes, consequences and environment, Croom Helm, London, 252. CROZIER, M.J., 1999. Prediction of rainfall triggered landslides: a test of the antecedent water status model. Earth Surf. Proces. Landf. 24 (9), 825 833. CROZIER, M.J. and GLADE, T., 2005. Landslide hazard and risk: Issues, Concepts and Approach. In: Glade T, Anderson M, Crozier M (Eds) Landslide hazard and risk. Wiley, Chichester, 1-40. CRUDEN, D.M., 1991. A simple denition of a landslide. Bulletin of the International Association of Engineering Geology, 43: 27-29. CRUDEN D.M. AND VARNES, D.J., 1996. Landslide types and processes. In: Turner A.K.; Shuster R.L. (Editors), Landslides: Investigation and Mitigation. Transportation Research Board, Special Report No. 247, pp. 36-75. CRUDEN, D. AND FELL, R., 1997. Landslide risk assessment. A.A.Balkema, Rotterdam, 371 pp. DAI, F.C., LEE, C.F. AND NGAI, Y.Y., 2002. Landslide risk assessment and management an overview. Engineering Geology, 64(1): 65-87.

DAI, F.C. AND LEE, C.F., 2001. Frequency-volume relation and prediction of rainfall induced landslides. Engineering Geology 59 (3-4), 253-266. DAVIES, M.C.R., HAMZA, O., HARRIS, C., 2001. The effect of rise in mean annual temperature on the stability of rock slopes containing ice-lled discontinuities. Permafrost and Periglacial Processes 12 (1), 137144. DHAKAL, A. S., AMADA, T. AND ANIYA, M., 2000. Landslide hazard mapping and its evaluation using GIS: An investigation of sampling schemes for a grid-cell based quantitative method.Photogrammetric engineering and remote sensing, Vol. 66(8), American Society of Photogrammetry and Remote Sensing, p. 981989, Falls Church. DORREN, L.K.A., BERGER, F., PUTTERS, U.S., 2006. Real size experiments and 3D simulation of rockfall on forested and nonforested slopes. Natural Hazards and Earth System Sci. 6, 145153. EINSTEIN, H.H., 1988. Special lecture: Landslide risk assessment procedure. In: C. Bonnard (Editor), Proc 5th International Symposium on Landslides. Publ Rotterdam: A A Balkema, Lausanne, pp. 1075-1090. EINSTEIN, H.H., 1997. Landslide Risk Systematic approaches to assessment and management in Landslide Risk Assessment. In: Cruden, D., Fell, R. (eds). Landslide Risk Assessment, 25-50. ERCANOGLU, M., 2005. Landslide susceptibility assessment of SE Bartin (West Black Sea region, Turkey) by articial neural networks, Natural Hazards and Earth System Sciences 5, 979992. ERCANOGLU, M. AND GOKCEOGLU, C., 2002. Assessment of landslide susceptibility for a landslide-prone area (north of Yenice, NW Turkye) by fuzzy approach. Environmental Geology 41, 720730. ERISMANN, T.H., 1986. Flowing, rolling, bouncing, sliding, synopsis of basic mechanisms. Acta Mechanica 64, 101110. Ermini, L., Catani, F. and Casagli, N., 2005. Articial neural networks applied to landslide susceptibility assessment, Geomorphology 66 (14), 327343. EVANS, S.G. AND HUNGR, O.,1993. The assessment of rockfall hazard at the base of talus slopes. Canadian Geotechnical Journal 30, 620636. FANNIN, R.J. AND WISE, M.P., 2001. An empiricalstatistical model for debris ow travel distance, Canadian Geotechnical Journal 38, 982994. FELL, R., 1994. Landslide risk assessment and acceptable risk. Canadian Geotechnical Journal, 31(2): 261-272. FELL, R., HO, K. K. S., LACASSE, S., AND LEROI, E., 2005. A framework for landslide risk assessment and management, in: Landslides Risk Management, edited by: Hungr, O., Fell, R., Couture, R., and Eberhardt, E., Taylor and Francis, London, 326. FERNANDEZ, T., 2003. Methodology for Landslide Susceptibility Mapping by Means of a GIS. Application to the Contraviesa Area (Granada, Spain). Natural Hazards, 30 (3), 297 308. FIORILLO, F., WILSON, R.C., 2004. Rainfall induced debris ows in pyroclastic deposits, Campania (Southern Italy). Engineering Geology, 75, 263- 289. FIRDAINI, 2008. Land capability and landslide hazard assessment for landuse priority in Gintung watershed. Purworejo Regency, Central Java, Indonesia. Jogjakarta Enschede, Gadjah Mada- ITC. Master. FINLAY, P.J.,1996. The risk assessment of slopes. School of Civil Engineering, University of New South Wales, Australia, PhD thesis.

Literatur / References:
AGS, 2000. Landslide risk management concept and guidelines. Australian Geomechanics Society sub- committee on landslide risk management, Australian Geomechanics 35 (1), 4492. ALCANTARA-AYALA, I., 2004. Hazard assessment of rainfall-induced Geomorphology 61, 19 40. landsliding in Mexico.

CAINE, N., 1980. The rainfall intensity duration control of shallow landslides and debris ow. Geograska Annaler Series A Physical Geography, 62: 23-27. CAINE, N.AND MOOL, P.K. 1982. Landslide in the Kolpu Khola drainage, middle mountains, Nepal. Mountain Research and Development 2 (2), 157-173. CANNON, S. H. AND ELLEN, S. 1985: Rainfall conditions for abundant debris avalanches, San Francisco Bay Region, California,- California Geology, Vol. December 1985, Depatrment of Conservation, Division of Mines and Geology, p. 267272, Sacramento. CANUTI, P., FOCARDI, P., GARZONIO, C.A., 1985. Correlation between rainfall and landslides. Bull. Int. Assoc. Eng. Geol. 32, 49 54. CARRARA, A., CARRATELLI, E.P., MERENDA, L., 1977. Computer-based data bank and statistical analysis of slope instability phenomena. Zeitschrift fr Geomorphologie 21, 187222. CARRARA, A. 1983: Multivariate models for landslide hazard evaluation. - Mathematical Geology, Vol. 15, Kluwer Academic Publishers, p. 403426, Dordrecht. CARRARA, A., CARDINALI, M., GUZZETTI, F., PASQUUI, V. AND REICHENBACH, P., 1991. GIS techniques and statistical models in evaluating landslide hazard. Journal Earth Surface Processes and Landform 16 (5), 427-445. CARRARA, A., 1991. GIS techniques and statistical models in evaluating landslide hazard, Earth Surface Processes and Landforms, v. 16, p. 427-445. CASTELLANOS, E. AND VAN WESTEN, C.J., 2003. Landslide Hazard Assessment Using The Heuristic Models. ITC, Enschede, The Netherland. CASTELLANOS ABELLA, E.A. AND VAN WESTEN, C.J., 2005. Development of a system for landslide risk assessment for Cuba. In: Proceedings of the international conference on landslide risk management, 31 May - 3 June, 2005 Vancouver. / ed. by O. Hungr, R. Fell, R. Couture and E. Eberhardt. London, 1-10. CASCINI, L., 2004. Risk assessment of fast landslide- from theory to practice. Department of Civil Engineering, University of Salerno, Italy. CATANI, F., CASAGLI, N., ERMINI, L., RIGHINI, G. AND MENDUNI, G., 2005. Landslide hazard and risk mapping at catchment scale in the Arno River Basin, Landslides 2 (4), 329343.

ANDERSON, L.R., BOWLES, D.S., PACK, R.T. AND KEATON, J.R., 1996. A risk based method for landslide mitigation. In: K. Senneset (Editor), VII International Symposium on Landslide. A.A.Balkema, Trondheim, Norway, 135-140. ARORA, M.K., GUPTA, A.S.D. AND GUPTA, R.P., 2004. An articial neural network approach for landslide hazard zonation in the Bhagirathi (Ganga) Valley, Himalayas, International Journal of Remote Sensing 25 (3), 559572. AZZONI, A., BARBERA, G.L., ZANINETTI, A., 1995. Analysis and prediction of rockfalls using a mathematical model. International J. Rock Mechanics and Mining Sci. 32, 709724. BARREDO, J.I., BENAVIDES, A., HERVAS, J., VAN WESTEN, C.J., 2000. Comparing heuristic landslide hazard assessment techniques using GIS in the Tirajana basin, Gran Canaria Island, Spain. International Journal of Applied Earth Observation and Geoinformation 2, 9 23. BERGER, F., COROMINAS, J., LOPEZ-CARRERAS, C., BRAUNER, M., KIENHOLZ, H., GRASSL, H., 2002. ROCKFOR, RockfallForest Interrelation, Efciency of the Protective Function of Mountain Forest against Rockfall. Second Annual Report (CD); nanced 84 M. Brauner et al. / Forest Ecology and Management 207 (2005) 7585 by the European Commission (Quality of Life and Management of Living Resources, FP5). BERTOLO, P. AND WIECZOREK, G.F., 2005. Calibration of numerical models for small debris ows in Yosemite Valley, California, USA. Natural Hazards and Earth System Sciences 5, 9931001. BLCHL, A. AND BRAUN, B. 2005. Economic assessment of landslide risks in the Swabian Alb, Germany research framework and rst results of homeowners and experts surveys. Nat Hazards Earth Syst Sci 5, 389396. BOMMER, J.J. AND RODRIGUEZ, C.E., 2002. Earthquake-induced landslides in Central America. Engineering Geology 63, 189 220.

Seite 59

Key-note papers

Seite 60

FINLAY, P.J., FELL, R., MAGUIRE, P.K., 1997. The relationship between the probability of landslide occurrence and rainfall. Can. Geotech. J. 34, 811 824. FINLAY, P.J., MOSTYN, G.R. AND FELL, R., 1999. Landslide Risk Assessment: prediction of travel distance. Canadian Geotechnical Journal, 36: 556-562. GARDNER, J.S., 1983. Rockfall frequency and distribution in the Highwood Pass area, Canadian Rocky Mountains. Zeitschrift fr Geomorphologie N.F. 27, 311324. GIANNECCHINI, R., 2006. Relationship between rainfall and shallow landslides in the southern Apuan Alps (Italy). Natural Hazards Earth System Science 6, 357364. GLADE, T., 2000. Modelling landslides: triggering rainfall thresholds at a range of complexities. Proceedings of the 8th int. symp. On landslides held in Cardiff on 2630 June, Landslides in Research, Theory and Practice, vol. 2, pp. 633 640. GLADE, T., ANDERSON, M. AND CROZIER, M.J., 2005. Landslide Hazard and Risk. JohnWiley & Sons, Ltd., Chichester, Eangland, 802 pp. GOMEZ, H.T. AND KAVZOGLU, T., 2005. Assessment of shallow landslide susceptibility using articial neural networks in Jabonosa River Basin, Venezuela, Engineering Geology 78 (12), 1127. GOVI, M., MORTARA, G., AND SORZANA, P. F., 1985. Eventi idrologici e frane, Geol. Appl. e Idrog., 20(2), 359375. GORSEVSKI, P. V., GESSLER, P. AND FOLTZ, R. B., 2000. Spatial prediction of Landslide hazard using logistic regression and GIS.- 4th International Conference on Integrating GIS and Environmental Modeling (GIS/EM4): Problems, Prospect and Research Needs, Banff, Alberta. GRUBER, S., HOELZLE, M., HAEBERLI, W., 2004. Permafrost thaw and destabilization of Alpine rock walls in the hot summer of 2003. Geophyiscal Res. Lett. 31, L13504. GUINAU, M., VILAJOSANA, I., VILAPLANA, J. M., 2007. Gis-based debris ow source and runout susceptibility assessment from DEM data a case study in NW Nicaragua, Natural Hazards and Earth System Sciences, 7: 703-716. GUZZETTI, F., 2000. Landslide fatalities and the evaluation of landslide risk in Italy. Engineering Geology 58(2), 89107. GUZZETTI, F., PERUCCACI, S., ROSSI, M. AND STARK, C.P., 2007. Rainfall thresholds for the initiation of landslides in central and southern Europe. Meteorol. Atmos. Phys. 98, 239-267. GUZZETTI, F., CROSTA, G., DETTI, R., AGLIARDI, F., 2002. STONE: a computer program for the three-dimensional simulation of rockfalls. Computers & Geosciences 28 (9), 10791093. GUZZETTI F., CARDINALI M., REICHENBACH P., CIPOLLA F., SEBASTIANI C., GALLI M., SALVATI P., 2004. Landslides triggered by the 23 November 2000 rainfall event in the Imperia Province, Western Liguria, Italy, Engineering Geology 73, 229245. GUZZETTI F., GALLI M., REICHENBACH P., ARDIZZONE F. AND CARDINALI M., 2006. Landslide hazard assessment in the Collazzone area, Umbria, Central Italy. Natural Hazards and Earth System Sciences, 6, 115131. HARP, E.L., WILSON, R.C., 1995. Shaking intensity thresholds for rock falls and slides: Evidence from 1987 Whittier Narrows and Superstition Hills earthquake strongmotion records. Bulletin of the Seismological Soci. of America 85 (6), 17391757. HOLLENSTEIN, K., 2005. Reconsidering the risk assessment concept: Standardizing the impact description as a building block for vulnerability assessment, Nat. Hazards Earth Syst. Sci., 5, 301307. http://www.nat-hazards-earth-syst-sci. net/5/301/2005/.

HORTON, R.E., 1945. Erosional development of streams and their drainage basins: hydro physical approach to quantitative morphology, Bull. Geol. Soc. Am. 56 (1945), pp. 275370. HUNGR, O. AND EVANS, S.G., 1988. Engineering evaluation of fragmental rockfall hazards. Proceedings of the 5th International Symposium on Landslides in Lausanne. The Netherlands, Balkema, Rotterdam, pp. 685690. HUNGR, O., 1995. A model for the runout analysis of rapid ow slides, debris ows and mavalanches, Canadian Geotechnical Journal 32, 610623. HUNGR, O., COROMINAS, J., EBERHARDT, E., 2005. Estimating landslide motion mechanism, travel distance and velocity. In: Hungr, O., Fell, R., Couture, R., Eberhardt, E. (Eds.), Landslide Risk Management. Taylor & Francis Group, Vancouver, pp. 99128. HUNTER, G. AND FELL, R., 2003. Travel distance angle for "rapid" landslides in constructed and natural soil slopes, Canadian Geotechnical Journal 40, 11231141. IADB, 2005. Indicators of disaster risk and risk management. Summary report for WCDR, Inter-American Development Bank (IADB), Manizales, Colombia. ISSMGE TC32- TECGNICAL COMMITTEE ON RISK ASSESSMENT AND MANAGEMENT, 2004. Glossary of Risk Assessment Terms- Version 1, July 2004, pp.7. IVERSON, R.M., REID, M.E., LAHUSEN, R.G., 1997. Debris-ow mobilization from landslides. Annu. Rev. Planet Sci. 25, 85 138. JAKOB, M. AND WEATHERLY, H. 2003. A hydroclimatic threshold for landslide initiation on the North Shore Mountains of Vancouver, British Columbia. Geomorphology, 54, 137-156. JAKOB, M., HOLM, K., LANGE,O. AND SCHWAB, J.W., 2006. Hydrometeorological thresholds for landslide initiation and forest operation shutdowns on the north coast of British Columbia. Landslides 3, 228-238. JOHNSON, A.M., 1970. Physical Processes in Geology. W.H. Freeman, 557 p. JTC-1 JOINT TECHNICAL COMMITTEE ON LANDSLIDES AND ENGINEERED SLOPES, 2008. Guidelines for landslide susceptibility, hazard and risk zoning, for land use planning. Engineering Geology 103, 8598. KANUNGO, D.P., ARORA, M.K., SARKAR, S. AND GUPTA, R.P., 2006. A comparative study of conventional, ANN black box, fuzzy and combined neural and fuzzy weighting procedures for landslide susceptibility zonation in Darjeeling Himalayas. Engineering Geology 85, 347 366. KEEFER, DK, 2000. Statistical analysis of an earthquake-induced landslide distribution the 1989 Loma Prieta, California event. Engineering Geology 58:231-249. KHAZAI, B., SITAR, N., 2004. Evaluation of factors controlling earthquake-induced landslides caused by Chi-Chi earthquake and comparison with the Northridge and Loma Prieta events. Engineering Geology 71, 7995. KIRKBY, M.J. AND STATHAM, I., 1975. Surface stone movement and scree formation. Journal of Geology 83, 349 362. KOIRALA, N.P. AND WATKINS, A.T., 1988. Bulk appraisal of slopes in Hong Kong. In Landslides, Proc. Of the Fifth Int. Symp. on Landslide, (Ed. C. Bonnard), Lausanne, Switzerland. A.A. Balkema, Rotterdam, The Netherlands, 2, 1181-1186. KOMAC, M., 2006. A landslide susceptibility model using the analytical hierarchy process method and multivariate statistics in perialpine Slovenia. Geomorphology 74, 17-28. KOMAC, M., AND RIBII, M., 2008. Landslide susceptibility map of Slovenia 1:250.000. Geological Survey of Slovenia, Ljubljana.

KOMAC, M., KUMELJ, . AND RIBII, M., 2010. Debris-ow susceptibility map of Slovenia 1:250,000. Ljubljana: Geological survey of Slovenia. KO KO, C., FLENTJE, P. AND CHOWDHURY, R., 2004. Landslides qualitative hazard and risk assessment: method and its reliability, Bull. Eng. Geol. Env. 63 (2), 149165. LAN, H. X., ZHOU, C. H., WANG, L. J., ZHANG, H. Y., LI, R. H., 2004. Landslide hazard spatial analysis and prediction using GIS in the Xiaojiang watershed, Yunnan, China, Engineering Geology, 76: 109128. LAPRADE, W.T., KIRKLAND, T.E., NASHEM, W.D., ROBERTSON, C.A., 2000. Seattle landslide study. Shannon & Wilson, Inc. Internal Report W-7992 01. LATELTIN, O., 1997. Bercksichtigung der Massenbewegungsgefahren bei raumwirksamen Ttigkeiten. Empfehlungen 1997. Swiss Federal Ofce for Water and Geology (FOWG). http://www.bwg.admin.ch/themen/natur/e/index.htm LEE, E.M. AND JONES, D.K.C., 2004. Landslide risk assessment. Thomas Telford, London, 454 pp. LEE, S. AND J. CHOI, 2004. Slide susceptibility mapping using GIS and Weight of evidence model. Geogr. Inform. Sci., 188: 789-814. LEE, S., 2007. Application and verication of fuzzy algebraic operators to landslide susceptibility mapping, Environmental Geology 52, 615623. LEE, S. AND EVANGELISTA, D.G., 2006. Earthquake-induced landslide-susceptibility mapping using an articial neural network, Natural Hazards and Earth System Sciences 6, 687695. LEE, S., RYU, J.H., LEE, M.J. AND WON, J.S., 2003a. Use of articial neural network for analysis of the susceptibility to landslides at Boun, Korea. Environmental Geology 44, 820-833. LEE, S., RYU, J.H., MIN, K. AND WON, J.S., 2003b. Landslide susceptibility analysis using GIS and articial neural network. Earth surface processes and landforms 28, 1361-1376. LEE, S., RYU, J.H., WON, J.S. AND PARK, H.J., 2004. Determination and application of the weights for landslide susceptibility mapping using articial network. Engineering Geology, 71(3-4), 289-302. LEONE, F., ASTE, J. P., AND LEROI, E., 1996. Vulnerability assessment of elements exposed to mass movements: working toward a better risk perception, in: Landslides, Glissements de terrain, Proceed. VII Int. Sym. Landslides, Trondheim, edited by: Senneset, K., Rotterdam, 263270. LIDA, T., 1999. A stochastic hydro-geomorphological model for shallow landsliding due to rainstorm. Catena 34 (3-4), 293-313. LU, P. R., 2003. Articial neural networks and grey systems for the prediction of slope instability. Natural Hazards 30(3),383398. LUCKMAN, B.H., FISKE, C.J., 1995. Estimating long-term rockfall accretion rates by lichenometry. In: Slaymaker, O. (Ed.), Steepland Geomorphology. Wiley, Chichester, UK, pp. 233255. LUI, Y., GUO, H.C., ZOU, R. AND WANG, L.J., 2006. Neural network modelling for regional hazard assessment of debris ow in Lake Qionghai Watershed, China, Environmental Geology 49, 968976. MCCARROLL, D., SHAKESBY, R.A., MATTHEWS, J.S., 1998. Spatial and temporal patterns of Late Holocene rockfall activity on a Norwegian talus slope: lichenometry and simulation-modelling approach. Arctic and Alpine Res. 30, 5160. MALET, J.-P., LAIGLE, D., REMAITRE, A., MAQUAIRE, O., 2005. Triggering conditions and mobility of debris ows associated to complex earth ows. Geomorphology 66, 215235.

MAGLIULO, P., DI LISIO, A., RUSSO, F. AND ZELANO, A., 2008. Geomorphology and landslide susceptibility assessment using GIS and bivariate statistics: a case study in southern Italy. Springer Science. Natural Hazard 47, 411-435. MARZORATI, S., LUZI, L., DE AMICIS, M., 2002. Rock falls induced by earthquakes: a statistical approach. Soil Dynamics and Earthquake Engineering 22 (7), 565577. MATSUOKA, N., SAKAI, H., 1999. Rockfall activity from an alpine Geomorphology 28, 309328. cliff during thawing periods.

MATSUOKA, N., 2006. Frost wedging and rockfalls on high mountain rock slopes: 11 years of observations in the Swiss Alps. Geophysical Res. Abstracts 8, 05344. MAZENGARB, C., 2004. Map 3, Hobart - Potential Debris- Flow Hazard. Tasmanian Landslide Hazard Series. Mineral Resources Tasmania, Department of Infrastructure Energy and Resources, Hobart. MEJA-NAVARRO, M., WOHL, E.E. AND OAKS, S.D., 1994. Geological hazard, vulnerability, and risk assessment using GIS: model for Glenwood Springs, Colorado, Geomorphology 10, 331354. MELCHIORRE, C., MATTEUCCI, M. AND REMONDO, J., 2006. Articial neural network and robustness analysis in landslide susceptibility zonation. International Joint Conference on Neural Networks, July 16-21, Vancouver, BC, Canada, 8808-8814. MELCHIORRE, C., MATTEUCCI, M., AZZONI, A. AND ZANCHI, A., 2008. Articial neural networks and cluster analysis in landslide susceptibility zonation, Geomorphology 94, 379400. MILLER, D. J. AND BURNETT, K. M., 2008. A probabilistic model of debris-ow delivery to stream channels, demonstrated for the Coast Range of Oregon, USA, Geomorphology, 94:184-205. MISKA, L. AND JAN, H., 2005. Evaluation of current statistical approaches for predictive geomorphological mapping. Geomorphology 67 (3-4), 299-315. MONTGOMERY, D. R. AND DIETRICH, W. E., 1994. A physically based model for the topograc control on shallow landsliding.Water resources research, Vol. 30/4, American Geophysical Union, p. 11531171, Washington. MONTGOMERY, D. R., SCHMIDT, K. M., GREENBERG, H. M. AND DIETRICH, W. E., 2000. Forest clearing and regional sliding. Geology, Vol. 28/4, The Geologic Society of America, p. 311314, Boulder. MONTGOMERY, D. R., SULLIVAN, K. AND GREENBERG, H. M., 1998. Reginal test of a model for shallow landsliding. Hydrological Processes, Vol 12, John Wiley & Sons, Ltd., p. 943955, New York. MORGAN, G.C., 1992. Quantication of risks from slope hazards. Open le report No 1992-15, Geological Survey of Canada, Canada. MWASI, B., 2001. Land use conicts resolution in a fragile ecosystem using Multi-Criteria Evaluation (MCE) and a GIS-based Decision Support System (DSS). Int. Conf. on Spatial Information for Sustainable Development, Nairobi, Kenya, FIG International Federation of Surveyors. 11 pp. NaDIM, F., KJEKSTAD, O., DOMAAS, U., RAFAT, R. AND PEDUZZI, P. 2006. Global landslide risk case study. In: Arnold M, Chen R.S., Deichemann U, Dilley M (eds) Natural disaster hotspots. Case studies. The World Bank, Washington, DC, p 204 NEAUPANE, K.M. AND ACHET, S.H., 2004. Use of backpropagation neural network for landslide monitoring: a case study in the higher Himalaya, Engineering Geology 74 (34), 213226.

Seite 61

Key-note papers

Seite 62

NEFESLIOGLU, H.A., GOKCEOGLU, C. AND SONMEZ, H., 2008. An assessment on the use of logistic regression and articial neural networks with different sampling strategies for the preparation of landslide susceptibility maps. Engineering Geology 97, 171191. NEUHUSER, B. AND TERHORST, G., 2007. Landslide susceptibility assessment using weights-ofevidence applied to a study area at the Jurassic escarpment (SW-Germany). Geomorphology 86, 1224. NEULAND, H. A., 1976. A prediction model of landslips. Catena, Vol. 3, Elsevier Science B.V., p. 215230, Amsterdam. NIE, H.F., DIAO, S.J., LIU, J.X., HUANG, H., 2001. The application of remote sensing technique and AHP-fuzzy method in comprehensive analysis and assessment for regional stability of Chongqing City, China. Proc. of the 22nd Asian Conf. on Remote Sensing, Vol. 1, Centre for Remote Imaging, Sensing and Processing (CRISP), National University of Singapore, Singapore Institute of Sorveyors and Valuers (SISV) and Asian Association on Remote Sensing (AARS), Singapore, pp. 660665. OKURA, Y., KITAHARA, H., SAMMORI, T., KAWANAMI, A., 2000. The effects of rockfall volume on runout distance. Engineering Geology 58, 109 124. PIKE, R. J., 1988. The geometric signature: Quantifying landslide-terrain types from digital elevation models. - Mathematical Geology, Vol. 20, Kluwer Academic Publishers, p. 491511, Dordrecht. PIRULLI, M., BRISTEAU, M. O., MANGENEY, A., AND SCAVIA, C., 2007. The effect of earth pressure coefcient on the runout of granular material, Environ. Modell. Softw., 22(10), 14371454. POWELL, G., 2000. Discussion Landslide Risk Management Concepts And Guidelines. Australian Geomechanics, Volume 35, No 1, 49-52. POZZONI, M., AMBROSI, C., SALVETTI, A., THRING, M., GERMANNCHIARI, C., 2009. Conceptual debris ow modeling for risk assessment at the municipal scale, SUPSI, Manno. (www.ist.supsi.ch/Content/main/uploaded/img/ progetti/dfwalk_big.pdf, 12. 1. 2009) PRADHAN, B. AND LEE, S., 2007. Utilization of optical remote sensing data and GIS tools for regional landslide hazard analysis by using an articial neural network model, Earth Science Frontier 14 (6), 143152. PRADHAN, B. AND LEE, S., 2009a. Delineation of landslide hazard areas on Penang Island, Malaysia, by using frequency ratio, logistic regression, and articial neural network model, Environmental Earth Sciences. PRADHAN, B. AND LEE, S., 2009c. Regional landslide susceptibility analysis using back-propagation neural network model at Cameron Highland, Malaysia, Landslides. PRADHAN, B. AND LEE, S., 2009b. Landslide risk analysis using articial neural network model focusing on different training sites, International Journal of Physical Sciences 4 (1), 115. PRADHAN, B., LEE, S. AND BUCHROITHNER, M.F., 2009. Use of geospatial data for the development of fuzzy algebraic operators to landslide hazard mapping: a case study in Malaysia, Applied Geomatics 1, 315. R2 RESOURCE CONSULTANTS, 2005. Upper Nehalem Watershed Analysis. Oregon Department of Forestry (Salem) 1231. RAGOZIN, A.L., 1996. Modern problems and quantitative methods of landslide risk assessment. Senneset K (Ed.) Landslides - Gliessements de Terrain. Rotterdam, A.A. Balkema. 1:339-44. RAGOZIN, A.L. AND TIKHVINSKY, I.O., 2000. Landslide hazard, vulnerability and risk assessment, 8th International symposium on landslides. Thomas Telford, Cardiff, Wales.

RITCHIE, A.M., 1963. Evaluation of rockfall and its control. Washington, DC: Highway Research Board, National Research Council, Highway Research Record 17, pp.13 28. RUFF, M. AND CZURDA, K., 2007. Landslide susceptibility analysis with a heuristic approach in the Eastern Alps, Vorarlberg, Austria. Geomorphology 94, 314-324. RUMELHART, D.E., HINTON, G.E. AND WILLIAMS, R.J., 1986. Learning representations by back-propagating errors, Nature 323, 533 536. SAATY, T.L., 1977. A scaling method for priorities in hierarchical structures. Journal of Mathematical Psychology 15, 234 281. SAATY, T.L. AND VARGAS, L.G., 2001. Models, methods, concepts and applications of the analytic hierarchy process. Kluwer Academic Publishers, 333 pp. SAHA, A.K., GUPTA, R.P., SARKAR, I., ARORA, M.K. AND CSAPLOVICS, E., 2005. An approach for GIS based statistical landslide susceptibility zonation-with a case study in the Himalayas. Springer-Verlag. SALDIVAR-SALI, A. AND EINSTEIN, H.H., 2007. A landslide risk rating system for Baguio. Philippines. Engineering Geology 91 (2-4), 85-99. SASS, O., 2005. Temporal variability of rockfall in the Bavarian Alps, Germany. Arctic, Antarctic, and Alpine Res. 37 (4), 564573. SCHMIDT, J. AND DIKAU, R., 2004. Modelling historical climate variability and slope stability. Geomorphology 60 (3-4), 433-447. SIRANGELO, B. AND BRACA, G., 2004. Identication of hazard conditions for mudow occurrence by hydrological model. Application of FLaIR model to Sarno warning system, Engineering Geology 73, 267276. SOETERS, R. AND VAN WESTEN, C.J., 1996. Slope Instability. Recognition, analysis and zonation. In: A.K. Turner and R.L. Schuster (Editors), Landslide: Investigations and Mitigation. Special Report 247. Transportation Research Board. National Research Council. National Academy Press., Washington, D.C, pp. 129-177. STATHAM, I. AND FRANCIS, S.C., 1986. Inuence of scree accumulation and weathering on the development of steep mountain slopes. In: Abrahams, A.D. (Ed.), Hillslope Processes. Allen and Unwin, Winchester, pp. 245 267. STOFFEL, M., WEHRLI, A., KHNE, R., DORREN, L.K.A., PERRET, S., KIENHOLZ, H., 2006. Assessing the protective effect of mountain forests against rockfall using a 3D simulation model. Forest Ecol. Management 225, 113122. SUZEN, M.L. AND DOYURAN, V., 2004. A comparison of the GIS based landslide susceptibility assessment methods: multivariate versus bivariate, Environmental Geology, 45, 665-679. TERLIEN, M.T.J., VAN WESTEN, C.J. AND ASCH, T.W.J., 1995. Deterministic modelling in GIS based landslide hazard assessment. In: A. Carrara and F. Guzzetti (Editors), Geographical Information Systems in Assessing Natural Hazards. Kluwer Academic Publishers, Netherlands, pp. 57-77. VAN ASCH, T.W.J., BUMA, J. AND VAN BEEK, L.P.H., 1999. A view on some hydrological triggering system in Geomorphology, 30(1-2), 25-32. landslides.

VAN ASCH, T. W. J., VAN BEEK, L.P.H. AND BOGAARD, T.A., 2007b. problems in predicting the mobility of slow mowing landslides. Engineering Geology, 91 (1), 55-46. VAN WESTEN, C.J., 1993. Remote Sensing and Geographic Information System for Geological Hazard Mitigation. ITC-Journal, 4: 393-399. VAN WESTEN, C.J. AND TERLIEN, M.T.J., 1996. An approach towards deterministic landslide hazard analaysis in GIS. A case study form manizales (Colombia). Earth surface processes and landforms 21, 853-868. VAN WESTEN, C.J., 2002. Hazard, vulnerability and risk analysis. ITC. VAN WESTEN, C.J., RENGERS, N. AND SOETERS, R., 2003. Use of geomorphological information in indirect landslide susceptibility assessment. Natural Hazards, 30(3): 399-419. VAN WESTEN, C.J., 2004. Geo-infromation tools for landslide risk assessment: an overwiev of development. In: W.A. Lacerda, M. Ehrlich, S. Fontoura and A. Sayao (Editors), Proc 9th International Symposium on Landslides. Landslides: Evaluation and Stabilization. ISL 04. Balkema, Taylor & Francis Group, Rio de Janeiro, pp. 39-56. VAN WESTEN, C.J., CASTELLANOS ABELLA, E.A. AND SEKHAR, L.K., 2008. Spatial data for landslide susceptibility, hazards and vulnerability assessment : an overview. In: Engineering geology, 102 (3-4), 112-131. VARNES, D.J., 1978. Slope movement types and processes. In: Schuster R.L., Krizek R. J. (Editors), Landslides, analysis and control. Transportation Research Board, Special Report No. 176, National Academy of Sciences, pp. 11-33. VARNES, D.J. AND IAEG, 1984. Landslide Hazard Zonation: a rewiev of principles and practise. UNESCO, Darantiere, Paris, 61 pp. WALKER, B.F., 2002. Response to discussin by G. Powell. Australian Geomechanics 35 (3), 111113. WANG, F.W., SASSA, K. AND WANG, G., 2002. Mechanism of a long runout landslide triggered by the August 1998 heavy rainfall in Fukushima Prefecture, Japan. Engineering Geology, 63 (1-2), 169- 185. WANG, H.B. AND SASSA, K., 2006. Rainfall-induced landslide hazard assessment using articial neural networks. Earth Surface Processes and Landforms, 31(2):235-247. WANG, H.B., XU, W.Y. AND XU, R.C., 2005, Slope stability evaluation using Back Propagation Neural Networks. Engineering Geology, 80 (3-4), 302-315. WEN, B. P. AND AYDIN, A., 2005. Mechanism of a rainfall-induced slide-debris ow: constraints from microstructure of its slip zone, Engineering Geology, 78: 6988. WHITMAN, R.V., 1984, Evaluating calculated risk in geotechnical engineering. ASCE Journal of Geotechnical Engineering, 110(2), 145-188. WIECZOREK, G.F., 1984. Preparing a detailed landslide inventory map for hazard evaluation and reduction. Bulletin of the Association of Engineering Geologists, XXI (3): 337-342. WIECZOREK, G. F. AND SARMIENTO, J., 1983. Signicance of storm intensity-duration for triggering of debris ows near La Honda, California.- Geological Society of America, Abstracts with Programs, Vol. 15, No. 5, Geological Society of America, p. 289, Boulder. WU, T.H., TANG, W.H. AND EINSTEIN, H.H., 1996. Landslide hazard and risk assessment. In: A.K. Turner and R.L. Schuster (Editors), Landslide Investigation and Mitigation, Transportation Research Board, National Research Council, pp. 106-128.

WU, C.H. AND CHEN, S.C.,2009. Determining landslide susceptibility in Central Taiwan from rainfall and six site factors using the analytical hierarchy process method. Geomorphology 112, 190-204. YESILNACAR, E. AND TOPAL, T., 2005. Landslide susceptibility mapping: a comparison of logistic regression and neural networks method in a medium scale study, Hendek region (Turkey). Engineering Geology 79, 251266. YOUSSEF, A.M., PRADHAN, B., GABER, A.F.D. AND BUCHROITHNER, M.F., 2009. Geomorphological hazard analysis along the Egyptian red sea coast between Safaga and Quseir, Natural Hazards and Earth System Sciences 9, 751766. ZAITCHIK, B.F., ES, H.M. AND SULLIVAN, P.J., 2003. Modelling Slope Stability in Honduras: Parameter sensitivity and Scale of Aggregation. Soil Science Society of American Journal 67, 268 -278. ZEZERE, J. L., TRIGO, R. M., AND TRIGO, I. F., 2005. Shallow and deep landslides induced by rainfall in the Lisbon region (Portugal): assessment of relationships with the North Atlantic Oscillation, Natural Hazards Earth Syst. Sci., 5, 331344. ZHOU, G., ESAKI, T., MITANI, Y., XIE, M., MORI, J., 2003. Spatial probabilistic modeling of slope failure using an integrated GIS Monte Carlo simulation approach, Engineering Geology, 68: 373386. ZINGGERLE, A., 1989. Steinschlagsimulation in Gebirgswaldern: Modellierung der relevanten Teilprozesse. Diploma Thesis, Department of Geography, University of Berne.

VAN ASCH, T. W. J., MALET, J.-P., AND VAN BEEK, L. P. H., 2006. Inuence of landslide geometry and kinematic deformation to describe the liquefaction of landslides: some theoretical considerations, Engineering Geology 88 (12), 5969. VAN ASCH, T. W. J., MALET, J.P., VAN BEEK, L.P.H. AND AMITRANO, D., 2007a. Techniques, issues and advances in numerical modelling of landslide hazard. Bulletin de la Societe Geologique de France, 178 (2), 65-88.

Seite 63

Key-note papers

Seite 64

ROLAND NORER

1. Introduction A glance at the legal framework on assessment and mapping of geological hazards is difcult.
1

Both provisions were classied as binding and directly applicable.5 In


6

Legal Framework for Assessment and Mapping of Geological Hazards on the International, European and National Levels Rechtlicher Rahmen fr Analyse und Kartierung geologischer Gefahren auf internationaler, europischer und nationaler Ebene
Summary: Legal standards for the assessment and mapping of geological hazards are rather scarce at the international and European level. Certain protocols to the Alpine Convention provide for the obligation to map geological hazards, but they fail to adopt substantive standards for it. At a European level, standards such as those for priority areas are only provided for in drafts such as the proposal for a Directive establishing a framework for the protection of soil or are mentioned in the Communication on the Community approach to prevent natural disasters. At a national level, there are legal provisions in connection with preventive planning on natural disasters, although the general problem on the coexistence of multiple area-related denitions persists. The extensive exposition of hazards in forestry law remains a central issue. The sources and materials encountered to this end are, however, not enough to derivate consistent standards and provisions for the assessment and mapping. Zusammenfassung: Rechtliche Vorgaben betreffend Analyse und Kartierung geologischer Gefahren sind sowohl auf internationaler als auch europischer Ebene selten. Bestimmte Protokolle zur Alpenkonvention sehen Kartierungspichten fr geologische Risiken vor, ohne allerdings materielle Vorgaben zu treffen. Im Europarecht nden sich solche Regeln lediglich in Entwrfen wie bei den prioritren Gebieten im Vorschlag einer EU-Bodenrahmenrichtlinie oder sie werden wie im Gemeinschaftskonzept zur Verhtung von Naturkatastrophen erst in Aussicht gestellt. Auf nationaler Ebene bestehen in der Regel Rechtsvorschriften im Zusammenhang mit prventiven Planungen bei Naturgefahren, wenngleich das allgemeine Problem des Nebeneinanders von mehreren gebietsbezogenen Festlegungen besteht. Als zentrale Vorschriften gelten die chenhaften Gefahrendarstellungen im Forstrecht. Das vorgefundene Material reicht jedenfalls nicht aus, um einheitliche Standards und Vorgaben fr Analyse und Kartierung ableiten zu knnen.

addition,

the

Mountain

Forests

Protocol

aims to preserve and, whenever

No coherent legal system on the

necessary, to develop or increase mountain forests as a near-natural habitat (art. 1.1) and imposes the duty of the Contracting Parties to give priority to the protective function of mountain forests (art. 6.1). The Spatial Planning and Sustainable Development Protocol7 establishes the obligation to determine the areas subject to natural hazards, where building of structures and installations should be avoided as much as possible (art. 9.2.e). The spatial planning policies also take into account the protection of the environment, in particular with regard to the protection against natural hazards (art. 3.f). 2.2. Findings

management of natural disasters can be found at either the international or European level. Also, a legal fragmentation can be detected at a national level. Therefore, the art is to lter something like a legal essence out of diverse dispersed norms, which are often only partly related to this topic and follow different legal approaches.2 This will be the attempt in the following sections. Naturally, the essay will not exceed a more or less abundant outline of the issue. 2. International law 2.1. Alpine Convention The Alpine Convention3 and its protocols

In international law, only certain provisions established in the protocols to the Alpine Convention refer to the obligation to map geological hazards. But farther-reaching, additional substantive elaborations arising out of these duties are not revealed before the respective national implementation measures. 3. European law 3.1. Soil protection law The communication from the European

are the only source of international law. The Soil Conservation Protocol4 provides for the obligation to draw up maps of Alpine areas which are endangered by geological, hydrogeological and hydrological risks, in particular by land movement (mass slides, mudslides, landslides), avalanches and oods, and to register those areas and to designate danger zones when necessary (art. 10.1). Likewise, areas damaged by erosion and land movement shall be rehabilitated in as far as this is necessary for the protection of human beings and material goods (art. 11.2).

Commission in 2002 about a Strategy for Soil Protection8 aims at the further development of
BMLFUW (ed.), Die Alpenkonvention: Handbuch fr ihre Umsetzung (2007), p. 112. Implementation analysis by SCHMID, Das Natur- und Bodenschutzrecht der Alpenkonvention. Anwendungsmglichkeiten und Beispiele, in: CIPRA sterreich (ed.), Die Alpenkonvention und ihre rechtliche Umsetzung in sterreich Stand 2009, Tagungsband der Jahrestagung von CIPRA sterreich, 21.-22.Oktober 2009, Salzburg (2010), p. 33 et seq. 6 BGBl. III 2002/233. 7 BGBl. III 2002/232..
5

For the Natural hazards prole of landslips, rock fall, avalanches and landslides, see RUDOLF-MIKLAU, NaturgefahrenManagement in sterreich (2009), p. 21 et seq. 2 For an overview regarding norms of prevention, see RUDOLFMIKLAU (fn. 1), p. 97 et seq. 3 BGBl. 1995/477. 4 BGBl. III 2002/235.
1

Seite 65

Key-note papers

Seite 66

political commitment to soil protection in order to achieve a more comprehensive and systematic protection. As soil formation is an extremely slow process, soil can essentially be considered as a non-renewable resource.9 It proceeds to mention eight main threats to soil in the EU , including
10

In particular, the EU Directive establishing a Framework for the Protection of Soil turned out to be ercely disputed.16 Since 2007, after an attenuated version failed to obtain the majority in the EU Environment Council, the future of this proposal remains uncertain. 3.2. Environmental law In the remaining European environmental laws, certain provisions about erosion can be found.17 However, there are no further provisions dealing with the topic of this essay. 3.3. Agricultural law The situation is rather similar in the area of European agricultural law. Different standards are included in the general provisions on direct payments (cross compliance)18, in which there is an obligation to maintain all agricultural land in good agricultural and environmental condition, such as those regarding soil erosion.19 In contrast, the regulation on support for rural development measures, such as afforestation (cf. art. 50.6).
21 20

3.4. Spatial planning law Regarding the quantitative aspects of soil

to collect and unify information about hazard/ risks by developing Community guidelines for hazard and risk mapping, building upon existing Community initiatives. However, these should focus on disasters with potential cross-border impact, exceptional events, large-scale disasters, and disasters for which the cost of recovery measures appears to be disproportionate when compared to that of preventive measures. Also, a more efcient targeting of Community funding25 is dealt with (3.3.1) by establishing an inventory of existing Community instruments capable of supporting disaster prevention activities, as well as by developing a catalogue of prevention measures (e.g. measures integrating preventive action in reforestation/afforestation projects). Furthermore, a a Council Civil Decision Protection establishing Community

protection, a separate communication on the topic of Planning and Environment the Territorial Dimension has been announced for a some time now. This communication should deal with rational land-use planning, as addressed by the Sixth Environment Action Programme. The
22

erosion and oods and landslides. These are intimately related to soil and land management. Floods and mass movements of soil cause erosion, pollution with sediments and loss of soil resources with major impacts for human activities and human lives, damage to buildings and infrastructures, and loss of agricultural land.
11

announced content, however, does not refer to a special relevance for the prevention of landslides. Hence, at present the only object of an integrated and sustainable management at the EU level is the ood prevention programme in transnational river areas included in the European Spatial Development Perspective (ESDP).23 3.5. Disaster law The Communication of the European Commission of February 200924 was another attempt to establish measures, based on the already existing instruments, for a Community approach on the prevention of natural and man-made disasters. Three key elements were mentioned for the Community approach: creating the conditions for the development of knowledge based disaster prevention policies at all levels of government, linking the actors and policies throughout the

In 2006, the European Commission followed suit with a Thematic Strategy for Soil Protection
13 12

and with a Proposal for a Directive establishing a framework for the protection of soil , the latter of which provides in its art. 6 for priority areas (rst draft: risk areas) with regard to landslides. The addendum landslides brought about by the down-slope, moderately rapid to rapid movement of masses of soil and rock material fell victim to the changes made by the European Parliament.14 Also, a programme of measures shall be adopted within ve years of the implementation of the Directive (art. 8). A list of common elements for the identication of areas at risk of landslides can be found in the appendix.
8

Mechanism25 deals with assistance intervention in the event of major emergencies, or the imminent threat thereof. However, a regulation on geological mass movements similar to the EU Directive on the assessment and management of ood risks27, with its ood hazard maps and ood risk maps, does not currently exist. 3.6. Findings Some relevant regulations can be found at the European level. However, only one of them, Cross Compliance, is in force and affects the topic dealt with in this essay in a rather marginal way. By contrast, the Proposal for a Directive establishing a Framework for the Protection of Soil, which has been put on hold, contemplates the designation of landslide risk areas and the establishment of
Especially the European Agricultural Fund for Rural Development, the Civil Protection Financial Instrument, LIFE+, the ICT Policy Support Programme, the Research Framework Programme. 26 Council Decision 2007/779/EC of 8 November 2007, OJ 2007 L 314/9. 27 Directive 2007/60/EC on the assessment and management of ood risks, OJ 2007 L 288/27.
25

includes in its Axis 2 some links with supporting

15

Communication from the Commission to the Council, the European Parliament, the Economic and Social Committee and the Committee of the Regions Towards a Thematic Strategy for Soil Protection, COM(2002) 179 nal. 9 Communication from the Commission to the Council, the European Parliament, the Economic and Social Committee and the Committee of the Regions Thematic Strategy for Soil Protection, COM(2006) 231 nal, Section 1. 10 Towards a Thematic Strategy for Soil Protection (fn. 8), Section 3. 11 Towards a Thematic Strategy for Soil Protection (fn. 8), Section 3.8. 12 Thematic Strategy for Soil Protection (fn. 9). 13 Proposal for a Directive of the European Parliament and of the Council establishing a Framework for the Protection of Soil and amending Directive 2004/35/EC, COM(2006) 232 nal.2.. 14 European Parliament legislative resolution of 14 November 2007 on the proposal for a directive of the European Parliament and of the Council establishing a framework for the protection of soil and amending Directive 2004/35/EC, P6_TA(2007)0509. 15 Annex I Section 5: soil typological unit (soil type), properties, occurrence and density of landslides, bedrock, topography, land cover, land use (including land management, farming systems and forestry), climate and seismic risk.

Cf. in detail NORER, Bodenschutzrecht im Kontext der europischen Bodenschutzstrategie (2009), p. 17 et seq. 17 Like the Directive 2000/60/EC establishing a framework for Community action in the eld of water policy (Wasserrahmenrichtlinie), OJ 2000 L 327/1. 18 Art. 4 et seq. Council Regulation (EC) No. 73/2009 establishing common rules for direct support schemes for farmers under the common agricultural policy and establishing certain support schemes for farmers, OJ 2009 L 30/16. 19 Art. 6 in conjunction with Annex III Regulation (EC) 73/2009; 5.1 in conjunction with Annex INVEKOS-CC-V 2010, BGBl. II 2009/492. 20 Council Regulation (EC) No. 1698/2005 on support for rural development by the European Agricultural Fund for Rural Development (EAFRD), OJ 2005 L 277/1. 21 Cf. Recital 32, 38, 41 and 44 Regulation (EC) 1698/2005. For Austrian implementation see Sonderrichtlinie zur Umsetzung der forstlichen und wasserbaulichen Manahmen im Rahmen des sterreichischen Programms fr die Entwicklung des lndlichen Raums 2007 2013 Wald & Wasser, BMLFUWLE.3.2.8/0054-IV/3/2007 idF BMLFUW-LE.3.2.8/0028IV/3/2009.
16

disaster management cycle and making existing instruments perform better for disaster prevention. In particular, the subsection Developing guidelines on hazard/risk mapping (3.1.3) is of great interest. Here, the Commission tries
Towards a Thematic Strategy for Soil Protection (fn. 8), Section 2.1, 6.1.; REISCHAUER, Bodenschutzrecht, in: Norer (ed.), Handbuch des Agrarrechts (2005), p. 491. 23 European Commission (ed.), ESDP European Spatial Development Perspective. Towards Balanced and Sustainable Development of the Territory of the European Union (1999), Section 146. 24 Communication from the Commission to the European Parliament, the Council, the European Economic ad Social Committee and the Committee of the Regions. A Community approach on the prevention of natural and man-made disasters, COM(2009) 82 nal, 23.02.2009.
22

Seite 67

Key-note papers

Seite 68

action programmes. Furthermore, a Community approach on the prevention of natural disasters sets out guidelines for the unication of hazard mapping in large-scale disasters. 4. National law 4.1. Forestry law Many times, the catchment area of mountain torrents and avalanches, as well as references to rock fall and landslip areas, are established within the national forestal spatial planning. It can even
28

4.3. Soil protection law The rules on soil protection can be divided in two categories with different aims: on the one hand, qualitative soil damage such as contaminating activities and structural damages and on the other hand, quantitative soil loss, such as soil degradation and erosion.
35

4.6. Findings In the light of the arid gain at the international and European legal level, at a rst glance the respective national systems seem to constitute the determining factor, by implementing higher-ranking guidelines or autonomously. However, norms related to the assessment and mapping of geological hazards, such as the law of natural disaster management at all40, remain fragmentated between the various regulations (Querschnittsmaterien). Relevant provisions exist, primarily in forestry law with its extensive hazard descriptions, but also marginally in spatial planning law. This fact, however, would not allow the development of uniform standards and provisions for assessment and mapping of geological hazards. 5. Conclusion
41

A convincing and coherent overall view cannot be offered. Whereas the available legal set of tools remains within the same course of action, no relevant changes coming from the international and European level are to be expected in the near future. Admittedly, the creation of uniform technical standards by all those involved as a further step towards self-regulation should be brought to mind. Anschrift des Verfassers / Authors address: Univ.-Prof. Dr. Roland Norer University of Lucerne School of Law Hofstrae 9 P.O. Box 7464 CH-6000 Luzern 7 Switzerland

The second category


36

could also be of interest for mass movements. 4.4. Spatial planning law

As a general rule, rules on areas with a higher risk of mass movements in connection with the designation of building sites
37

include the layout of forests with a protective function


29

or the extensive hazard description

or special use in

structured in risk levels.30 The protective effect of the forest especially implies the protection against natural peril and contaminating environmental inuences as well as the conservation of the soil against torrents and drift, boulders accumulation and landslides.31 Thus, forests with a direct protective function against the above-mentioned hazards could be signalised by means of an administrative act (Bannwlder). 4.2. Water law Such regulations are limited to measures for ood prevention33, although geological risks are at times also included .
34 32

grassland can be mainly found in spatial planning law. Further contents in this regard remain missing.38 4.5. Building law

Legal provisions regarding the assessment and A similar situation applies to building law. The suitability as a building site for areas with a higher risk of mass movements is not given.
34

mapping of geological hazards are tenuously sown at the international and European level. Unlikely enough, at the national level more legal provisions exist in connection with preventive planning42 for natural hazards. Here, the existing instruments partially conduct the assessment of mass movements, although the general problem of the coexistence of different area-related denitions still remains.43
39 In Austria e.g. 5.1.5 Styria Building Act (Steiermrkisches Baugesetz), LGBl. 1995/59, according to which a plot area is only suitable for building if the risks posed by ood debris accumulation, rockfall, landslides are not to be expected. From the perspective of avalanche protection see in detail KHAKZADEH (fn. 37), p. 58 et seq. 40 For Austria see e.g. HATTENBERGER, Naturgefahren und ffentliches Recht, in: Fuchs/Khakzadeh/Weber (ed.), Recht im Naturgefahrenmanagement (2006), p. 67 ; RUDOLF-MIKLAU (fn. 1), p. 57 and list 61 et seq., speaking of Kompetenzlawine. 41 WEBER/OBERMEIER, Verwaltungs- und zivilrechtliche Aspekte von Steinschlaggefhrdung und schutz, Studie im Auftrag des Bundesministeriums fr Land- und Forstwirtschaft, Umwelt und Wasserwirtschaft (2008, unverffentlicht), p. 29, suggest for Austria f.i. an extension of the competence Wildbach- und Lawinenverbauung towards other natural hazards. The political feasibility seems little realistic. 42 For Austria see in detail RUDOLF-MIKLAU (fn. 1), p. 129 et seq.; HATTENBERGER (fn. 40), p. 73 et seq. 43 For Austria see HATTENBERGER (fn. 40), p. 84 et seq.

39

In Austria e.g. the mapping of risk areas is based on 11 Austrian Forestry Act 1975, BGBl. 1975/440, in conjunction with 7.a Regulation on the mapping of risk areas, BGBl. 1976/436, including brown areas of reference, which posed other hazards than mountain torrents and avalanches, such as rock fall and landslips. Cf. JGER, Raumwirkungen des Forstrechts, in: Hauer/Nubaumer (ed.), sterreichisches Raum- und Fachplanungsrecht (2006), p. 181 et seq.; STTTER/FUCHS, Umgang mit Naturgefahren Status quo und und zuknftige Anforderungen, in: Fuchs/Khakzadeh/Weber (ed.), Recht im Naturgefahrenmanagement (2006), p. 21 et seq. 29 In Austria e.g. Forestry Development Plan (Waldentwicklungsplan) based on 9 Austrian Forestry Act 1975. 30 In Austria e.g. hazard and risks mapping (Gefahren- und Risikokarten), here geological hazard mapping (no legal basis). 31 Such as in 6.2b Austrian Forestry Act 1975. 32 Such as in 27.2.a Austrian Forestry Act 1975. 33 In Austria e.g. Section 4 of the Water Law Act 1959, BGBl. 1959/215 (Wv).
28

In Austria e.g. Water Construction Development Act (Wasserbautenfrderungsgesetz), BGBl. 1985/148 (Wv), expressly mentions the necessary protection against rock fall, mudow and landslides in the requirements for granting and allocation of federal funds to pursuit the objectives in the Act ( 1.1.1.b). 35 Cf. HOLZER/REISCHAUER, Agrarumweltrecht. Kritische Analyse des Grnen Rechts in sterreich (1991), p. 47; REISCHAUER (fn. 22), p. 477. 36 In Austria e.g. the pertinent national provisions only provide for land-use measures for soil in erosion areas; see 5 Burgenland Soil Protection Act (Burgenlndisches Bodenschutzgesetz), LGBl. 1990/87; 27 Upper Austria Soil Protection Act 1991 (Obersterreichisches Bodenschutzgesetz), LGBl. 1997/63; 7 Salzburg Soil Protection Act (Salzburger Bodenschutzgesetz), LGBl. 2001/80; 6 Styria Agricultural Soil Protection Act (Steiermrkisches landwirtschaftliches Bodenschutzgesetz), LGBl. 1987/66. 37 In Austria e.g. 37.1.a Tyrol Spatial Planning Act (Tiroler Raumordnungsgesetz), LGBl. 2006/27, according to which certain areas are excluded as building sites when f.i. there is a risk of rockfall, landslide or other gravitated natural hazards. From the perspective of avalanche protection see in detail KHAKZADEH, Rechtsfragen des Lawinenschutzes (2004), p. 37 et seq. 38 F.i. the Recommendation Nr. 52 of the Austrian Spatial Planning Conference (ROK) about preventive handling with natural hazards in Spatial Planning (2005) also puts an emphasis in oods. Cf. for Austria altogether KANONIER, Raumplanungsrechtliche Regelungen als Teil des Naturgefahrenmanagements, in: Fuchs/Khakzadeh/Weber (ed.), Recht im Naturgefahrenmanagement (2006), p. 123 et seq.

Seite 69

Key-note papers

Seite 70

KARL MAYER, BERNHARD LOCHNER

Internationally Harmonized Terminology for Geological Risk: Glossary (Overview) Internationale Harmonisierung der Fachterminologie fr geologische Risiken: Glossar (berblick)
Summary: Purpose and motivation for this project are the difculties traditionally encountered when using or dening mass movements terms in scientic papers. This results in different methods and concepts being used by geological agencies and nally leads to misunderstandings and problems in cooperative international projects. In order to tackle that complexity and ambiguity, found not only in the German-speaking geology, but generally throughout Europe, a multilingual glossary shall be created. This glossary aims at an international harmonization by providing the user with a selection of ofcial terms used by the geological agencies in a specic country and by setting relations to similar terms employed in other countries. The resulting harmonized terms and denitions should be made available to all partners and to the general public on the internet through the Bavarian Environment Agency homepage. The rst step is to design and implement the technical infrastructure required to store and query the terms. For this purpose, a relational database management system will be used as a back-end.

Zusammenfassung: Ausgangslage und Motivation fr dieses Projekt ist die schon traditionelle Problematik der unterschiedlichen Verwendung und Denition der Begrifichkeiten in der Fachliteratur zum Themenbereich Massenbewegungsprozesse. Dies hat zur Folge, dass die Arbeitsweisen der Experten in den verschiedenen geologischen mtern in den Projektpartnerlndern nicht einheitlich sind und es daher immer wieder zu Missverstndnissen und Schwierigkeiten bei der Abstimmung gemeinsamer Projekte kommt. Aufgrund dieser Komplexitt und der Unklarheit, die speziell im deutschsprachigen Raum, aber auch europaweit, besonders im Hinblick auf die Klassikation der Massenbewegungen existiert, soll ein mehrsprachiges Glossar erstellt werden, in welchem im Sinne der internationalen Harmonisierung in Absprache mit den einzelnen Projektpartnerlndern die von den jeweiligen geologischen mtern verwendeten administrativen Begriffe eingestellt und in Beziehung gesetzt werden. Das gesamte Projekt gliedert sich grundstzlich in einen technischen und einen inhaltlichen Teil, wobei die erste Projektphase vom technischen Bereich bestimmt wird. Da die harmonisierten Begrifichkeiten und Denitionen fr alle beteiligten Lnder und auch fr eine breitere ffentlichkeit zugnglich gemacht werden soll, wird eine relationale Datenbank erstellt, in welcher die Inhalte logisch verknpft werden und welche zu Projektende in die LfU-Homepage integriert wird.
combination for one language and one country. 1. Requirements for the relational database Before the actual database is deigned, it is essential to assess the exact requirements for the glossary. This eases the following conceptional work a lot and minimizes time-consuming adjustments and changes to the model later on. First a list of attributes needed for a single glossary term as well as a type for those attributes (e.g. numbers, text, keys etc.) is to be dened. The type of attribute determines which relations can be saved in the database and what kind of information can be queried using them. Every attribute corresponds at least to one column in the main glossary table. The unique language to which a term is assigned is a fundamental attribute in a multilingual glossary. Because of the panEuropean character of the glossary, it is necessary to specify the languages more precisely by linking them to a specic country, resulting in a unique It is particularly relevant for this project, as the usage of a term varies greatly within a language depending on the region where it is used, as it is the case for German (Germany, Austria, Switzerland). Easy and intuitive queries are essential for the usability of the glossary. Although the user friendliness mostly depends on the graphical user interface and is hard to control through the database design, there are still aspects that need to be considered in conception. It is important to determine what possible queries will be offered to the user (e.g. a search by synonyms, case and special character insensitive searches, etc.) and to adapt the database design accordingly. Editing and adding glossary terms after the initial import should also be possible and requires saving metadata for each entry, e.g. time and date of the creation or the last edit of a term. Using that information, it is easy to reconstruct the history of an entry at a later point in time.

Seite 71

Key-note papers

Seite 72

tdtaTerm PK idterm idworkowstatus metacreator metaowner idreadaccess idwriteaccess deleted metamasterlang metalastedit

tdtaTermLng PK, FK1 PK idterm lang term description


Fig. 1: Example of a multilingual glossary where each term has exactly one translation in each other language. The primary key of the language table ('tdtaTermLng') is dened by its ID and language Abb. 1: Beispiel eines mehrsprachigen Glossars, in dem jeder Begriff genau eine bersetzung fr jede weitere Sprache hat. Der Primrschlssel der Tabelle mit dem Textinhalt ('tdtaTermLng') ist somit ber ID und Sprache deniert.

The nomenclature used throughout the database follows a simple naming convention. Depending on the function or content of a particular table, its name is prexed differently. The prex tdta- stands for tables in which actual data is being stored, tkey- is used for key tables (key attributes can only take a value from a predened set of keys) and trel- for relation tables. Unique IDs are prexed with id- and meta-attributes with meta-. For most of the tables the multilingual concept required by the direct translation provides a second table with an identical name and the sufx -Lng. Those language tables hold the text values of the different glossary terms. The rst section is the core of the database, with its element tables tdtaElement and tdtaEleGlossarTerm. The glossary terms are stored in the latter, whereas the main element table holds additional information related to the system and not to the glossary itself (mostly through the usage of foreign keys).
FK4 FK2 FK1 FK5 FK3 PK

tdtaElement idelement

tdtaEleGlossarTerm PK,FK1 idelement term reference idtopic idlang idcountry searchterm searchsynonyms

elementtype idworkowstatus metaowner metacreator idreadaccess idwriteaccess deleted metamasterlang

FK3 FK4 FK2

Finally, the database should, to some

meters. The relation to rock fall (i.e. similar meaning) would be a looser one. The relations between cliff falls, block falls, boulder falls and Felssturz, Steinschlag, Blockschlag could be dened in a similar manner.

extent, be expandable if future needs for extensions or additional functions arise. 1.1 Relations The classical approach followed by most

tdtaEleGlossarTermLng tdtaElementLng PK,FK1, PK,FK2 idelement lang title summary metacreated metalastedit metatranslator PK,FK1, FK2 PK,FK1 idelement lang title description

(Note: the values used above are examples and do not necessarily match any ofcial values) 1.2 Database model This chapter describes in detail the different sections of the database. For the purpose of clarity, the database was divided into four sections or areas which correspond to a set of interrelated tables. The following diagram shows the relations between those sections.
Glossary
Terms Relations Translation tables

glossaries is a single translation layer; a direct translation of each term into exactly one term of another language. This corresponds to a 1: n relation between the entities (i.e. glossary terms) in an entity-relationship model (ERM). Such a direct translation supposes an equivalence of the terms denition and meaning. In this new glossary, the relations between the different terms should be dened solely by their technical meaning, resulting in two possible relations: same meaning or similar meaning. A direct translation is still required in order to provide the user with the exact translation of a denition in his language. Following example should help clarifying The English term rock fall is usually
Metadata
Workow History

Fig. 3: Main tables Abb. 3: Haupttabellen

tkeyCountry PK idcountry tdtaEleGlossarTerm PK,FK1 idelement term reference idtopic idlang idcountry searchterm searchsynonyms
tkeyLang PK idlang langsort

tdtaEleGlossarTermLng PK,FK1 PK,FK2 idcountry lang countryterm tkeyLanguage PK idlanguage lang languagesort

countrysort

Auxiliary
Key tables Relation tables

tkeyLangLng PK,FK1 PK,FK2 idcountry lang langterm idlanguage

the concept of meaning vs. denition: translated into Felssturz or Bergsturz in German, but that translation usually doesn't consider the effective volume transported. However, if the technical meaning is taken into account, Bergsturz, which corresponds to a minimum volume of 106 cubic meters, would have the same meaning as rock avalanche, also characterized by volume values above 106 cubic
Fig. 2: Overview of the database model components Abb. 2: bersicht ber die Komponenten des Datenbankmodells

FK3 FK4 FK2

User Management
Users & groups Permissions

tkeyTopic PK idtopic topicsort

tkeyTopicLng PK,FK1 PK,FK2 idtopic lang topicterm

tkeyLanguageLng PK,FK1 idlanguage PK lang languagesort

Fig. 4: Auxiliary tables Abb. 4: Behelfstabellen

Seite 73

Key-note papers

Seite 74

For each term, following elds are available: 'term': the actual text value (direct translation using the -Lng table) reference: source of information and date 'idlang' and 'idcountry': foreign keys pointing to a unique combination of language/country 'idtopic': foreign key specifying the topic of this term 'searchterm' and 'searchsynonyms': used for insensitive searches 'picture': paths to pictures depicting a term tkeyWorkowStatus PK idworkowstatus workowstatussort

The auxiliary tables are mainly key tables dening the different languages, countries and topics used in the main table. They also contain the relation table used to specify relations between terms based on a relation code (similar or same). Metadata is partly stored in the tdtaElement table using foreign keys. Those keys point to external metadata tables such as tkeyWorkowstatus or tdtaUser, where, for example, information about the status, author or owner of an element are dened. tdtaHistory works similarly to a log by saving all actions performed on a specic tkeyLanguage PK idlanguage lang languagesort PK FK4 FK2 FK1 FK5 FK3 PK

tdtaElement idelement elementtype idworkowstatus metaowner metacreator idreadaccess idwriteaccess deleted metamasterlang PK PK

tkeyPermissionLevel idpermissionlevel permissionlevelsort

tkeyPermissionLevelLng tdtaGroup idgroup groupname PK,FK1 idpermissionlevel PK lang permissionlevelterm

tkeyWorkowStatusLng PK,FK1 idworkowstatus PK,FK2 lang workowstatusterm

tdtaUser iduser username password email organisation fullname inactive superadmin lastlogin loginip maingroup

description

tdtaElement PK idelement elementtype idworkowstatus metaowner metacreator idreadaccess idwriteaccess deleted metamasterlang tdtaEleGlossarTerm PK,FK1 idelement term reference idtopic idlang idcountry searchterm seyrchsynonyms tkeyLanguageLng PK,FK1 idlanguage PK lang languagesort FK1

trelUserGroup PK,FK2 iduser PK,FK1 idgroup

FK4 FK2 FK1 FK5 FK3

FK3 FK4 FK2

Fig. 6: User and group management Abb. 6: Benutzer- und Gruppenverwaltung

tkeyelementActionLng PK,FK1 idelementaction PK lang FK2 elementactionterm idlanguage

element, which can be displayed as a list to an authorized user. Finally, user and group management denes the group(s) a user belongs to and which read/write rights a group or a specic user owns (through the tdtaElement table) 1.3 Data capture and import The primary data capture is done via an Excel table with a predened format. This table is used as an interface to import data records in the database. The person responsible for lling out this table must ensure that the relations between the terms are set correctly. Other errors, such as

duplicate IDs, can be handled to some extent by the database itself. The integration of the database into the homepage from the Bavarian Environment Agency (LfU) and a graphical user interface to manually add or edit single terms is planned in the nal stage of the project. 2. Contents of the glossary In view of a different use of landslide-terms in the European countries, a multilingual glossary can help to improve the collaboration between the experts. Also, progress concerning the comparability of the methods dealing with geological hazards in the several countries is to be achieved.

tdtaUser PK iduser username password email organisation fullname inactive superadmin lastlogin loginip maingroup PK FK2 FK1 tdtaHistory idhistory idelement lang iduser logdatetime info idelementaction

FK1

tkeyElementAction PK idelementaction

FK1 elementactionsort idhistory

Fig. 5: Metadata tables Fig. 5: Metadata tables

Seite 75

Key-note papers

Seite 76

In general, the glossary implies terms and denitions to landslides and corresponding maps, considering danger, hazard and risk caused by several kinds of geological hazards. Due to the alpine character of the project, the glossary contains all the languages spoken in the Alpine region plus English and Spanish for two additional European countries dealing with geological hazards. Therefore, the glossary consists of the following six languages: German Germany, Switzerland, Austria (three different lists) Italian Italy French France Slovenian Slovenia Spanish Spain (Castilian and Catalan) English United Kingdom 2.1 Basic list for Germany For the development of such a glossary, it is necessary to create a basic list in which all the desired terms and denitions are included. Therefore a table with 92 terms and denitions for geological hazards (in German) was drafted. Based on this, the other language lists were developed. More information on the approach of this Harmonization is available in chapter 3.2. In order to facilitate this process, all the terms are structured in different topics. This classication is very useful for simplifying the comparability between the languages. For example, its much easier to get the English term for Stauchwulst if the English expert knows that you are searching for an accumulation term. This topical limitation helps the translator to get the several experts on the right track. The basic list is structured into the Accumulation (Ablagerungen - z.B. Schuttkegel) following topics:

General geomorphology (Allgemeine Geomorphologie - z.B. Grat) General (Allgemeines - z.B. Primrereignis) Fracture forms (Anbruchformen - z.B. Bergzerreissung) Path of movement (Bewegungsbahnen z.B. Sturzbahn) Flow process slow (Flieprozess langsam - z.B. Soliuktion) Flow process rapid (Flieprozess schnell - z.B. Blockstrom) Flow process very rapid (Flieprozess sehr schnell - z.B. Murgang) Risk (Gefahr-Gefhrdung-Risiko - z.B. Restrisiko) Maps (Karten - z.B. Gefahrenkarte) Classication processes (Klassikation Prozesse - z.B. Sturzprozess) Measures (Manahmen - z.B. aktive Manahmen) Slides combined (Rutschprozess Kombinierte Rutschung - z.B. Rutschung mit kombinierter Gleitche) Slides rotational (Rutschprozess Rotationsrutschung - z.B. Rotationsrutschung) Slides translational (Rutschprozess Translationsrutschung - z.B. Translationsrutschung) Landslide dynamics (Rutschungsdynamik z.B. aktuelle Hangbewegung) Landslide features (Rutschungsmerkmale z.B. Rutschungkopf) Falls (Sturzprozess Bergsturz - z.B. Bergsturz) Falls (Sturzprozess Blockschlag - z.B. Blockschlag) Falls (Sturzprozess Felssturz - z.B. Felssturz)
id

Falls (Sturzprozess Steinschlag - z.B. Steinschlag) Subrosion (Subrosionsprozess - z.B. Doline)


term lang country denition

As mentioned above, the different terms lists will be integrated in the ofcial homepage of the Bavarian Environment Agency in a nal step. Therefore, the terms are collected in a predened
reference topic same_ rel similar_ rel

2016

Abusslose Senke

de

DE

Senke ohne natrlich mglichen oberirdischen Wasserabuss. In einem uviatil geprgten Relief stellt sie eine Anomalie dar, die u.U ein Hinweis auf Hangbewegungen sein kann Schutzmanahme, die dem Naturereignis aktiv entgegenwirkt, um die Gefahr zu verringern oder um den Ablauf eines Ereignisses oder dessen Eintretenswahrscheinlichkeit wesentlich zu verndern. Neben den klassischen, punktuellen technischen Schutzmanahmen wie zum Beispiel Sttzmauer oder Felsanker sind auch chendeckende Manahmen im Einzugsgebiet, beispielsweise Aufforstungen oder Entwsserungen, dieser Kategorie zuzuordnen. Hangbewegung die zum Zeitpunkt der Aufnahme aktiv oder bezglich ihres Alters fr die Untersuchungen relevant war. Hangbereich aus dem eine Hangbewegung ihren Ausgang nimmt. Der Auslser/Anlass fr das Versagen eines Hanges liegt in externen Faktoren. Dieser lst eine quasi sofortige Reaktion aus, die ihrerseits wieder Auslser fr die nchste Reaktion sein kann (Kausalittskette). Die Auslser reduzieren zum Beispiel die Festigkeit der im Hang anstehenden Gesteine. Mgliche Auslser knnen sein: Niederschlge, Schneeschmelze, Frost- Tauwechsel, Erdbeben, Menschlicher Eingriff. ffnungen an der Erdoberche ber die Oberchenwasser in den Untergrund eindringt. Hangbewegung mit groem Volumen und hoher Dynamik, die oftmals dafr sorgt, dass die Massen am Gegenhang weit aufbranden. Volumen > 1.000.000m.

LfU Bayern

Allgemeine Geomorphologie

2066

aktive Manahmen

de

DE

LfU Bayern

Manahmen

2070

Aktuelle Hangbewegung Anbruch

de

DE

LfU Bayern

Rutschungsdynamik Anbruchformen

2029

de

DE

LfU Bayern

2027

Auslser

de

DE

LfU Bayern

Allgemeines

2092

Bachschwinde (Ponor)

de

DE

LfU Bayern

Subrosionsprozess/Allgemein

2079

Bergsturz

de

DE

LfU Bayern

Sturzprozess Bergsturz

Fig. 7: Extract of the Basic-Terms-Table in German Abb. 7: Auszug aus der Deutschen Begriffstabelle

Seite 77

Key-note papers

Seite 78

Excel table with a unique ID for each term. This ID is used to establish the relations between the different languages and also to integrate these in the relational database. Fig. 6 shows an extract of this Excel table with the basic terms from Germany. 2002 2.2 Harmonisation of terms and methods A glossary will facilitate transdisciplinary and translingual cooperation as well as support the harmonization of the various methods (www.adaptalp.org). Striving for Harmonization of regional terms and methods seems to be a guiding principle not only in WP 5 of the AdaptAlp project but in multiple European cooperation projects. In the literature, a lot of denitions are used for the term harmonization. According to the business dictionary, harmonization is an adjustment of differences and inconsistencies among different measurements, methods, procedures, schedules, specications, or systems to make them uniform or mutually compatible (www.businessdictionary.com). This denition implies some important points which are mentioned as main goals in many projects supported by the EU. The adjustment of differences and the achievement of compatibility also play a major role in work package 5: AdaptAlp will evaluate, harmonise and improve different methods of hazard zone planning applied in the Alpine area. The comparison of methods for mapping geological and water risks in the individual countries (www.adaptalp.org) will be brought into focus. Concerning the development of the 2011 multilingual glossary for geological hazards, the Harmonization is implemented by the following approach. 2010 2007 2008 2009 2006 2005 2004 id 2001

German term Stauchwulst denition


Wulst aus Gesteinsmaterial. Sie tritt vor allem an der Stirn einer Rutsch- oder Kriechmasse auf Murablagerung am seitlichen Rand des Murkanales Gelnde, in dem weitrumig Blcke und Gesteinsschollen verteilt sind. Herkunft der Blcke in der Regel von groen Fels- od. Bergstrzen, aber auch von Talzuschben. Unter Murkegel sind kegelfrmige Ablagerungen v.a. an Gerinnen zu verstehen, deren Bschungswinkel meist mehr als 8-10 betrgt Sie sind oft noch durch die typischen dammartigen Wlste entlang des Randes eines ehemaligen Murstromes gekennzeichnet. Schwemmkegel weisen im Gegensatz zu Murkegeln meist Bschungswinkel von weniger als 10 auf, grere Geschiebeblcke fehlen. Schuttkegel entstehen v. a. durch Steinschlag. Sie lagern sich an Steilwnde und dort bevorzugt im Bereich von Steinschlagrinnen an Gelnde, das durch unruhige Morphologie (weiche Formen) gekennzeichnet ist. Ablagerung infolge eines Sturzprozesses. Ablagerung infolge eines Rutschprozesses Teilweise im Verband bendlicher Gesteinskomplex, der als ganze Scholle abrutscht. Einzelblock >1m, infolge eines Sturzprozesses.

English reference
LfU Bayern

topic
Ablagerungen

topic
accumulation

term
toe????

denition
accumulation at the toe/foot of the main body. accumulation at ank of the main body.

Murwall

LfU Bayern

Ablagerungen

accumulation

2003

Blocklandschaft

LfU Bayern

Ablagerungen

accumulation

Bloc Landscape????

Area in which blocs are shared spacious. Bloc are comming from rock collapses, block falls or sags.

Murkegel, -fcher

LfU Bayern

Ablagerungen

accumulation

Coned accumulation espacially at channels with a naturel slope of 8-10.

Schwemmkegel, -fcher

LfU Bayern

Ablagerungen

accumulation

Coned accumulation espacially at channels with a naturel slope less than 10 and with no big blocs.

Schuttkegel

LfU Bayern

Ablagerungen

accumulation

coned debris/ detritus????

"coned debris/detritus" are caused by rock falls. They accumulate at the rock face.

Buckelche Sturzmasse Rutschmasse Rutschscholle Sturzblock

LfU Bayern

Ablagerungen

accumulation

undulating area????

Area which is characterized by undulating morphologie. Accumulation caused by a fall process. Accumulation caused by a slide process.

LfU Bayern LfU Bayern

Ablagerungen Ablagerungen

accumulation accumulation sliding bloc/clod/ massif????

LfU Bayern

Ablagerungen

accumulation

A coplex of rocks which is sliding as one bloc/clod/ massif. One bloc (<1m) of an fall process.

LfU Bayern

Ablagerungen

accumulation

(fall) bloc????

Fig. 8: Extract of the suggested-terms list for England Abb. 8: Auszug aus der vorgeschlagenen Begriffsliste fr England

Seite 79

Key-note papers

Seite 80

2.2.1 Basic rules In order to tackle the complexity and ambiguity, found not only in German-speaking geology, but generally throughout Europe, a multilingual glossary shall be created. This glossary aims at international harmonization by providing the user with a selection of ofcial terms used by the geological agency in a specic country and by setting relations to similar terms employed in other countries. Unlike many other glossaries, which are more like dictionaries working with direct translations; this glossary consists of terms and denitions which are used by the ofcial agencies from the involved countries. So the big difference from many other word lists is the way of getting the topics. 2.2.2 Data acquisition Basically the data acquisition is made during

short visits in the involved countries. Building on the German basic list, in these talks term after term is discussed with the respective person responsible. With regard to linguistic problems, each Harmonization is carried out with the help of native speakers who also be well versed in the thematic of geological hazards. The terms are related in the following three forms: Same meaning (the term has the same meaning in both languages) Similar meaning (the term has a similar meaning in both languages) Not existing (no term with the same or similar meaning exists) To facilitate the harmonization process, in the run-up to the visits, several national literature lists with suggested terms are worked out with the native speakers. These lists also contain short descriptions of the desired expressions and they

are sent to the responsible persons for orientation and preparation. Furthermore, Fig. 7 shows an extract of the suggested terms list for England. A picture paints a thousand words, therefore also pictures and illustrations are used within the talks. 2.2.3 Data preparation and presentation Concerning the data preparation, the main issues are already described in the technical description above. The central point to fully exploit the possibilities of the database structure is the correct setting of the relations between the different terms (over the ID). Regarding to the data presentation, at this stage of the project no nal results can be shown. As mentioned in the introduction of this article, the main output of the project will be an online glossary which is linked to the homepage of the Bavarian Environment Agnecy (LfU). The layout of this web page should be clear and simple for everyone to use. Therefore existing online glossaries are compared and bestpractice examples are pulled out as inspiration. Fig. 8 shows the Inter Active Terminology for Europe glossary from the European Union which approximately fulls the desired criteria for the geological hazard glossary. 3. Conclusion As mentioned in the introduction, this article presents no nal results because the project runs until February 2011. Nevertheless, provisional results, theoretical and practical approaches could be shown. The database model presented in this article fulls all requirements stated by a multilingual glossary focusing on mass movements and other geological hazards. The multilingual concept provides the user with a direct translation of a term in a foreign language and sets relations to other terms based on its

technical meaning. Although the structure of the model may seem complex, the multiple functions offered by external tables and the stronger data integrity fully compensate for a higher level of complexity. To achieve this complexity, not only the structure of the relational database but also the contents should satisfy the guidelines. The term Harmonisation is playing a central role in the work for the glossary where the contents are concerned. Only terms, which are ofcially used by the regional responsible agencies, are registered in the glossary and the relations between the different expressions are also dened by several experts. The topics in this glossary are not dened by a translation agency, which undoubtedly would have the linguistic ability but not the specialist background. Due to this approach, every involved country or region gets the chance to determine the terms and denitions they use and that procedure improves the overall result. The connection to the LfU Homepage ensures accessibility for all interested persons. This is an important contribution to one of the main goals of the whole project, namely the improvement of the cooperation by the European countries in dealing with geological hazards.

Anschrift der Verfasser / Authors addresses: Karl Mayer Bavarian Environment Agency (LfU) (Ofce Munich) Lazarettstrae 67 80636 Munich GERMANY Bernhard Lochner alpS Centre for Natural Hazard and Risk Management Grabenweg 3 6020 Innsbruck - AUSTRIA

Fig. 9: Screenshot of the online Inter Active Terminology for Europe from the EU (Source: http://iate.europa.eu) Abb. 9: Screenshot de online Inter Active Terminology for Europe der EU (Quelle: http://iate.europa.eu)

Seite 81

Hazard assessment and mapping of mass-movements in the EU

Seite 82

Introduction In Austria there are several public organizations ([12] HBL et al. 2009) involved in the assessment of rapid gravitational mass movements such MICHAEL MLK, THOMAS SAUSGRUBER, RICHARD BK, ARBEN KOCIU as rock falls and landslides. Inventories of such events are maintained by the Austrian Torrent and Avalanche Control (WLV) and the Geological Survey of Austria (GBA) apart from independent assessments done by the national railway and road administrations. different On the level of the federal administrations, approaches to documenting and/or

but it is not digitally available. And then there are states that can rely on a lot of digitally available data and are working on generating landslide susceptibility maps. The following provides a short summary about the efforts in the federal states. Mass-movement inventories in Austria Since 1978 the Geological Survey of Austria has been gathering and displaying information (e.g. documents, photos, inventory maps) about gravitational mass movements and other hazardous processes. Due to the increasing amount of data, the Department of Engineering Geology of the Geological Survey of Austria developed a complex data management system called GEORIOS. It consists of a Geographical Information System (GIS), which is the basis for the digital storage and display of data and overlay of different data types. Additionally the data management system consists of a relational data base, which manages additional thousands of meta-information (documents, photos etc.).

Standards and Methods of Hazard Assessment for Rapid Mass Movements (Rock Fall and Landslide) in Austria Standards und Methoden der Gefhrdungsanalyse fr schnelle Massenbewegungen (Steinschlge und Rutschungen) in sterreich
Summary: This presents the Austrian approach for the documentation and prediction of landslides and rock falls from various inventories (GEORIOS - Geological Survey, Torrent and Avalanche Control, inventories of the federal states) via the hazard zone planning leading to the development of process related susceptibility maps. The different legal obligations of the respective organizations leads to different results regarding the type, the extent and the quality of the expertise. Zusammenfassung: Der sterreichische Weg zur Erfassung von historischen bzw. zur Vorhersage von zuknftigen Steinschlagprozessen und Rutschungen von den verschiedenen Ereigniskatastern (GEORIOS Geologische Bundesanstalt, Wildbach- und Lawinenkataster, Ereigniskataster der Lnder) ber die Gefahrenzonenplanung bis zur Erstellung von Prozessdispositionskarten wird dargestellt. Dabei sind unterschiedliche gesetzliche Verpichtungen und Zielsetzungen fr die damit befassten Organisationen mageblich fr die Art, den Umfang und die Qualitt der erreichten Aussagen.

forecasting such mass movements are being followed. These organizations deal with those hazards using different approaches (method and target). As there are no legal instructions in Austria as to how to deal with the evaluation of mass movements, the federal states all follow a different course of action. Also, the status of available historical data is very different in the individual states. In some of the federal states, almost no data is available, others have collected a lot of data

Fig. 1: Inventory of mass movements in Austria (source Geol. B.-A.: www.geologie.ac.at) Abb. 1: Karte der Massenbewegungen in sterreich (Quelle: Geol. B.-A.: www.geologie.ac.at)

Seite 83

Hazard assessment and mapping of mass-movements in the EU

Seite 84

The

database

includes

detailed

information about the mass movements (geology, hydrology, geometric and geographical data, studies or tests carried out, mitigation measures) and the source of information (archives, etc.), and also information about who carried out the eld work and added the data into the database. There are already in the 22,000 database. mass The
Fig. 4: WLV-Inventory of mass movements in Austria (source: www.die-wildbach.at) Abb. 4: Ereignisdatenbank der WLV (Quelle: www.die-wildbach.at)

movements

stored

compilation of a part of the mass movements in Austria is publicly accessible via the internet (www.geologie.ac.at) in German and English. However, the web application includes only events such as slides, rock falls, or more complex mass movements which have been published already in the media or the internet and are freely available for everyone ([16]KOCIU et al 2007). An engineering geological database, as well as a bibliographical database is also included in the GEORIOS system. In cooperation with the Geological Survey of Carinthia, the Geological Survey of Austria has created not just one inventory map, but a level of information, as is explained in the following ([17] KOCIU et al 2010): Level of information: Process index map, map of phenomena (Prozesshinweiskarte, Karte der Phnomene): These kinds of maps can have different scales (1:50,000 and bigger) and can be of varying quality with information about process areas as phenomena of mass movements that have already happened. The event inventory (Ereigniskataster) records only those processes for which an event date is known (5Wquestions), it is independent of a scale and can contain processes without information on location. In Carinthia, a digital landslide inventory was created with historical events of the last 50 years ([1] BK et al 2005).
Fig. 3: Event map of Carinthia (brown landslides; blue earth ow; red rock fall; green earth fall) Abb. 3: Ereigniskarte von Krnten Fig. 2: Event inventory of Carinthia with 5W-questions and quality remarks MAXO (M-sure; A-estimate; X-uncertain; O-unknown) Abb. 2: Ereignisdatenbank von Krnten mit 5W-Fragen und Qualittskriterien MAXO

The Austrian Torrent and Avalanche Control (WLV) also maintains an inventory covering torrential oods, avalanches, landslides and rock falls the so called Wildbach- und Lawinenkataster. Standards of susceptibility/hazard assessment in Austria Because of the lack of a regulatory framework or technical standard concerning landslides and rock falls in Austria - only the course of actions concerning oods, avalanches and debris ows are regulated by law (ordinance of hazard zone mapping,[33] RUDOLF-MIKLAU F. & SCHMIDT F., 2004) - the federal states all follow a different course of action. For example, in Vorarlberg risk maps (susceptibility map, vulnerability map, risk map) were produced in the course of a university dissertation ([34] RUFF, 2005). For modelling, bivariate statistics (for landslides) and cost analysis (for rock fall) were used, working with a 25x25m raster. The susceptibility, meaning spatial susceptibility, is presented in 5 classes (very low, low, medium, high, very high). The inventory map is included in the susceptibility map. On the other hand, the local department of the Austrian Service for Torrent and Avalanche Control (WLV) creates hazard maps within the hazard zoning plan. In Upper Austria, Lower Austria, Burgenland and Carinthia, different approaches

are chosen to develop susceptibility maps (different scales, processes) derived from existing data sets and maps ([30] POSCH-TRTZMLLER G., 2010): Main focus of Burgenland is concentrated on shallow landslides with an annual rate of movement of 1-2cm. For the prediction of landslide susceptibility based on morphological and geological factors, the method called Weights of Evidence was chosen ([15] KLINGSEISEN et al., 2006). Three (respectively 4) hazard zones were classied ([high Hazard], hazard, hazard cannot be excluded, no hazard, [15] KLINGSEISEN et al., 2006). In Lower Austria up until now the susceptibility maps have been created using a heuristic approach based on geological expertise, historical data and interpretation of DEM and aerial photos. Three to ten classes of susceptibility are delineated at a scale ranging from 1:50,000 to 1:25,000 ([36] SCHWEIGL & HERVAS 2009). To offer assistance for the municipalities in land-use planning, landslide susceptibility maps were generated for the major settled areas in Upper Austria (O). For each type of mass movement, the priority, which is a susceptibility class, was evaluated on the basis of the intensity and the probability of an event. The priority was classied in 3 stages (high medium low; [18] KOLMER, 2005). As these maps include the intensity and the frequency of mass movements, they can be called hazard maps by denition. Nevertheless it has to be

The

inventory

map/event

map

(Ereigniskarte) contains only information about processes for which an event date is known (5Wquestions: What, When, Where, Who, Why). The symbols are correlated to process type and magnitude (triangle small events, pentagon great events).

The thematic inventory map contains only information related to a type of process, categorized according to the quality of the data.

Seite 85

Hazard assessment and mapping of mass-movements in the EU

Seite 86

taken into account that the method of generating these maps included neither eld work nor remote sensing techniques. The method of assessment is based solely on geological expertise. Using the digital geological map of Carinthia (1:50,000), the inventory map of mass movements (landslides and rock falls), DEM (10m x10m raster), land-use and lithologicalgeotechnical characteristics of bedrock and unconsolidated sediments, process-related susceptibility maps for Carinthia were generated in a collaboration of the Geological Survey of Austria (GBA) and the Geological Survey of Carinthia at a scale of 1:200,000 ([1] BK et al., 2005). Of course these maps still lack information about intensity and recurrence period or probability of occurrence. Due to the imprecision of input data used, the accuracy of predictions regarding the susceptibility for rapid mass-movements based on maps like the ones mentioned above is limited.

For a small study area in Styria, the Geological Survey of Austria generated a susceptibility map for spontaneous landslide (soil slips and earth ows) at a scale of 1:50,000 using neural network analysis ([35] SCHWARZ et al., 2009). Any susceptibility class is not a ranking of the degree of slope stability, but a description of the relative propensity/probability of a landslide of a given type and of a given source area to occur.). At the Geological Survey of Austria (GBA), susceptibility maps in different scales and with different methods (heuristic approach, neural network analysis) have already been generated. ([17] KOCIU et al., 2010, [21] MELZNER et al., 2010, [38] TILCH et al., 2009, [39] TILCH et al., 2010, [40] TILCH et al., 2010, [41] TILCH et al 2009). Legal situation, requirements by the law, The key feature for susceptibility/hazard responsibility of different authorities mapping is a good documentation of historic

events, a thorough mapping of the phenomena involved and an accurate interpretation of the failure with the subsequent processes. The WLV is legally obliged to do an inventory of all events regarding natural hazards, such as torrential processes, avalanches, rock-falls and landslides in the so called Wildbach- und Lawinenkataster WLK ([8] Forstgesetz 1975). The GBA denes its very own tasks, among others: the assessment and evaluation of geogenically induced natural hazards". These inventories (WLV, GBA, geological surveys of provinces like Carinthia) are established to guarantee a complete documentation of processes and events that can eventually endanger infrastructure and/or people. The data collected in the inventories allow for better information and further evaluation of where, when, how often and with which intensities those events took place. These inventories can form an important basis for the elaboration of hazard maps and related hazard zones, which give the authorities good evidence to optimize land-use planning and avoid areas that tend to be exposed to natural hazards. For already developed areas, the assessment of the type of process, magnitude, run-out, location, frequency etc. allows for a better priority-rating and design of mitigation measures. The elaboration of hazard zone maps

issuing building permits to consult an expert to evaluate the hazard for the planned construction site explicitly, otherwise the community can be excluded from public funding for the nancing of mitigation measures in the future. Standards, guidelines, ofcial and legal documents Several standards issued by the IAEG (Internat. Association of Engineering Geology UNESCO Working Party of World Landslide Inventory [42] to [47]) exist for the documentation and classication of landslides. Furthermore, for the documentation of landslide and rock fall events (avalanches and torrential processes are covered as well) there is a short course of the Universitt fr Bodenkultur Wien, Dpt. f. Bautechnik und Naturgefahren, Inst. f. Alpine Naturgefahren, which certies documentalists for those processes. For the assessment and evaluation of rock fall processes and the design of protection measures an Austrian Standard is currently under development ([28] ONR 24810: Technischer Steinschlagschutz). State of the art in the practice The code of practice is to be brought up to the state of the art due to the absence of binding standards. The state of the art according to the Wasserrechtsgesetz WRG 1959 12a(1) is dened in Austria as the following: The use of modern technological methods, equipment and modes of operation with proven functionality which represent the status of progress based on relevant scientic expertise. Rock fall hazard assessment The state of the art regarding the assessment and evaluation of hazard for rock fall processes can

([8] Forstgesetz 1975 and [2] BGBl. 436/1976) for potentially endangered zones caused by natural hazards (except ooding by rivers and earthquakes, which are done by other authorities) for all communities is the task of the Austrian Torrent and Avalanche Control (WLV). The delineation of potential emmissionzones of rapid mass movements, such as rock falls and landslides, are not mandatory and therefore can be illustrated as brown hazard indication
Fig. 5: Susceptibility map for spontaneous shallow landslide at Gasen Haslau ([35] Schwarz et al 2009). Abb. 5: Dispositionskarte fr spontane, achgrndige Rutschungen im Bereich Gasen-Haslau ([35]Schwarz et al 2009).

areas by the WLV. The legal implication of these indication areas lies in the obligation of the authorities

Seite 87

Hazard assessment and mapping of mass-movements in the EU

Seite 88

be described by the following workow. The methods to be applied are just roughly described, for a detailed description see the cited literature. Depending on the objective of the assessment, the tools to be applied may vary in respect to the scale of the result, being more coarse at regional scale and detailed at slope-scale. Standard procedure for the assessment of rock fall hazards (best practice): Preparation Denition of the boundaries of the project area in compliance with the stakeholder Acquisition of basic data (topograc maps, geology, land use, literature, studies etc.) Collection of historic event information (written and oral) Field work: Collection of properties of the forest (if relevant), identication (by eld work and/ or according to e. g. [12] JABOYEDOFF 1999) and Evaluation of detachment areas description of discontinuities (type, dip/direction, opening, lling ), properties of rock mass, relevant failure mechanisms, probabilistic distribution of joint-bordered rock bodies Scree slopes: block-size distribution (statistics) Analysis of rock fall processes ([22] MELZNER et al 2010, [23] MELZNER et al 2010, [24] MLK 2008): Rough estimation of run out e. g. by shadow angle (regional scale) 2D or 3D modelling (probabilistic): provides run out length, energy and bouncing-height distributions for slopescale problems For the design of mitigation measures, a probabilistic approach is going to be dened as a standard procedure in Austria ([28] ONR 24810) following the concept of partial factors of safety ([26] EUROCODES) for actions/resistances and varying accepted probabilities of failure depending on the casualty and reliability-classes of [27] Eurocode 0.
Fig. 6: Delineation of potential conict areas at regional extent using an empirical model ([21] Melzner et al 2010). Abb. 6: Abgrenzung potenzieller Wirkungsbereiche mittel einfachen empirischen Modellanstzen ([21] Melzner et al 2010).

Landslide hazard assessment General Landslides present complex natural phenomena for both the variability of processes and the dimensions. A landslide may exhibit a translational sheet slide of some square meters involving the ground surface or a deep seated mass movement of several cubic kilometres. Rapid landslides with reference to [6] CRUDEN & VARNES (1996) feature velocities of some metres per minute to several meters per second. In Austria, the main processes exhibit different slides and debris slides. Very rapid to rapid ow slides, which one can nd for example in Scandinavia or in Canada, have no relevance in Austria. Slides include rotational, translational and compound slides. Rotational slides own a circular sliding surface, which results from shear failure in relatively homogenous rock or soil of low strength. Translational slides take place in rock on forgiven more or less planar features like bedding planes, joints etc. The failure results when the shear resistance on the plane is exceeded. Relatively often one can nd these slides in the soil cover of the ground, called where sheet the slides, sliding

The combination of a rotational and a translational sliding mechanism is called a compound slide. These may develop in horizontally stratied soils and rocks, where the upper part of the slope shows a rotational failure which is constrained by a plane of weakness at the base (e. g. a claystone layer). A process that frequently can be observed in Austria are debris slides (e. g. Gasen and Haslau 2005, Vorarlberg). These failures occur in porous soils, especially after extraordinary water input resulting from precipitation and/or snow melt leading to an excess of pore water pressure. The mass movement often starts as a rotational slide, which turns into a debris ow down slope. When assessing landslide hazards, it is important to distinguish between preparatory factors and the triggers ([46] WL/WPLI 1994). The triggering of the occurrence of a mass movement is the last step of destabilization over a longer period of time. Concerning [37] THERZAGHI (1950) the stability of slopes is stated by the factor of safety, which is expressed by the ratio between driving

surface is formed by a weak clay layer, such as a gley horizon in the range of groundwater uctuations.
Fig. 7: An Example of changes of the factor of safety with time after [46] WL/WPLI (1994) Abb. 7: Beispiel fr die Vernderung der Sicherheit eines Einhanges ber die Zeit, nach [46] WL/WPLI (1994)

Seite 89

Hazard assessment and mapping of mass-movements in the EU

Seite 90

forces and resisting forces. Stable slopes feature a factor of safety over one, meaning that the resisting forces exceed the driving forces. If the driving forces are greater than the resisting forces the slope fails, i.e. the factor of safety drops under one. Fig. 5 ([46] WL/WPLI 1994) shows the development of a stable slope to one that fails. Since the slope is exposed to weathering, erosion processes etc. the factor of safety of the slope decreases to the point where it is close to failure (marginally stable). At this point the slope is susceptible to many triggers. When assessing landslide hazard the following information is needed regarding the ground conditions: geology and structures hydrogeology, type of process velocity of the process geotechnical involved potential role of human activities (triggers?). State of the practice in landslide assessment Conventional methods are based on observations of potentially unstable slopes. Aerial photos, both stereographic and orthophotos, have been used since decades to detect these slopes by characteristic geomorphological phenomena in combination with available geological maps ([4] BUNZA 1996, [14] KIENHOLZ 1995). This rst analysis is completed by mapping in the eld. The data are commonly presented in landslide hazard maps, which show the spatial distribution of different hazard classes. Additionally chronicles, which occasionally exist at the town halls, turned out to be very useful. properties of materials

State of the art in landslide assessment For several years, high resolution Lidar data have been available for most regions in Austria bearing landslide activity. They are a powerful tool to recognize geomorphological structures of landslides ([49] ZANGERL et al., 2008). A main advantage of Lidar data in comparison to conventional photos is the information on shaded areas and of areas covered with wood. Additionally, remote sensing systems (e.g. airborne and satellite-based multispectral and radar images) provide information on unstable, slowly creeping slopes, which may fail and transfer into a rapid moving masses ([31] PRAGER et al., 2009). Until recently, susceptibility/hazard maps in Austria were often made on demand. For some years authorities (LReg Krnten, WLV Obersterreich und Vorarlberg) are going to make comprehensive hazard maps giving a basis on decision-making for land use and development. Landslide inventories (databases of WLV, GBA, several federal states) in combination with GIS applications are used to get rapid information to areas prone to landslides. Collected surface data in combination with subsurface data gained from trenches and boreholes or seismic refraction, groundpenetrating radar and electrical resistivity proles allow for the drawing of an underground-model and deduce the type of failure mechanism which is most likely to occur. Geotechnical data are also required to assess the factor of safety and the probability of failure by means of analytical calculations or numerical modelling (e.g. [29] Poisel et al. 2006). Additional information on the process can be provided by a monitoring system. This serves as a check for the taken assumptions

and an evaluation of the mechanical model. Furthermore, a monitoring allows the prediction of failure time under certain circumstances (e.g. [9] FUKUZONO 1985, [19] KRHENBHL 2006, [32] ROSE & HUNGR 2007) Future development The development of forecast-models for the prognosis of the location and/or time of rapid gravitational mass movements to take place or even the meteorological settings which will trigger such events is at an early stage. Due to the fact that the authorities are strongly asking for such tools, many practitioners and scientists are focusing on that topic. The multitude of parameters inuencing the development of the erosion processes in question will keep the stakes high and will not allow for providing the authorities with the accurate models they ask for within a considerable time. Given the necessary detailed parameters, such as geology, hydrogeology, geotechnical parameters etc., triggering, inuencing or allowing for the processes in question are at hand, and all the necessary models are developed, it is highly likely that they will work in certain regions with similar or corresponding geological, morphological and meteorological conditions only. The accuracy of these models will depend with highly on a thorough events. well-documented necessarily calibration

future. Models showing the disposition of a given environment to tend to mass-movements and also forecasting the location, time and run-out of such processes will be a precious tool for the experts although a replacement of a thorough evaluation of the conditions on site is not to be expected anytime.

Anschrift der Verfasser / Authors addresses: Michael Mlk Forsttechnischer Dienst fr Wildbach und Lawinenverbauung, Geologische Stelle Liebeneggstr. 11 6020 Innsbruck michael.moelk@die-wildbach.at Thomas Sausgruber Forsttechnischer Dienst fr Wildbach und Lawinenverbauung Geologische Stelle Liebeneggstr. 11 6020 Innsbruck thomas.sausgruber@die-wildbach.at Richard Bk Abt. 15 Umwelt Geologie+Bodenschutz Flatschacher Strae 70 9020 Klagenfurt richard.baek@ktn.gv.at Arben Kociu Geologische Bundesanstalt Fachabteilung Ingenieurgeologie Neulinggasse 38 1030 Wien arben.kociu@geologie.ac.at

This emphasizes the necessity of a consistent documentation of events, to provide the modeldevelopers with calibration data. This means that the expertise of experts applied at dened locations with all the necessary eld work and assessment of natural parameters, fed in apt models will not become obsolete in the near and very probably not even in the far

Seite 91

Hazard assessment and mapping of mass-movements in the EU

Seite 92

Literatur / References:
[1] BK, EBERHART, GOLDSCHMIDT, KOCIU, LETOUZE-ZEZULA & LIPIARSKI (2005): Ereigniskataster und Karte der Phnomene als Werkzeug zur Darstellung geogener Naturgefahren (Massenbewegungen), Arb. Tagg. Geol. B.-A., Gmnd 2005 [2] BGBl. Nr. 436/1976: Verordnung des Bundesministers fr Land- und Forstwirtschaft vom 30. Juli 1976 ber die Gefahrenzonenplne [3] BMLFUW (2010): Richtlinie fr die Gefahrenzonenplanung-LE.3.3.3/0185-IV/5/2007 vom 12. Jnner 2010 [4] BUNZA, G. (1996): Assessment of landslide hazards by means of geological and hydrological risk mapping [5] BUWAL: Symbolbaukasten zur Kartierung der Phnomene. Mitt. Bundesamt f. Wasser u. Geologie 6, p. 41, 2004 [6] CRUDEN, D.M.; VARNES D. J. (1996): Landslide Types and Processes. In: Turner A.K. and Schuster R.L. (eds.): Landslides: Investigation and mitigation. Special report 247. Washington D.C.: National Academic Press, 36-45,1996. [7] DORREN, L., JONNSON, M., KRAUTBLATTER, M., MOELK, M. AND STOFFEL, M. (2007): State of the Art in Rock-Fall and Forest Interactions, Schweizerische Zeitschrift fr Forstwesen 158 (2007) 6: S 128-141 [8] Forstgesetz 1975, 11 [9] FUKOZONO T. (1985): A new method for predicting the failure time of a slope. Proc. 4th Int. Conf. and eld workshop on landslides, Tokyo, 145-150, 1985. [10] HUNGR, O.; EVANS, S.G. (2004): The occurrence and classication of massive rock slope failure. Felsbau 22, 16-23, 2004. [11] HUTCHSINSON, J.N. (1988): General Report: Morphological and geotechnical parameters of landslides in the relation to geology and hydrogeology. In: Bonnard (ed.): Proceedings of the 5th International Symposium on Landslides, Vol 1. Rotterdam: Balkema, 3-35, 1988. [12] HBL, J., KOCIU, A., KRISSL, H., LANG, E., LNGER, E., RUDOLFMIKLAU, F., MOSER, A., PICHLER, A., RACHOY, Ch., SCHNETZER, I., SKOLAUT, Ch., TILCH, N. & TOTSCHNIK, R. (2009): Alpine Naturkatastrophen Lawinen-Muren-Felsstrze-Hochwsser, 120 S..Leopold Stocker Verlag, Graz. [13] JABOYEDOFF Michel, BAILLIFARD Francois, MARRO Christian, PHILIPPOSSIAN Frank & ROUILLER Jean-Daniel (1999): Detection of Rock Instabilities: Matterock Methodology. Joint japa-Swiss [14] KIENHOLZ, H, KRUMMENACHER, B, LIENER, S. (1995): Erfassung und Modellierung von Hangbewegungen als Beitrag zur Erstellung von Gefahren-Hinweiskarten. Report, Mnchner Forum fr Massenbewegungen, Mnchen [15] KLINGSEISEN, B., LEOPOLD, Ph., TSCHACH, M. (2006): Mapping Landslide Hazards in Austria: GIS Aids Regional Planning in NonAlpine Regions. ArcNews 28 (3): 16, 2006. [16] KOCIU A. et al. (2007): Massenbewegungen in sterreich. JB der Geol. B.-A. Band 147, Heft 1+2, S 215-220. Wien 2007 [17] KOCIU, A., TILCH N., SCHWARZ L,. HABERLER A., MELZNER S. (2010): GEORIOS - Jahresbericht 2009; Geol.B.-A. Wien 2010. [18] KOLMER, Ch. (2009): Geogenes Baugrundrisiko bersterreich. Vortrag im Rahmen des Landesgeologentages 2009, St. Plten 2009. [19] KRHENBHL R. (2006): Der Felssturz, der sich auf die Stunde genau ankndigte. Bull. Angew. Geol., 11(1), 49-63, 2006.

[20] MELZNER, S., LOTTER, M. & A. KOCIU (2009): Development of an efcient methodology for mapping and assessing potential rock fall source areas and runout zones. European Geosciences Union (EGU), General Assembly, 19-24th April 2009, Vienna. (http://www. geologie.ac.at/pdf/Poster/poster_2009_egu_melzner.pdf) [21] MELZNER, S., DORREN, L. , KOCIU, A. & R. BK (2010B): Regionale Ausweisung potentieller Ablse- und Wirkungsbereichen von Sturzprozessen im Oberen Mlltal/Krnten. Poster Prsentation beim Geoforum Umhausen 2010, Niederthai, Tirol. (Poster download on GBA homepage www.geologie.ac.at) [22] MELZNER, S., TILCH, N., LOTTER, M., KOIU, A. & BK, R. (2010C): Rock fall susceptibility assessment using structural geological indicators for detaching processes such as sliding or toppling. European Geosciences Union (EGU), General Assembly, 02-07 Mai 2010, Wien. (http://www. geologie.ac.at/pdf/Poster/poster_2010_egu_melzner_etal.pdf) [23] MELZNER, S., MLK, M., DORREN, L. & R. BK (2010A): Comparing empirical models, 2D and 3D process based models for delineating maximum rockfall runout distances. European Geosciences Union (EGU), General Assembly, 02-07 Mai 2010, Vienna. (http://www. geologie.ac.at/pdf/Poster/poster_2010_egu_melzner_2d_3d.pdf) [24] MLK, M. (2008): Regionalstudie Wipptal Sdost: Erfassung und Darstellung von Naturgefahrenpotentialen im Regionalen Mastab nach EtAlp Standards. Poster Prsentation beim Geoforum Umhausen 2008, Niederthai, Tirol. [25] MLK M. und NEUNER G. (2004): Generelle Legende fr Geomorphologische Kartierungen des Forsttechnischen Dienst fr Wildbach und Lawinenverbauung, Geologische Stelle, Innsbruck, S.49, 2004 [26] NORM EN 1990: Eurocode: Grundlagen der Tragwerksplanung [27] NORM EN 1997-1: Eurocode 7: Entwurf, Berechnung und Bemessung in der Geotechnik. Teil 1: Allgemeine Regeln [28] ONR 24810: Technischer Steinschlagschutz: Begriffe und Denitionen, geologischgeotechnische Grundlagen, Bemessung und konstruktive Ausgestaltung, Instandhaltung und Wartung. In preparation, foreseen publication: 2011 [29] POISEL, R., ANGERER, H., PLLINGER, M., KALCHER, T., KITTL, H. (2006): Assessment of the Risks Caused by the Landslide Lrchberg ? Galgenwald, Austria. Felsbau 24, No. 3, S. 42-49 (2006) [30] POSCH-TRZMLLER, G. (2010): Adapt Alp WP 5.1 Hazard Mapping - Geological Hazards. Literature Survey regarding methods of hazard mapping and evaluation of danger by landslides and rock fall. Final Report, Geologische Bundesanstalt, Wien, 2010 (www.ktn.gv.at/Verwaltung/Abteilungen/Abt.15 Umwelt, Thema Geologie und Bodenschutz) [31] PRAGER, Ch.; ZANGERL, Ch.; NAGLER, Th. (2009): Geological controls on slope deformations in the Kfels rockslide area (Tyrol, Austria). AJES 102/2 (2009), 4-19 [32] ROSE, N.D. and HUNGR O. (2007): Forecasting potential rock slope failure in open pit mines using the inverse-velocity method. Int. Jour. of Rock Mech. and Min. Science, 44, 308-320, 2007. [33] RUDOLF-MIKLAU F. & SCHMIDT F. (2004): Implementation, application and enforcement of hazard zone maps for torrent and avalanches control in Austria, Forstliche Schriftenreihe, Universitt fr Bodenkultur Wien, Bd. 18, p. 83-107, 2004 [34] RUFF, M. (2005): GIS-gesttzte Risikonanalyse fr Rutschungen und Felsstrze in den Ostalpen (Vorarlberg, sterreich). Georisikokarte Vorarlberg. Diss. Univ. Karlsruhe, 2005. [35] SCHWARZ, L., TILCH, N. & KOCIU. A. (2009): Landslide sucseptibility mapping by means of articial Neuronal Networks performed for the region Gasen-Haslau (eastern Styria, Austria) 6th European Congress on regional Geoscientic Cartography and Information Systems. (http://www.geologie.ac.at/pdf/Poster/poster_2009_euregio.pdf)

[36] SCHWEIGL, J.; HERVAS, J. (2009): Landslide Mapping in Austria. JRC Scientic and Technical Report EUR 23785 EN, Ofce for Ofcial Publications of the European Communities, 61 pp. ISBN 978-92-79-11776-3, Luxembourg, 2009. [37] TERZAGHI, K. (1950): Mechanism of landslides. Geological Society of America. Berkey Volume 1950, 83-124 [38] TILCH, N. (2009): Datenmanagementsystem GEORIOS (Geogene Risiken sterreich). Vortrag im Rahmen des Landesgeologentages 2009, St. Plten 2009. [39] TILCH, N. (2010): Rumliche und skalenabhngige Variabilitt der Datenqualitt und deren Einuss auf mittels heuristischer Methode erstellte Dispositionskarten fr Massenbewegungen im Lockergestein - eine Fallstudie im Bereich Niedersterreichs , 12. Geoforum Umhausen 14.-15.10.10, Niederthai, (http://www.geologie.ac.at/pdf/Poster/poster_2010_geoforum_tilch.pdf). [40] TILCH, N. (2010): Erstellung von Dispositionskarten fr Massenbewegungen Herausforderungen, Methoden, Chancen, Limitierungen.- Vortrag Innsbrucker Hofgesprche 26.05.2010, Innsbruck; (http://bfw.ac.at/050/ pdf/IHG_26_05_2010_Tilch_Schwarz.pdf) [41] TILCH, N., MELZNER, S., JANDA, C. & A. KOCIU (2009): Simple applicable methods for assessing natural hazards caused by landslides and erosion processs in torrent catchments. European Geosciences Union (EGU), General Assembly, 19-24th April 2009, Vienna. (http://www.geologie.ac.at/pdf/Poster/poster_2009_egu_tilch_etal.pdf) [42] WP/WLI - Working Party on Landslide Inventory (International Geotechnical Societies of UNESCO) (1990): Suggested Nomenclature for Landslides . Bull. Intern. Ass. Eng. Geology, No. 41, Paris 1990

[43] WP/WLI - Working Party on Landslide Inventory (International Geotechnical Societies of UNESCO) (1990): Suggested Method for Reporting a Landslide . Bull. Intern. Ass. Eng. Geology, No. 41, Paris 1990 [44] WP/WLI - Working Party on Landslide Inventory (International Geotechnical Societies of UNESCO) (1991): A Suggested Method for a Landslide Summary. Bull. Intern. Ass. Eng. Geology, No. 43, Paris 1991 [45] WP/WLI - Working Party on Landslide Inventory (International Geotechnical Societies of UNESCO) (1993): A Suggested Method for describing the Activity of a Landslide. Bull. Intern. Ass. Eng. Geology, No. 47, Paris 1993 [46] WP/WLI - Working Party on Landslide Inventory (International Geotechnical Societies of UNESCO) (1994): A Suggested Method for Reporting Landslide Causes. Bull. Intern. Ass. Eng. Geology, No. 50, Paris 1994 [47] WP/WLI - Working Party on Landslide Inventory (International Geotechnical Societies of UNESCO) (1995): A Suggested Method for the Rate of Movement of a Landslide. Bull. Intern. Ass. Eng. Geology, No. 52, Paris 1995 [48] WYLLIE D. C. (2006): Risk management of rock fall hazards. Sea to Sky Geotechnique, Conference Proceedings, 25-32, Vancouver 2006. [49] ZANGERL C., PRAGER C., BRANDNER. R., BRCKL E., EDER S., FELLIN W., TENTSCHERT E., POSCHER G., & SCHNLAUB H. (2008): Methodischer Leitfaden zur prozessorientierten Bearbeitung von Massenbewegungen. Geo.Alp, Vol. 5, S. 1-51, 2008. [50] ZANGERL C.; PRAGER, Ch. (2008): Inuence of geologcial structures on failure initiation, internal derformation and kinematics of rock slides. American Rock Mechanics Association, 0863, (2008)

Seite 93

Hazard assessment and mapping of mass-movements in the EU

Seite 94

Introduction Switzerland is a country exposed to many natural hazards. These hazards include earthquakes, oods, forest res, snow avalanches, rock falls and debris ows. More than 6% of Switzerland is affected by hazards due to slope instability. These areas occur mainly in the Prealps and in the Alps. The Randa rock avalanches of 1991 are a good example of the potential of such hazards. Thirty million m3 of fallen debris cut off the valley for two weeks. In another HUGO RAETZO, BERNARD LOUP case, a landslide was reactivated with historically unprecedented rates of displacement up to 6 m/ day, causing the destruction of the village of FalliHlli in the year 1994. The legal and technical background conditions for the protection against landslides have undergone considerable changes since the 80s. The ooding of 1987 promoted the federal authorities to review criteria governing natural

step the hazard of landslides is assessed according to the methods used in the Swiss strategy against all natural hazards (e.g. oods, avalanches). The hazard assessment is then integrated into land use planning and in the risk management (3. step). First step: Hazard identication Landslides can be classied according to the estimated depth of the sliding plane (< 2m: shallow; 2-10 m: intermediate; >10 m: deep) and the long term mean velocity of the movements (< 2 cm/year: substabilised; 2-10 cm/year: slow; > 10 cm/year: active). These depth and velocity parameters are not always sufcient to estimate the potential danger of a landslide. Differential movements must also be taken into account since they can generate buildings to topple or cracks to open. Rock falls are characterized by their speed (< 40 m/s), the size of their elements (stone < 0.5 m, block > 0.5 m) and the volumes involved. Rock avalanches with huge volumes (v > 1million m3) and high speed (> 40 m/s) can also happen although these are rare. Due to heavy rainfall, debris ows and very shallow landslides are frequent in Switzerland. These are moderate volume (< 20,000 m3) and high speed features (1-10 m/s). These phenomena are very dangerous and annually cause important trafc disruptions and fatalities. A map of landslide phenomena and an associated technical report provide signs and indications of slope instability as observed in the eld. The map represents phenomena related to dangerous processes and delineates the vulnerable areas. Field interpretation of these phenomena allows areas vulnerable to landslides to be mapped. This is based on the observation and interpretation of landforms, on structural and geomechanical properties of slope instabilities,

Geological Hazard Assessment in Switzerland Geologische Gefahrenbeurteilung in der Schweiz


Summary: Geological hazard assessments are based on Swiss laws dealing with natural hazards. Guidelines are published by the Federal Ofce for the Environment (FOEN/BAFU). According to the integrated risk management, the methods are applied for all natural hazards (landslides, oods, snow avalanches). The hazard maps are dealing with ve degrees: high (red), medium (blue), low (yellow), residual (yellow-white), no hazard (white). Zusammenfassung: Geologische Gefahren werden in der Schweiz gem den eidgenssischen Gesetzen ber den Wald und den Wasserbau erhoben und beurteilt. Dazu hat das zustndige Bundesamt (heute das Bundesamt fr Umwelt BAFU) entsprechende Empfehlungen und Richtlinien verffentlicht. Im Sinne des integralen Risikomanagements werden fr alle Gefahrenprozesse vergleichbare Methoden angewendet und anschlieend in der Planung umgesetzt. Das gilt fr geologische Massenbewegungen, Hochwasser und Lawinen. Fr diese Prozesse werden Gefahrenkarten erstellt, die immer fnf Gefahrenstufen ausscheiden: Hohe, mittlere und geringe Gefahr sowie Restgefhrdung und keine Gefhrdung. Daraus entstehen die roten, blauen, gelben, gelb-wei gestreiften und weien Zonen auf den Gefahrenkarten.

hazard protection. The Federal Flood Protection Law and the Federal Forest Law came into force in 1991. Their purpose is to protect the environment, human lives and property from the damage caused by water, mass movements, snow avalanches and forest res. Following the promulgation of these new regulations, greater emphasis has been placed on preventive measures. Consequently, hazard assessment, the identication of protection objectives, purposeful planning of preventive measures and the limitation of the residual risk are of central importance. The cantons are now required to establish inventories and maps denoting areas of hazards, and to take them into account in the land use planning. For the improvement of the inventories and the hazard maps, the federal government provides subsides to the cantonal authorities (50%). In a rst step the landslides are identied and classied. During this phase inventories and maps of phenomena are established. In a second

Seite 95

Hazard assessment and mapping of mass-movements in the EU

Seite 96

and on historical traces. Extensive knowledge of past and current events in a catchment area is essential if zones of future instability are to be identied. Some recommendations for the uniform classication, representation and documentation of natural processes have been established by the Swiss federal administration. Consequently, the denition of features on a natural hazard map is based on a uniform legend for landslides, oods and snow avalanches. The different phenomena are represented by different colours and symbols.
RED: high hazard

An additional distinction is made between potential, inferred or proved events. According to the scale of mapping (e.g. 1:50,000 for the Master Plan, 1:5,000 for the Local Plan), this legend may contain a large number of symbols. Inventories: Recommendations for

allow an overview of the different natural disasters and potential associated damage in Switzerland. Second step: Hazard assessment of landslides Hazard is dened as the occurrence of a potentially damaging natural phenomena within a specic period of time in a given area. Hazard assessment implies the determination of the magnitude or intensity of an event over time. Mass movements often correspond to gradual (landslides) or unique (falls, debris ows) events. It is sometimes difcult to make an assessment of the return period of a massive rock avalanche, or to predict when a dormant landslide may reactivate. Some federal recommendations have been proposed in the 90s for the management of landslides and oods. Since 1984 similar recommendations have already existed for snow avalanches. Hazard maps, according to the federal recommendations (guidelines), express three degrees of danger, represented by corresponding colours: red, blue and yellow (Fig. 1). The various hazard zones are delineated according to the landslide phenomena maps, the register of slope instability events and additional documents. Numerical models (analysis of block trajectories, calculations of factors of safety) may be used to determine the extent of areas endangered by rock falls, or to present quantitative data on the stability of a potentially unstable area. A chart of the degrees of danger has been developed in order to guarantee a homogeneous and uniform means of assessment of the different kinds of natural hazards across Switzerland (oods, snow avalanches, landslides) for example, Fig.1 for fall processes. Two major parameters are used to classify the danger: the intensity, and the probability (frequency or return period). Three degrees of danger have been dened. These are represented by the colours red,

blue and yellow. The estimated degrees of danger have implications for land use. They indicate the level of danger to people and to animals, as well as to property. In the case of mass movement, people are considered safer inside the buildings than outside. A description of the magnitude of potential damage caused by an event is based on the identication of threshold values for degrees of danger, according to possible damage to property. The intensity parameter is divided into three degrees: High intensity: People and animals are at risk of injury inside buildings; heavy damage to buildings or even destruction of buildings is possible. Medium intensity: People and animals are at risk of injury outside buildings, but are at low risk inside buildings; lighter damage to buildings should be expected. Low intensity: People and animals are slightly threatened, even outside buildings (except in the case of stone and block avalanches, which can harm or kill people and animals); supercial damage to buildings should be expected. Criteria for the intensity assessment: There is generally no applicable measure to dene the intensity of slope movements. However, indicative values can be used to dene classes of high, mean and low intensity. Applied criteria usually refer to the zone affected by the process, or to the threatened zone. For rock falls, the signicant criterion is the impact energy in the exposed zone (translation and rotation energy). The 300 kJ limit corresponds to the impact energy to which can be resisted by a reinforced concrete wall, as long as the structure is properly constructed. The 30 kJ limit

the denition of a uniform Register for slope instability events has been developed, including special sheets for each phenomenon (landslides, oods, snow avalanches). Each canton is currently compiling the data for its own register. These databases (StorMe) are transferred to the FOEN to

People are at risk of injury both inside and outside buildings. A rapid destruction of buildings is possible. or: Events occurring with a lower intensity, but with a higher probability of occurrence. In this case, people are mainly at risk outside buildings, or buildings can no longer house people. The red zone mainly designates a prohibition domain (area where development is prohibited). BLUE: moderate hazard People are at risk of injury outside buildings. Risk is considerably lower inside buildings. Damage to buildings should be expected, but not a rapid destruction, as long as the construction type has been adapted to the present conditions. The blue zone is mainly a regulation domain, in which severe damage can be reduced by means of appropriate protective measures (area with restrictive regulations). YELLOW: low hazard People are at slow risk of injury. Slight damage to buildings is possible. The yellow zone is mainly an alerting domain (area where people are notied at possible hazard). YELLOW-WHITE HATCHING: residual danger Low probability of high intensity event occurrence can be designated by yellow-white hatching. The yellow-white hatched zone is mainly an alerting domain, highlighting a residual danger. WHITE: no danger or negligible danger, according to currently available information.

Seite 97

Hazard assessment and mapping of mass-movements in the EU

Seite 98

Phenomena Rock fall Rock avalanche Landslide

Low intensity E < 30 kJ v 2 cm/y dv, D, T

Medium intensity 30 < E < 300 kJ v : 2-10 cm/y dv, D, T

High intensity E > 300 kJ E > 300 kJ v>10 cm/year dv, D, T v > 0.1 m/day for shallow landslides; displacement > 1 m per event

correlated with recurrent meteorological conditions. The probability of mass movement occurrence should mainly be established for a given duration of land use. Thus, the probability of potential damage during a certain period of time, or the degree of safety of a specic area should be taken into account, rather than the frequency of dangers. The probability of occurrence and the return period can be mathematically linked, if attributed to the same reference period: p = 1 (1 1/T)n

For example, considering a time period of 30 years, an event with a 30-year return period has a 64% probability of occurrence (or about 2 in 3), of 26% (or about 1 in 4) for a 100-year return period, and of 10% (or about 1 in 10) for a 300year return period. The calculation of the probability of occurrence clearly shows that even for a rather high return period (300 years), the residual danger remains not signicant. In principle, the probability scale does not exclude very rare events, neither does it exclude the intensity scale for high magnitude events. Hazards with a very low probability of occurrence are usually classied as residual dangers under the standard classication. In the

Earth ows and debris ows potential real e < 0.5 m 0.5 m < e < 2 m h<1m e>2m h>1m

Whereby p is the probability of occurrence, n represents the given time period (for example 30 or 50 years), and T is the return period.

E: kinetic energy; e: thickness of the unstable layer; h: height of the earthow deposit; v: long term mean velocity, dv: variation of velocity (accelerations), D: differential movements, T: thickness of the landslide. corresponds to the maximum energy that oakwood stiff barriers can resist (e.g. rail sleeper). For rock avalanches, the high intensity class (E > 300 kJ) is always reached in the impact zone. The target zones affected by block avalanches of low to medium intensity can only be roughly delineated. Therefore, it is recommended not to articially delineate zones affected by low to medium intensities. Most landslides: A low intensity movement has an annual mean speed of lower than 2 cm per year. A medium intensity has a speed ranging from one to 10 cm per year. The high intensity class is assigned to velocities higher than 10 cm per year and to shear zones or zones with clear differential movements (D). It may also be assigned if reactivated phenomena have been observed or, if horizontal displacements greater than one meter per event may occur. Finally, the high intensity class can also be assigned to very rapid shallow landslides (speed > 0.1 m/day). In the area affected by landsliding eld, intensity criteria can be directly converted to danger classes. Other criteria as velocity changes or accelerations (dv), differential movements (D) and thickness of the landslide (T) can lead to increase resp. to reduce the intensity For earth ows and debris ows, the intensity depends on the thickness of the potentially unstable layer. The boundaries dening the three intensity classes are set at 0.5 m and 2 m. Probability: Probability of landslides is dened according to three classes. The class limits are set at 30 and 300 years and are equivalent to those class as derived from the long term velocity.

high

RED YELLOW / WHITE

INTENSITY

medium

BLUE

low

established for snow avalanches and oods. The 100-year limit corresponds to a value applied in the design of ood protection structures. The results of probability calculations to determine if mass movements occur remain very uncertain. Unlike oods and snow avalanches, mass movements are usually non-recurrent processes. The return period, therefore, only has a relative meaning, except for events involving stone and block avalanches and earth ows, which can be

YELLOW

high
Fig. 1: Matrix for the assessment of hazards Abb. 1: Matrix fr die Gefahrenbeurteilung

medium PROBABILITY

low

very low

Seite 99

Hazard assessment and mapping of mass-movements in the EU

Seite 100

domain of dangers related to mass movements, the limit for a residual danger has been set for an event with a 300-year return period. The degree of hazard is dened in a hazard matrix based on intensity and probability criteria (Raetzo & Loup 2009). The resulting hazard map is mainly used for planning (land use), while the design of protection measures needs more detailed investigations. In general the methods used are related to the product, scales and the risk in order to respect economic criteria: low efforts are done for the Swiss indicative map (level 1), important efforts are done when a hazard map is established or reviewed (level 2). Detailed analyses and engineering calculations are foreseen for the planning of countermeasures (level 3). Applying this concept rising efforts for geological investigations are planned when the assessment on the second or third level takes place. Third step: Land use planning and risk management The hazard map is a basic document used in land use planning. Natural hazards should be taken into account particularly in the following situations: Elaboration and improvement of cantonal Master Plan and Communal Local Plans for land use. Planning, construction, transformation of buildings and infrastructures. Granting of concessions and planning for construction and infrastructural installations. Granting of subsidies for building and development (road and rail networks, residences), as well as for slope stabilisation and protection measures.

According to Art. 6 of the Federal Law for Land use Planning, the cantons must identify all areas that are threatened by natural hazards. The cantonal Master Plan is a basic document for land use planning, infrastructural coordination and accident prevention. It consists of a map and a technical report, and is based on studies. The Master Plan allows for deciding the following: It shows how to coordinate activities associated with different land uses. It identies the goals of planning and species the necessary stages. It provides legal constraints to the authorities in charge of land use planning. The objectives of the Master Plan with respect to natural hazards are: To early detect conicts between land use, development and natural hazards. To rene the survey of basic documents concerning natural hazards. To formulate principles that can be applied by the cantons to the issue of protection against natural hazard. To dene necessary requirements and mandates to be used in subsequent planning stages. The constraints on Local Planning already allow and ensure appropriate management of natural hazards with respect to land use. The objective of these constraints is to delineate danger zones by highlighting restrictions, or to establish legal frameworks leading to the same ends. At the same time danger zones can be delineated on the local plan with areas suitable for construction as well as additional protection zones.

The degrees of danger are initially assigned according to their consequences for construction activity. They must minimise risks to the safety of people and animals, as well as minimising as possible damage to property. In agricultural zones, buildings affected by different degrees of danger are constrained by the same conditions as those in built-up areas.

Anschrift der Verfasser / Authors addresses: Hugo Raetzo Federal Ofce for the Environment FOEN Bundesamt fr Umwelt BAFU 3003 Bern Schweiz Bernard Loup

Conclusions In Switzerland legal and technical references are published to clarify which responsibilities the authorities have and how the assessment has to be done in order to apply the concept of integral risk management. The hazard map indicates which areas are unsuitable for use, according to existing natural hazard. The integration of hazard maps into land use planning (including construction conditions, building licences) and the development of protective measures to minimise damage to property are main objectives. When the hazard map is compared with existing land use conicts may occur. Since it is difcult or impossible to change land use, specic construction codes are required to reach the desired protection level. Hazard maps are also considered in planning protective measures as well as the installation of warning systems and emergency plans. The federal recommendations are on attempt to mitigate natural disasters by restricting development on unstable areas.

Federal Ofce for the Environment FOEN Bundesamt fr Umwelt BAFU 3003 Bern Schweiz

Literatur / References:
BUNDESAMT FR RAUMPLANUNG, BUNDESAMT FR WASSERWIRTSCHAFT & BUNDESAMT FR UMWELT, WALD UND LANDSCHAFT, (1997). Empfehlungen, Bercksichtigung der Massenbewegungsgefahren bei raumwirksamen Ttigkeiten, EDMZ, 3000 Bern. CRUDEN D.M. UND VARNES D.J.: Landslide types and processes. In: A. Keith Turner & Robert L. Schuster (eds): Landslide investigation and mitigation: 36-75. Transportation Research Board, special report 247. Washington: National Academy Press, 1996. KIENHOLZ, H., KRUMMENACHER, B. et al.: Empfehlungen Symbolbaukasten zur Kartierung der Phnomene Ausgabe 1995, Mitteilungen BUWAL Nr. 6, 41 S., Reihe Vollzug Umwelt VU7502-D, Bern 1995. RAETZO et al.: Hazard assessment of mass movements codes of practice in Switzerland, International Association of Engineering Geology IAEG Bulletin, 2002. RAETZO, H. & LOUP, B.; BAFU: Schutz vor Massenbewegungen. Technische Richtlinie als Vollzugshilfe. Entwurf 9. Sept. 2009. VARNES, D.J. and IAEG Commission on Landslides and other MassMovements: Landslide hazard zonation: a review of principles and practice. The UNESCO Press, Paris, 1984.

Seite 101

Hazard assessment and mapping of mass-movements in the EU

Seite 102

Introduction When facing a natural hazard, risk management can be divided in several stages: a) danger characterization, hazard assessment and vulnerability analysis; b) risk evaluation and assessment; STEFANO CAMPUS c) risk prevention (protective works, land use regulation, monitoring, etc.); d) crisis and post-crisis management; e) feedback from experience. It is essential to properly distinguish the three aspects of landslides studies: DANGER. Threat characterization (typology, morphology even quantitative, inventory); HAZARD. Spatial and temporal probability, intensity and forecasting of evolution (scenarios) are needed; RISK. Interaction between a threat having particular hazard and human activities. We need vulnerability and damage analysis. These differences are theoretically well known by all technicians but often there are some problems when they have to be applied in a legal framework. So, it is not so unusual to nd inventory maps used as hazard maps or damage maps called risk maps. Therefore, we have to distinguish two situations: 1) Landslides studies that have no inuence from legal point of view. Typical cases are the studies carried out by universities about relevant landslides. The aim is, for example, to understand the mechanical features of instability or to study different ways of evolution of the phenomenon (scenarios) in order to assess residual risk. Any method to assess landslide hazard and risk can be used. They include statistical, deterministic, numerical, etc. methods for hazard and qualitative or matrix calculus for risk. Landslide inventory can be made by means of historical, morphological, etc. approach.

2) Landslide studies that have direct consequences to land planning laws, at local scale or higher. GIS methods allow for performing analyses over wide areas that are useful to be included in basin plans or master plans. National or local laws can require standard ways to present the results (common graphical signs on the maps, for example). Legal framework in Italy and Piemonte High Level Legislation (national level) The national Law n. 445/1908 (Transfer and consolidation of unstable towns) and Royal Decree R.D. n. 3267/1923 (Establishment of areas subject to hydro-geological constrains) were the rst public regulations on land use planning. At the beginning of 70s, land use management was transferred to the regions. The national Law n. 183/1989 introduced land use planning at a basin scale: the government sets the standards and general aims without xing a methodology to analyze and evaluate the dangers, hazards, and risks related to natural phenomena. The same law designated the Autorit di Bacino (Basin Authorities) whose main goal is to draw up the Basin Plan, a tool for planning actions and rules for conservation and protection of the territory. About Po basin, the last plan adopted in 2001 is called PAI (Piano per lAssetto Idrogeologico or Hydrogeological System Plan of River Po Basin). It tries to verify the geological instability of the whole territory as regards the land use planning through a process of upgrading and feedback with the local urban management plans. Moreover, all the municipalities are classied according different risk levels, mainly from a qualitative point of view. For landslides it has two atlases (1:25,000 scale):

Landslide Mapping in Piemonte (Italy): Danger, Hazard & Risk Kartierung von Rutschungen im Piemont (Italien): Gefahren & Risiken
Summary: This paper briey describes the legal framework of landslide danger, hazard and risk mapping in Italy and Piemonte. Laws or rules that indicate how a landslide analysis (danger, hazard, risk) has to be done, do not exist. As a general remark, it has to be observed that public legislation denes general principles and lines of conduct, functions, activities and authorities involved, while the regional administrations apply restrictions on land use through different regional laws. Keywords: Landslide, danger, hazard, risk, Piemonte, Italy Zusammenfassung: Diese Abhandlung beschreibt kurz den gesetzlichen Rahmen der Kartograe von Rutschungsgefahren und -risiken in Italien und im Piemont. Es gibt keine Gesetze oder Verordnungen darber, wie eine Rutschungsanalyse (Gefahren und Risiken) auszufhren ist. Als eine allgemeine Bemerkung ist festzustellen, dass die ffentliche Gesetzgebung allgemeine Prinzipien und Richtlinien, Funktionen, Aktivitten und betreffende Befugnisse festlegt, die Regionalverwaltungen hingegen erlegen auf der unterschiedlichen landesgesetzlichen Basis Einschrnkungen hinsichtlich der Bodennutzung auf. Schlsselwrter: Rutschung, Gefahr, Gefhrdung, Risiko, Piemont, Italien

Seite 103

Hazard assessment and mapping of mass-movements in the EU

Seite 104

1) Atlas of Hydro-geological Risks (landslides, oods, alluvial fans, avalanches) at the municipal level. Every municipality is valued on the basis of the hazard, vulnerability and expected damage. Landslide hazard is function of ratio between area of landslides within municipal boundaries and whole area of municipality. It has 4 qualitative classes: R1-moderate risk. Social damages and few economic losses are possible. R2-medium risk. Few damages to buildings and infrastructures without loss of functionality. R3-high risk. Problems to human safety. Many damages and economic losses. R4-very high risk. Deaths and severe injuries are possible.

2) Atlas of Landslides. It is an inventory, in which polygons and points are divided in 3 classes (g. 1): Fa-Area with Active Landslides (very high hazard). No new buildings or infrastructures are allowed. Only measures of protection and reduction of vulnerability; Fq-Area with Quiescent Landslides (high hazard). Some enlargements are allowed. New buildings are allowed according to city development plan. Fs-Area with Stabilized Landslides The (medium-moderate hazard).

government to give answers for development regulation (to reduce or eliminate landslides losses). According to the national Law n. 267/1998, the government enforced legislative measures at the national level, including the procedure to dene landslide risk areas. Another important aspect of the Law n. 267/1998 regards the development of extraordinary plans to manage the situations of higher risk (R.M.E.-Aree a Rischio Molto Elevato), where safety problems or functional damages are possible. Local and regional authorities are obliged to dene, design and apply proper measures to risk mitigation, with national funding. In Piedmont, these actions have been applied in some signicant cases such as in Ceppo Morelli (Valle Anzasca in northern part of Piemonte), classied as a very high-risky area. Low Level Legislation (Local Urban Development Plan) The classication of areas made by the Po Basin Authority is a binding act. The municipality must adopt a new town development plan taking into account that classication. If the municipality wants to change PAI classication, a deep analysis of the areas has to be done to justify new land use destination. Regione Piemonte Regional Law for Urban Development L.R. n. 56/1977, which is the main legal instrument of land use management at a local scale, as well as the Regional Law L.R. n. 45/1989 which regulates land use modication and transformation in areas subject to environmental protection, divides areas in more detailed classes having (almost) same meaning of PAI classication. In Piemonte, the local management plan

of geological and morphological features and historical analysis. In a state of emergency (as established by the Regional Law n. 38/1978, which regulate and organise interventions related to severe instability phenomena), a specic article of the regional law 56/1977 (art. 9/bis) allows inhibiting or suspending development in the involved areas. Consequently, new land-use planning must be realised (upgrade/ revision of the local management plan). The last integrations to this law (Circolare del Presidente della Giunta Regionale, n. 7/LAP/1996 and Nota Tecnica Esplicativa, n. 12/1999) introduced the concept of hazard and risk zoning, classifying the whole territory in different classes where land uses are precisely regulated and dened, where building is forbidden, where preventive measures have to be taken, etc It is important to clarify that Regione Piemonte does not have an ofcial regional Geological Survey. Some geological functions are for executed by Arpa Piemonte (Agency two Environmental Protection) having

development of these areas is indicated in the city development plan. The catastrophic event of May 1998, which caused heavy damages and victims in municipalities of Sarno and Quindici (Campania), urged the

geological departments: one dedicated to Geological Informative System, research and applied projects, the other one deals with geological aspects of municipality urban plans. Therefore, we produce landslide danger, hazard and risk analyses that have not any legal consequences. Within many regional, national and projects, Arpa Piemonte carried European

out many experiences in elds of assessing methodology for landslides hazard assessment: for instance, the IMIRILAND Project within Fifth Framework Programme, Interreg PROVIALP Project Fall or national Project of Geological Cartography for shallow and planar landslides hazard maps in the southern hilly part of Piemonte region called Langhe (g. 2).

Fig. 1: Example of Atlas of Landslides published by Po River Basin Authority (elaboration by Arpa Piemonte). Abb. 1: Beispiel des Atlas of Landslides (Bergsturz-Atlas), verffentlicht von Po River Basin Authority (Ausarbeitung von ARPA Piemonte).

(required by the Regional Law L.R. n. 56/1977) includes the danger/hazard zoning in order to identify landslide prone areas on the basis

Seite 105

Hazard assessment and mapping of mass-movements in the EU

Seite 106

existing landslides (g. 3). Every region decided


Fig. 2: Extract from the shallow landslides hazard map of 1:50,000 scale sheet Dego in Piemonte. The trafc light colors indicate increasing hazard (from green to red), referring to return periods of critical rainfall (Arpa Piemonte, 2006). Abb. 2: Auszug aus dem Gefahrenzonenplan rutschgefhrdeter, oberchennaher Hnge im Mastab von 1:50.000 Dego im Piemont. Die Ampelfarben veranschaulichen die zunehmende Gefahr (von grn zu rot) mit Bezug auf Wiederkehrdauern kritischen Niederschlags (ARPA Piemonte, 2006).

Anschrift des Verfassers / Authors address: Stefano Campus Arpa Piemonte Dipartimento Tematico Geologia e Dissesto via Pio VII 9, 10135 TORINO (ITALY) stefano.campus@regione.piemonte.it

by itself if the results of IFFI Project (danger maps) do or do not have or a legal value. Currently, in Piemonte landslides inventory coming from IFFI Project is not a legal basis but it is one of the tools available that can be consulted. In any event, IFFI represents a very important tool for the planners who nally have the rst homogeneous, shared, detailed and most complete knowledge of the landslide occurrence on the whole territory. As a general remark for Italy, it has to be observed that public legislation denes general principles and lines of conduct, functions, activities and authorities involved, while the regional administrations apply restrictions on land use through different regional laws. Final remarks Laws or rules that indicate how a landslide analysis (danger, hazard, risk) has to be done, do not exist; There is often some confusion among danger, hazard and risk. An inventory map can be used as hazard map (i.e. susceptibility map), without any prevision of scenarios; There is some lack of trust in quantitative methods. Qualitative approach seems to be preferred; The technicians who make the maps have to think rstly: Who will be the end users? What will be the use of maps? Is the scale of work suitable for this? Are the complexity of methods (time, resources, needed input data) and results appropriate and understandable for decision makers? Literatur / References:
ARPA PIEMONTE, (2006), Note illustrative della Carta della Pericolosit per Instabilit dei Versanti alla scala 1:50,000 Foglio n. 211 Dego. (S. Campus, F. Forlati & G. Nicol editors), Apat, Roma. (in Italian); ARPA PIEMONTE, (2007), Evaluation and prevention of natural risks. (S. Campus, F. Forlati, S. Barbero & S. Bovo editors), Balkema Publisher; ARPA PIEMONTE, (2008), Interreg IIIa 2000-2006 Alpi Latine Alcotra. Progetto n. 165 PROVIALPProtezione della Viabilit Alpina. Final Report (in Italian); ARPA PIEMONTE, (2010), Geographic Information System on-line - http://webgis.arpa.piemonte.it V.A. (2004), Identication and mitigation of large landslides risks in Europe. The IMIRILAND project. (C. Bonnard, F. Forlati & C. Scavia editors), Balkema Publisher;

So complete coverage of basic information is available (lithology, geotechnical geo-database, landslides inventory, etc), but only few rigorous applications of hazard & risk assessment. One of the available tools produced by Arpa Piemonte is the regional part of Italian Landslides Inventory (IFFI). It is a national program of landslide inventory, sponsored by national

authorities and made locally by the regions. It is the rst try of an inventory based on common graphical legend and glossary. In Piemonte, over 35,000 landslides were recognized by interpreting aerial photos and eld surveys and the Informative System of Landslides is constantly updated with inclusion of new landslides or corrections and deepening of

Fig. 3: Arpa Piemonte Web-GIS Information Service of the IFFI Project. Abb. 3: ARPA Piemonte, Web-GIS Informationsdienst des IFFI-Projekts.

Seite 107

Hazard assessment and mapping of mass-movements in the EU

Seite 108

Zusammenfassung: Slowenien liegt in einem komplexen Raum Adria Dinaren Pannonisches Becken, und seine allgemeine geologische Struktur ist bestens bekannt. Aufgrund seiner auerordentlich heterogenen geologischen Lage ist Slowenien Hangmassenbewegungen (SMM = slope mass movement) sehr stark ausgesetzt. Die slowenische Gesetzgebung (und darauf beruhend auch die entsprechenden Manahmen) sind vorwiegend auf die Schadenbehebungsphase und die Begrenzung der Auswirkungen bereits aufgetretener SMM-Vorkommnisse ausgerichtet, es mangelt jedoch an vorbeugenden Manahmen. Der Zweck dieses Artikels ist die Prsentation von Gefahrenhinweiskarten ber Hangmassenbewegungen auf nationaler und regionaler Ebene, die zum Schutz vor schnellen Massenbewegungen in Slowenien erstellt wurden und die eine fachlich fundierte Grundlage fr die entsprechenden Prventivmanahmen bilden. Der nchste logische Schritt wre, dieses Know-how und diese Anstze in die Gesetzgebung zu integrieren. Schlsselwrter: Massenbewegungen, Gesetzgebung, Gefahrenhinweiskarte, Slowenien
MARKO KOMAC, MATEJA JEMEC but they can be mitigated or avoided, applying 1. Introduction Slovenian territory occupies the Eastern ank of the Alpine chain. As in other areas of the Alpine region, Slovenia is exposed to different slope mass movements (SMM) above the average of the rest of Europe. SMM that represent substantial problems can be generally divided into three groups, 1) landslides, 2) debris-ows, and 3) rock falls. The majority of SMM events cannot be prevented, adequate legislation measures supported by corresponding expert argumentation. Although Slovenian legislation (and hence also measures) mainly focuses on the remediation phase and mitigation of consequences of SMM events that have already occurred, its biggest deciency lays in the area of prevention measures. While, in the case of rare SMM events, the current approach of exclusively post-event measures is conditionally sustainable, in the case of frequent events it

Standards and Methods of Hazard Assessment for Rapid Mass Movements in Slovenia Standards und Methoden der Gefhrdungsanalyse fr schnelle Massenbewegungen in Slowenien
Summary: Slovenia is situated on the complex Adria Dinaridic Pannonian structural junction and its general geological structure is well known. As a consequence of an extraordinarily heterogeneous geological setting, Slovenia is highly exposed to slope mass-movement processes. While Slovenian legislation (and based on that also measures) mainly focuses on the remediation phase and mitigation of consequences of SMM events that have already occurred, its biggest deciency lays in the area of prevention measures. The purpose of this paper is to represent slope mass movement susceptibility maps on a national and a local level that have been developed for protection from rapid mass movements in Slovenia and which form an expert foundation for the prevention measures. The next logical step would be to incorporate this knowledge and approach into legislation.
Keywords: mass movement processes, legislation, susceptibility map, Slovenia

Fig 1: Relation between hazards on one side and elements at risk on the other, and the risk in between (after Alexander, 2002). Abb. 1: Beziehung zwischen Gefahren und gefhrdeten Elementen, und das dazwischen liegende Risiko (nach Alexander, 2002).

Seite 109

Hazard assessment and mapping of mass-movements in the EU

Seite 110

becomes unsustainable and brings a huge burden to the local, regional and state budget. The only reasonable approach would hence be minimising interaction between SMM events and elements at risk. Graphically this interaction would be presented as a cross-section between the natural hazard on one side and vulnerability of elements at risk on other side (Fig 1). 2. Legislation in the eld of slope mass movement domain In the area of systematic prevention measures regarding SMM, Slovenia lags behind other Alpine countries or regions. One of the basic approaches to solve the problem is to establish potentially hazardous areas due to natural phenomena and the inclusion of this information in spatial plans. Information on geology, upon which the slope mass movement occurrence heavily depends, it is not yet an integral part of spatial plans. Legislative acts deal mostly with remediation issues instead with the prevention measures. The protection strategy against landslides (within legislation the term landslide also other types of slope mass movements are included) varies substantially and is tailored according to different terrain conditions. They are mainly divided into prevention, emergency protective measures and permanent measures adopted in the process for remediation. In the frame of preventive actions, the emphasis is on creating a national database of active landslides (and other SMM) and intentions of government to include hazards doe to landslides into spatial planning. In the planning and implementation of emergency protective measures, the emphasis is on protecting human lives and property.

Law on protection against natural and other disasters (Ofcial Gazette of RS, no. 64/94)

Water Act (Ofcial Gazette RS, no. 67/02, 4/09) Protection against the harmful effects of water

included in the fth development priority, which is designed to achieve sustainable development.

The Act governs the protection against natural and other disasters and includes the protection of people, animals, property, cultural heritage and environment against any hazard or accidents (risk) that can threaten their safety. The main goal of the protection against natural and other disasters system is to reduce the number of disasters, and to forestall or reduce the number of victims and other consequences of disaster. The basic tasks of the system are: prevention, preparedness, and protection against threats, rescue and help, providing of basic conditions for life, and recovery. National program of protection against natural and other disasters (Ofcial Gazette of RS, no. 44/02) On the basis of the Resolution, the National Programme of Protection against Natural and Other Disasters for the period 2002 2007. The National Programme is oriented towards the prevention and its basic aim is to reduce the number of accidents and to prevent or minimise its consequences.

that is among other the issues dealt with this act also refers to protection against landslides. Threatened area is dened by Government, which is responsible for protecting the population, property and land in dangerous exposed areas. In order to protect against the harmful effects of water, land in the threatened area is categorized into classes based on the risk. Act on measures to eliminate the consequences of certain large-scale landslides in 2000 and 2001 (Ofcial Gazette RS, no. 21/02, 92/03, 98/05) Act denes the format and the method of nancing and form of allocating state aid for the implementation of remedial measures, to prevent the spread of landslide and stabilization of landslides on the specic area of inuence. It covers several major landslides in Slovenia. Spatial Development Strategy of Slovenia (Ofcial Gazette of RS, no. 76/04) The Spatial Development Strategy of Slovenia is a

Regulation of the spatial order of Slovenia (Ofcial Gazette of RS, no. 122/04) Regulation of spatial order in Slovenia provides the rules for managing the eld of landslide problematic. One of the important articles is Article 67, in which is mentioned how to plan according to the limitations which are caused by natural disasters and water protection. Resolution of the National Environmental Act (Ofcial Gazette of RS, no. 2/06) The National Environmental Action Programme (NEAP) is the basic strategic document in the eld of environmental protection, aimed at improving the overall environment and quality of life and protection of natural resources. NEAP was prepared under the Environmental Protection Act and complies with the European Community Environment Programme, which addresses the key environmental objectives and priorities that require leadership from the community. The objectives and measures are dened in the four areas, namely: climate change, nature and biodiversity, quality of life, and waste and industrial pollution. 3. Methodology Due to specics of different slope mass movement processes, a single approach would be hampered in its results / prognosis. The following chapter presents an overview of approaches to slope mass movements (1 landslides; 2 debris-ows; 3 rock falls) hazard assessment. The presented approaches are similar to a certain level, they also differ according to the scale of the assessment. The

Law on the Remediation of consequences of natural disasters (Ofcial Gazette of RS, no. 114/05) The Act denes a landslide as a natural disaster. According to the article 11, with some restriction and at some level of damage, state budget funds may be used to ease the effects of natural disasters. Damage assessment is made in accordance with the Regulation on the methodology for damage assessment (Ofcial Gazette of RS, no. 67/03, 79/04), after which the landslide is considered a landslide, which threats a property or infrastructure.

public document guiding development in the eld of landslide problematics. It provides a framework for spatial development throughout the country and sets guidelines for development in European space. It provides for the creation of spatial planning, its use and conservation. The spatial strategy takes into account social, economic and environmental factors of spatial development. Slovenia's Development Strategy Slovenia's Development Strategy sets out the vision and objectives of Slovenia and ve development priorities with action plans. The chapter on protection against natural disasters is

Seite 111

Hazard assessment and mapping of mass-movements in the EU

Seite 112

nal results (but not the only ones) of approaches presented in the following text were presented in a form of warning maps that are still the main product used by end users. All the analyses were conducted in GIS, which enables the end users to implement results also in a form of databases or a digital format. According to Skaberne (2001) the terminology of slope mass movements in Slovenia are as follows: landslides are processes of translational or rotational movement of rock or soil as a consequence of gravity at discontinuity plane(s). Rock falls are processes of falling or tumbling of a part of rock or soil along a steep slope. Debris-ows are processes of transportation of material composed of soil, water and air. The landslide susceptibility model for Slovenia at scale 1:250,000 was developed at the Geological Survey of Slovenia in 2006 (Komac & Ribii, 2006). The nal result of this approach was presented in a form of a warning map (Fig. 2). Based on the extensive landslide database that was compiled and standardised at the national level, and analyses of landslide spatial occurrence, a Landslide susceptibility map of Slovenia at scale 1 : 250,000 was completed. Altogether more than 6,600 landslides were included in the national database, of which roughly half are on known locations. Of 3,257 landslides with known locations, random but representative 65% were selected and used for the univariate statistical analyses (2) to analyse the landslide occurrence in relation to the spatio-temporal precondition factors (lithology, slope inclination, slope curvature, slope aspect, distance to geological boundaries, distance to structural elements, distance to surface waters, ow length, and land cover type) and in relation to the triggering factors (maximum 24-h rainfall, average annual rainfall intensity, and peak ground acceleration). The analyses were conducted using

GIS in raster format with a 25 25 m pixel size. Five groups of lithological units were dened, ranging from small to high landslide susceptibility. Furthermore, critical slopes for the landslide occurrence, other terrain properties and land cover types that are more susceptible to landsliding were also dened. Among triggering factors, critical rainfall and peak ground acceleration quantities were dened. These results were later used as a basis for the development of the weighted linear susceptibility model where several models with various factor weights variations based on previous research were developed. The rest of the landslide population (35 %) was used for the model validation. The results showed that relevant precondition spatio-temporal factors for landslide occurrence are (with their weight in linear model): lithology (0.3), slope inclination (0.25), land cover type (0.25), slope curvature (0.1), distance to structural elements (0.05), and slope aspect (0.05). Beside landslide susceptibility assessment, a rainfall inuence on landslide occurrence was analysed since rainfall plays an important role in the landslide triggering processes. Analyses of landslide occurrences in the area of Slovenia have shown that areas where intensive rainstorms occur (maximal daily rainfall for a 100-year period), and where the geo-logical settings are favourable an abundance of landslide can be expected. This clearly indicates the spatial and temporal dependence of landslide occurrence upon the intensive rainfall. Regarding the landslide occurrence, the intensity of maximal daily and average annual rainfall for the 30 years period was analysed. Results have shown that daily rainfall intensity, which signicantly inuences the triggering of landslides, ranges from 100 to 150 mm, most probably above 130 mm. Despite the vague inuence, if any at all, of the average annual rainfall, the threshold above which signicant number of landslides occurs is 1000 mm.
Fig. 2: Landslide susceptibility warning map of Slovenia at scale 1:250,000 (Komac & Ribii, 2006, 2008). Abb. 2: Gefahrenhinweiskarte fr Rutschungen in Slowenien im Mastab von 1:250.000 (Komac & Ribii, 2006, 2008).

The debris-ow susceptibility model for Slovenia at scale 1:250,000 was also developed at Geological Survey of Slovenia in 2009 (Komac et al., 2009). The nal result of this approach was presented in a form of a warning map (Fig. 3). For the area of Slovenia (20,000 km2), a debrisow susceptibility model at scale 1:250,000 was produced. To calculate the susceptibility to debrisow, occurrences using GIS several information layers were used such as geology (lithology and distance from structural elements), intensive rainfall (48-hour rainfall intensity), derivates of digital elevation model (slope, curvature, energy potential related to elevation), hydraulic network (distance to surface waters, energy potential of streams), and locations of sixteen known debris ows, which were used for the debris-ow susceptibility models evaluation. A linear model-

weighted sum approach was selected on the basis of easily acquired spatio-temporal factors to simplify the approach and to make the approach easily transferable to other regions. Based on the calculations of 672 linear models with different weight combinations for used spatio-temporal factors and based on results of their success to predict debris-ow susceptible areas, the best factors weight combination was selected. To avoid over-tting of the prediction model, an average of weights from the rst hundred models was chosen as an ideal combination of factor weights. For this model an error interval was also calculated. A debris-ow susceptibility model at scale 1:250,000 represent a basis for spatial prediction of the debris-ow triggering and transport areas. It also gives a general overview of susceptible areas in Slovenia and gives guidance for more detailed

Seite 113

Hazard assessment and mapping of mass-movements in the EU

Seite 114

(4) Mapping of problematic areas at scale 1:5000 or 1:10,000 for the purpose of the highest detail planning

(3) Development of detailed geohazard map at scale 1:25,000 as a combination of synthesis of phases (1) and (2)

(1) Synthesis of archive geological data into the overview geohazard map at scale 1:25,000 (2) Development of statistical geohazard at scale 1:25,000
Fig. 3: Debris-ow susceptibility warning map of Slovenia at scale 1:250,000 (Komac et al., 2009). Abb. 3: Muren-Gefahrenhinweiskarte Sloweniens im Mastab von 1:250.000 (Komac et al., 2009). Fig. 4: Schematic diagram of the process of production of landslide and rock-fall susceptibility at the municipal scale (1:25.000) (Bavec et al., 2005). Abb. 4: Schematische Darstellung der Erstellung von Gefahrenhinweiskarten ber Erdrutsch, Berg- und Felssturz im Mastab einer Wanderkarte (1:25.000) (Bavec et al., 2005).

research areas and further spatial and numerical analyses. The results showed that approximately 4% of Slovenias area is extremely high susceptible and approximately 11% of Slovenias area of susceptibility to debris-ows is high. As expected, these areas are related to mountainous terrain in the NW and N of Slovenia. In the frame of a research project, slope mass movement geohazard estimation The Bovec municipality case study an approach to assess the landslide and rock-fall susceptibility at the municipal scale (1:25,000) (Bavec et al, 2005; Komac, 2005). The production of a susceptibility map that should represent (ofcially not included among the documentation yet) one of basic layers in the spatial planning process shown in the Fig. 4. Methodology was developed for estimation of geohazard induced by mass movement

processes, taking the Bovec municipality as the case study area. The geohazard map at the scale 1:25,000 as the nal product is aimed to be directly applicable in spatial planning of local communities (municipalities). The requirements that were followed to achieve this aim were: expert correctness, reasonable time of elaboration, and easy to read product. Elaboration of the nal product comprises four consecutive phases, of which the rst three are done in the ofce: 1) synthesis of archive data, 2) probabilistic model of geohazard induced by mass movement processes, 3) compilation of phases 1 and 2 into the nal map at scale 1:25,000. As the last phase, eld reconnaissance of most hazardous areas is foreseen. The susceptibility model development was based on the upgrading of the expert geohazard map at scale 1:25,000 with a probabilistic model

development that included relevant inuence factors. For analytical purposes, 10,816 models were developed: 3,142 for landslide susceptibility and 7,674 for rock-fall susceptibility. In both cases, geology/lithology and slope angle showed to be the most important inuencing factors. Regarding landslides, additional important factors were land use and synchronism of strata bedding and slope aspect, and in the case of rock-falls an additional important factor was synchronism of strata bedding and slope aspect. The methodology is focused towards the direct use of the nal product in the process of spatial planning at the municipal level and is divided into four phases as shown in Fig. 4: (1) Synthesis of archive geological data

in the overview geohazard map at scale 1:25,000 (Budkovi, 2002). (2) Development of statistical geohazard at scale 1:25,000 (Komac, 2005). (3) Development of detailed geohazard map at scale 1:25,000 as a combination of synthesis geological map (1) and statistical geological model (2) and delineating the most problematic areas. (4) Mapping of problematic areas at scale 1:5,000 or 1:10,000 for the purpose of the highest detail planning. All presented approaches are based on a probability statistical model that is a part of a conceptual development model of general or detailed slope mass susceptibility maps represented in Fig 5.

Seite 115

Hazard assessment and mapping of mass-movements in the EU

Seite 116

Anschrift der Verfasser / Authors addresses:


Univariate analysis (x2) of SMM occurrence by classes within each of the inuence factor Inuence factors classes ranging based upon their inuence on the SMM occurrence Values normalisation within each inuence factor (0-1)

Literatur / References:
ALEXANDER, D.E., 2002. Principles of emergency planning and management. Oxford University Press, New York, 340 pp. BAVEC, M., BUDKOVI, T. AND KOMAC, M., 2005. Estimation of geohazard induced by mass movement processes. The Bovec municipality case study. Geologija, 48/2, 303-310. BUDKOVI, T., 2002. Geo-hazard map of the municipality of Bovec. Ujma, 16, 141-145. KOMAC, M. 2005. Probabilistic model of slope mass movement susceptibility - a case study of Bovec municipality, Slovenia. Geologija, 48/2, 311-340. KOMAC, M., RIBII, M., 2006. Landslide susceptibility map of Slovenia at scale 1:250,000. Geologija, 49/2, 295-309. KOMAC, M., KUMELJ, . AND RIBII, M., 2009. Debris-ow susceptibility model of Slovenia at scale 1: 250,000. Geologija, 52/1, 87-104. SKABERNE, D., 2001. Prispevek k slovenskemu izrazoslovju za pobona premikanja. Ujma, 1415, 454458.

Bad results Testing of different models developed on the weighted sum of inuence factors

Fig 5: Conceptual model of development of general or detailed slope mass susceptibility maps. Abb. 5: Konzeptionelles Modell fr die Entwicklung von allgemeinen oder detaillierten Gefahrenhinweiskarten ber Hangbewegungen.

Marko Komac Dimiceva ulica 14 1000 Ljubljana SI-Slovenia Marko.komac@geo-zs.si Mateja Jemec Dimiceva ulica 14 1000 Ljubljana SI-Slovenia Mateja.jemec@geo-zs.si

Field testing

Selection of optimal and most logical susceptibility model

Development of phenomenon susceptibility map

For all inuence factors included in the weighted sum model calculation, original values were transformed into the same scale, which ranged from 0 1 to assure the equality of the input data. In other words, within each factor original values were normalised with the eq. 1. (RV - Min) NVR = , Max - Min eq. 1

or discreet variable value. Final slope mass movements susceptibility values (the range is between 0 and 1) were classied into 6 susceptibility classes: 0 Negligible (or None); 1 Insignicant (or Very Low); 2 Low; 3 Medium (or Moderate); 4 High; 5 Very High. 4. Conclusion Slope mass movement processes are specic in their nature, hence separate analyses had to be performed and a different model development had to be developed. In Slovenia, slope mass movement susceptibility maps have been developed on national and on local level. In the case of the latter, which has an actual application, value maps were developed only for some test areas. Thus several questions remain open and these are: when will the geohazard layer be included as a compulsory part of the spatial planning document, to what extent quality geological data will be used for the assessment, and how the lack of detailed geological data would be tackled.

Where NVR represents new and normalised value, and RV the old (nominal) value. Min and Max represent the minimum and maximum original value within the factor, respectfully. For the purpose of the development of the best and at the same time the most logical susceptibility model, a weighted sum approach (Voogd, 1983) was used (eq. 2). Where wj x fij H= eq. 2. j=l H represents standardised relative n

phenomenon susceptibility (0 1), wj represents the factor weight, and fij represents a continuous

Seite 117
res od Go s ult

Hazard assessment and mapping of mass-movements in the EU

Seite 118

KARL MAYER, ANDREAS VON POSCHINGER

Standards and Methods of Hazard Assessment for Geological Dangers (Mass Movements) in Bavaria Standards und Methoden zur Verminderung von geologischen Gefhrdungen durch Massenbewegungen in Bayern
Summary: Information about geological hazards in the Bavarian Alps (e.g. rock falls, landslides) is available in the Internet or intranet section Georisk of the Bodeninformationssystem Bayern (BIS-BY) (www.bis.bayern.de). This information system is already used by a number of departments such as district administrations, water and trafc management ofces, forest management as well as private users. By now the BIS-BY only shows the sites of origin of geological hazards and not the whole endangered area, which would be relevant for land use planning. This area, the so called process area, can only be dened by empirical or numerical simulations and models. A hazard map gives an overview of the situation. It is based on model calculations and empirical analysis and can be veried by the Georisk-cadastre (BIS-BY). Concerning the spatial extent of the process areas, possible inaccuracies may impair an exact expression of the danger. The hazard map shows large areas where a special type of danger can be assumed. Therefore, will be easier to deduce possible conicts between hazards and land use. Hazard maps can be included in the land development plan or can be used to assign priorities while taking measures.

Zusammenfassung: Informationen ber geogene Gefhrdungen (z.B. Steinschlag, Felsstrze, Rutschungen) sind als GEORISK-Daten ber das Bodeninformationssystem Bayern (BIS-BY) im Internet oder Intranet abrufbar (www.bis.bayern.de). Dieses Informationssystem wird bereits von vielen Fachstellen genutzt. Neben den Landkreisen sowie vielen Kommunen sind die Behrden der Wasserwirtschaft, der Straen- und Forstverwaltung sowie private Planer die Hauptnutzer. Im BIS-BY ist bisher allerdings nur das Herkunftsgebiet von Gefhrdungen dargestellt, nicht der planungsrelevante Gefhrdungsbereich. Dieser kann nur durch empirische oder numerische Simulationen und Modellierungen abgegrenzt werden. Die Gefahrenhinweiskarte gibt eine bersicht ber die Gefhrdungssituation. Sie basiert sowohl auf Modellrechnungen als auch auf empirischen Untersuchungen und wird mit dem GEORISK-Ereigniskataster (BIS-BY) auf Plausibilitt geprft. Bezglich der rumlichen Abgrenzung kann sie Ungenauigkeiten enthalten und die Gefhrdung nicht in jedem Fall genau wiedergeben. Die Gefahrenhinweiskarte hlt fr groe Gebiete chendeckend fest, wo mit welchen Gefahren gerechnet werden muss. Daraus lassen sich mit geringem Aufwand mgliche Koniktstellen zwischen Gefahr und Nutzung ableiten. Die Gefahrenhinweiskarten knnen einerseits in Flchennutzungsplne mit einieen und dienen anderseits zur Priorittensetzung beim Erarbeiten weitergehender Manahmen.
Main data of the topic mass movements and 1. Introduction In Germany, geogenic natural hazards, such as mass movements, karstication, large scale ooding as well as ground subsidence and uplift affecting building ground, shall be recorded, assessed and spatially represented using a common minimal standard in the future. For this purpose, the Geohazards team of engineering geologists of the different German federal governmental departments of geology (SGD) are giving recommendations on how to create a hazard map. These recommendations of minimum requirements are directed at the employees of the SGD. An important component for developing hazard maps is the construction and evaluation of landslide inventories (e.g. landslide or sinkhole inventories). The recorded data in the inventories should have a minimal nationwide standard and are divided into: subrosion / karst with information about the spatial positioning, about determination of coordinates, etc. Commonly shared technical data of the subject mass movements and subrosion / karst with information about the date of origin, about the land use and about damage, etc. Specic technical data of the subject mass movement and subrosion / karst Data concerning subsidence and uplift Computerized modelling increasingly allows the identication of hazard areas that have been veried using the landslide inventory or through evaluation of the results of eld work. The current emphasis in Germany is on hydrological modelling of ood events that are used for water management issues in ood prevention. Geotechnical modelling is used increasingly for rock falls, avalanches and shallow landslides.

Seite 119

Hazard assessment and mapping of mass-movements in the EU

Seite 120

If necessary, in addition to the tools described above, eld studies will be needed for exact clarication and assessment of given situations. In Alpine regions, natural hazards are a common phenomenon. Landslides, rock falls and mudows occur in the course of mountain degradation that reects the natural slope instability of mountain areas. Landslides are mostly triggered by extreme rainfall that will, according to climate scientists, become more relevant in Alpine regions in particular (Umweltbundesamt 2008). With an increase in heavy rainfall events an increase in landslide events must be expected. With approximately 4450 km, the Bavarian Alps cover about 6.3 % of Bavaria. The Bavarian Alps are the most important tourist region of Bavaria and, therefore, of particular importance. Furthermore, they have a unique ecological value that has to be specially protected. Since it is more and more difcult to ensure this protection by structural activities, protective measures need to be involved in the planning process and also allow sustainable and cost effective strategies. The most effective and sustainable method to prevent losses arising from hazardous events is to avoid land use in the endangered areas. In areas where construction already has been established or where construction of new infrastructure or buildings is unavoidable, it is essential to determine areas endangered by geological hazards. In May 2008, the Bavarian Environmental Agency launched the project hazard map for the Bavarian Alps. The aim of the project is to create a hazard map for deep seated landslides, shallow landslides and rock fall areas for the whole of the Bavarian Alps. It will be nished during December 2011.

2. Denition of a hazard map The federal geological surveys of Germany agreed on denitions for the terminology used for mapping of geological hazards (Personenkreis Geogefahren 2008) based on BUWAL (2005). A hazard map gives a rst overview of areas affected by landslides (potentially endangered area) and can be a basis for the detection of conicts of interests. By dening a most probable design event and integrating it in the landslide modelling process, a hazard map also gives a qualitative statement about the probability of a landslide event. The potential process areas of the expected landslides vary depending on the design event, the geological, topographical and morphological situation and the existence of forest. Modelling parameters for rock fall and shallow landslide simulations can be deduced and trivialised from comprehensive data. Generally the scale of a hazard map ranges from 1:10,000 to 1:50,000. Within this project, despite the possibilities of the zoom function of a GIS, the hazard map is produced for a scale of 1:25,000. 3. Material and methods 3.1 Basis maps Essential data basis for modelling the hazard map is a high resolution digital elevation model (DEM) derived from airborne laser scanning. The datasets are used in different resolutions (1 m, 5 m, 10 m) depending on the modelling approach. The vertical resolution is better +/- 0.3 m, except for very few areas where currently no laser scanning data is available.

3.2 Basis data for landslide modelling Information about geological hazards such as landslides, rock falls and earth falls, especially in the densely populated areas in the Bavarian Alps, is available in the section Georisk of the Bodeninformationssystem Bayern (BIS-BY, www.bis.bayern.de), a GIS-based inventory of Bavaria including numerous geological data. By now (October 2010), about 4,500 landslide events have been detected and evaluated within the project area. Every event is described concerning its process type and dimension, the age and potential future trend of the landslide as well as annotations about the source and the degree of information. Origin and accumulation zones of landslides have been digitised and stored as well as signicant photos. With all of this the BIS-BY is the most important source of information. Also integrated in the BIS-BY are maps of active areas that have been mapped by eld work, aerial photo analysis and archive data for the main settlement areas. Within these maps landslides are classied into four levels of activity to give an indirect statement about the level of danger. These maps can be used to estimate the extension of deep-seated landslides, for example. Above all, results of two other projects have been used: Within the project HANG (historical analysis of alpine hazards), historical data of landslides have been evaluated and digitised. Within the project EGAR (catchment areas in alpine regions), the risk potential of alpine torrents has been estimated analysing the discharge and catchment potential.

4. Fall processes 4.1 Minimum requirements in Germany In many states of Germany, only medium to long term, large-scale numeric modelling of rock fall hazards are possible using high resolution terrain models and specialised software. In the rst stage, a black and white map is created showing veried / potential rock fall areas derived from the landslide inventories and / or remote sensing (DEM). This map shows veried as well as potential rock fall escarpments i.e. slopes with an inclination > 45 (in Alpine areas). The entire process area is, however, not depicted. In the second stage, the run-out zone, i.e. the entire process area, is depicted. That means areas prone to rock falls due to the inclination, but which are not conrmed. To dene these areas, estimated empiric angle methods or physical deterministic models can be used. To determine rock fall escarpments, the shadow angle and the geometric slope angle is applied. Both the shadow angle (e.g. 27) as well as the geometric slope angle (e.g. 32) can be used as the estimated angle (Mayer & Poschinger 2005). An angle of deection from the vertical slope can be used as a lateral boundary of the process area (e.g. 30). In Bavaria this method is used for huge rock masses. For single blocks, a physical trajectory model from Zinggeler + GEOTEST is used (MAYER 2010). 4.2 Modelling rock fall of single blocks (methods use in Bavaria) For the detection of potential starting zones of rock falls, two empirical approaches can be applied. In a rst step, potential starting zones

Seite 121

Hazard assessment and mapping of mass-movements in the EU

Seite 122

stored in the BIS-BY are extracted. These starting zones are detected by eld work. In areas where no information is available, an even more empiric approach must be applied: it has to be assumed that every slope steeper than 45 is a potential detachment zone (Wadge et al. 1993).

by eld work. As a result, a mean block size and geometry that represents the most probable event has been determined for every geological unit. This design event has been assigned to one of four volume classes. For each of these classes the mean block mass has been calculated. The block mass of a geological unit is an input parameter for the simulation. The simulation of the block movement is carried out according to physical principles of mechanics and is separated into falling, bouncing and rolling (Fig. 1). The calculation is a succession of these processes with intermediate contacts to underground and tree trunks. The loss of energy during tread mat

4.3 Modelling rock fall masses (Bavarian approach) The trajectory model for rock fall (chapter 4.2) calculates the reach of single blocks. For the runout zone of larger rock fall volumes, an empirical process model with a worst case approach is used. Numerous papers (Lied 1977, Onofri & Canadian 1979, Evans & Hungr 1993, Wieczorek et al. 1999, Meil 1998) show that a global angle method is an appropriate approach to determine the maximum run-out zone of rock fall. Two different global angles have been applied. The rst and more important one is the shadow angle ( in Fig. 3). It is dened as angle between the horizontal line and the connecting line from the block with maximum run out and the top of the talus. According to Evans & Hungr (1993) a shadow angle of 27 has been assumed. The other global angle is the geometrical slope angle that spans between the horizontal line and top of detachment zone ( in Fig. 3). A minimum geometrical slope angle of 30 is presumed (Meil 1998).

The application of the different global

angles depends on slope morphology. A proper decision for one global angle model can be reached by the quotient of shadow angle tangent and geometrical slope tangent. If the quotient is below 0.88, the shadow angle has to be used. Otherwise the geometrical slope angle is better suited to describe the maximum run-out zone (Mayer & von Poschinger 2005). Global angles can easily be modelled with implemented functionalities of standard GIS programs. Within the project, the viewshed function of Spatial Analyst in ArcGIS has been employed. This function identies all cells on a surface (DEM) that can be seen from selected observation points (Fig. 4). There are a number of important attributes of every starting point necessary for the modelling process: the vertical view angle, which is the predened global angle (Fig. 3), the horizontal view angle that is dened with 30, as well as the aspect that can be calculated out of the DEM.

Fig. 1: Basic processes during rock fall simulation (Krummenacher et al. 2005). Abb. 1: Schematische Darstellung der prinzipiellen Prozesse der Steinschlagmodellierung (Krummenacher et al. 2005).

is controlled by deformability and surface roughness. These parameters have to be deduced and trivialised from the basis data of the area to be investigated.

According to Meil (1998) or Hegg &

Kienholz (1995) the process model can be divided into two parts: the trajectory model calculating the paths of the blocks as vectors and the friction model calculating the energy along these paths as well as the run-out length. In this project, the vector based simulation model of Zinggeler & GEOTEST (Krummenacher et al. 2005) is used. Beside the topographical information derived from the DEM, damping and friction characteristics of the slope surface and the vegetation have to be known. Furthermore it is very important to dene a design event for rock fall. That means that, according to the geology, form and dimension of typical blocks have to be determined. As the block dimension is the only variable parameter within the simulation, it plays an essential role in the calculation of the run- out zone. To assess the design events, the starting zones already determined within the disposition model have been intersected with the geological map. The affected geological units have been checked
Fig. 2: 3D Trajectories with (red) and without (orange) the protecting function of forest. Abb. 2: 3D Sturztrajektorien mit (rot) und ohne (orange) Bercksichtigung der Schutzfunktion des Waldbestandes.

The simulation has been run for two

different scenarios. Within the rst scenario, the forest with the protecting function of the trees has been considered. To simulate a worst-case scenario, the forest has not been included in the second scenario.
Fig. 3: Global angle models: shadow angle () and geometrical slope angle () (Meil 1998, modied). Abb. 3: Pauschalgefllemodelle: Schattenwinkel () und Geometrisches Geflle (), verndert nach Meil (1998).

Seite 123

Hazard assessment and mapping of mass-movements in the EU

Seite 124

To identify of hazard areas, only important rock fall areas with evidence of activity have been processed. Due to long-lasting eld work, there is an excellent overview of the situation within the densely populated areas in the Bavarian Alps. Beyond those areas it is assumed that all important rock fall areas are known. To start the modelling process, rst the global angle approach has to be chosen (shadow angle or geometrical angle). After digitizing the starting points and determination of necessary attributes, the viewshed modelling with ArcGIS can be executed. 5. Slide processes 5.1 Minimum requirements in Germany In the rst stage, landslide inventories, e.g. all registered objects and the associated near-surface processes, should be visually displayed. That means affected by denite indications of active and inactive landslides and landslides that have already occurred (reactivation or enlargement of the landslide area is possible). The areas can be found using mapping (registers) or remote sensing (DEM) methods.

In

the

second

stage,

potential

demonstrated that deep-seated landslides mostly occur in areas already affected by landslides in the past. For this reason they can be used as design events. To detect these areas, information about known landslides, extracted from the databases listed in chapter 3.2 has to be evaluated. Permanent activity or more or less recurrent reactivation likely produces enlargement of the landslide area identied in the disposition model, both the detachment and run-out zone upward and downward. Since a numeric modelling of deep seated

more related to water-related hazards and for this reason not explained here in detail. The deep-seated landslides are handled in the same way as the slide processes. The process occurring in the run-out zone of shallow landslides is also mostly a ow process. To estimate this process as disposition model in Bavaria, the physical computer model SLIDISP is used. To nd the run-out zones and to simulate the process, the model SLIDEPOT (GEOTEST) is applied. 6.2 Modelling shallow landslides (methods used in Bavaria) Shallow landslides are usually triggered by heavy rainfall, depending on the predisposition of the slope. Like the rock fall simulation, the modelling of shallow landslides is carried out in two steps. The starting zones are calculated in the disposition model and the run-out zones are calculated in the process model. For the disposition model, the deterministic numerical model SLIDISP (Liener 2000 and GEOTEST AG) is used. This assumes an above average precipitation for a certain area. The Innite-Slope-Analysis is applied to calculate the slope stability for every raster cell. Fundamental basic data are the slope angle, derived from the DEM from which the thickness of soil will be deduced and the geology which allows to determine friction angle and cohesion as geotechnical parameters. The factor of safety F will be calculated for every raster cell to describe the ratio of retentive and impulsive forces (Fig. 5, Selby 1993). The natural range in the variation of a Monte-Carlo-Simulation. For every different input parameters will be considered using raster cell, the number of instable cases will be determined. The higher the number of instabilities the higher is the probability of slope failure. Since the occurrence of forest affects the stability in an enormous way, the root strength will be

landslide areas are determined in addition to the veried landslide areas. That means areas prone to landslides due to the geological and morphological situation and the land use (were landslides have not yet taken place). These areas can be found by using empirical methods due to the geological and morphological circumstances and the land usage; alternatively / additionally: Visualisation of semi-automatically derived areas (cross-over between DEM / geological entity); e.g. using an additional signature The distinction between shallow and deep-seated slides is optional when visualising the hazard map. Near-surface landslides of a small volume (shallow slides) are either separately determined using above procedure or are displayed simultaneously alongside the deepseated slides. 5.2 Modelling deep seated landslides (methods used in Bavaria) Deep-seated landslides are mostly result of the activation of predened failure zones, i.e. by long lasting rainfall. Experience shows that they can range from about 5 m up to more than 100 m in depth. To identify areas endangered by deep seated landslides, two different approaches have been applied. On the one hand, areas showing evidence of previous deep-seated landslides, with either ongoing activity or a clear probability of reactivation, have been evaluated. On the other hand, the terrain has been evaluated concerning an increased susceptibility for deep-seated

landslides is not available for a regional scale, the determination of the potential process area has to be worked out with empirical methods, taking into account the local geology and morphology. Under extreme conditions, the process area can reach the next ridge, terrace or depression in the greater surroundings of the landslide. In the case of small-scaled scars in smooth slopes, a margin of 20 30 m has been added to the detachment areas to assess the potential process area. To determine the potential run out of an active or reactivable landslide, the present runout length has been determined by databases, hillshades and eld work in a rst step. If there are indications for active movements in the landslide toe, it is assumed that the run-out length will proceed even further in case of a reactivation. The danger area has to be dimensioned according to geomorphologic conditions. 6. Flow processes 6.1. General approach The procedure and depiction of ow processes like deep-seated landslides (Talzuschub) is similar to the method used for slide processes. Flow processes rarely occur in low mountain ranges. In the German Alpine area, debris ows are

landslides.
Fig. 4: The viewshed function identies all raster locations to be seen from appointed starting points with dened global angle. Abb. 4: Die Viewshed-Funktion ermittelt alle Bereiche, die von festgelegten Startpunkten mit einem denierten Vertikalund Horizontalwinkel gesehen werden.

The locality of the origin of danger (areas

showing a higher probability for the development of a deep seated landslide) has been identied within the previously cited disposition model. Previous experiences and analysis have

Seite 125

Hazard assessment and mapping of mass-movements in the EU

Seite 126

inuences on karstication, can be noted in an additional category. Optionally, a differentiation between carbonate, sulphate and chloride karstication can be implemented in the rst or second stage of the hazard map. If the information is available in individual states, the spread of the inner and outer salt slopes as well as intact salt domes should be entered into the hazard map. 8. Discussion The hazard map has been worked out for a regional scale (1:25,000). Therefore the boundaries of the hazard areas are not sharply bounded lines and
Fig. 6: Calculation of accumulation: for the central cell with exposition of 210 230, the 20 sector identies 3 cells that are either starting zones or already show accumulation (orange cells). Fig. 5: Principle for the calculation of the factor of safety F for every raster cell (Selby 1993). Abb. 5: Grundlagen zur Berechnung des Sicherheitsgrades F einer Rasterzelle (Selby 1993). Abb. 6: Berechnung der Auslaufbereiche: Fr die Rasterzelle in der Mitte mit der Zellexposition 210230 wurden drei Rasterzellen im Sektor von 20 ermittelt, die sowohl Anbruchzone als auch Auslaufbereich sind (orange Rasterzellen).

a detailed view on particular areas or objects is not allowed. In addition, the modelling of the different processes can make no claim to be complete. The maps show potentially endangered areas that have been determined on the basis of available information and that has been computed with modern numerical models. Anthropogenic preventive measures have not been introduced into the models. Improbable and extreme events have not been considered. Instead, frequently occurring events have been modelled since they are more representative and felt more as a risk. From a geological view, rare and extreme events have to be accounted as an unavoidable residual and remaining risk. The hazard maps for rock fall of single blocks and rock fall masses and deep-seated landslides are based on eld work for the most part. On the contrary, the hazard areas of shallow landslides are solely based on computer models and represent a typical susceptibility map. Therefore, they are presented as hatched areas. In the eld, witnesses of former traces of shallow landslides are hard to nd due to weathering. However, if the predicted consequences of

integrated in the calculation of the factor of safety as an additional parameter. Considering the root strength and its effect on soil stability it is possible to simulate two scenarios with different intensities of the root effect (high and low). To calculate the run-out zones. the raster-based model SLIDEPOT is used (GEOTEST AG). For every raster cell in the starting zone, the accumulation will be modelled in the ow direction. The model is based on neighbourhood statistics. Above a potential accumulation cell, the raster cells inside a 20 sector will be analysed (Fig. 6). Accumulation will be calculated if there is a starting zone and if the topography in the sector named above is not convex. Every step of expansion will analyse the neighbourhood up to a dened distance (4 cells; red circle in Fig. 6). With every step, the hypothetical starting volume and the rest volume will be reduced by a degradation factor, which depends foremost on the slope

angle. The expansion stops if a dened number of expansion steps is achieved or if the calculated value falls below a dened threshold. The run-out zones will be calculated for both scenarios. In both cases, a maximum of 8 expansion steps have been calculated while the degradation factor has been reduced in the forest. Because of uncertainties concerning complex edge conditions, the degradation factors have been dened quite pessimistically. With this the run-out zones are large enough and rather too large in the case of doubt. 7. Subrosion / karstication Supercial or near-surface subrosion features (sinkholes) and the knowledge of subrodable sediments serve as criteria for the analysis of a process area. In the rst stage, the following hazard areas are distinguished:

Veried karstication features from the

Geological map, event register or remote sensing (e.g. DEM) methods. In the rst stage, supercial or near-surface subrosion features (e.g. sinkholes, depressions, clefts) are visualised. There is no differentiation between fossil and current subrosion features. The second stage includes the visualisation of the dispersion of karstiable sediments. Hazard elds can be derived using a point or area statistical evaluation (e.g. using the feature density or a raster based density calculation), as well as using inuencing factors, such as geology, tectonics and hydrogeology. The result of the second stage determines the differentiation of hazard areas. Where applicable, the hazard areas can be coupled with general geotechnical recommendations as to construction work in karst landscapes. Special conditions in individual states, e.g. mining

Seite 127

Hazard assessment and mapping of mass-movements in the EU

Seite 128

climate change with an increase in extreme rainfalls will come true, an increasing number of shallow landslides must be taken into account. Climate change predictions could be implemented in the model if maps with predicted precipitation on a local scale were available. This would allow the identication of hot spots with heavy rainfall and, therefore, a higher susceptibility for landslides. The identication of such hot spots is one target in the Alpine Space Programme project AdaptAlp that also focuses on evaluation, harmonizing and improvement of different methods for hazard mapping. 9. Conclusions A hazard map is a very helpful tool for planning authorities to get an overview about land use conicts and potentially endangered areas. It is a general map created under objective scientic criteria and indicating geological hazards that have been identied and localized but not analysed and evaluated in detail. A hazard map does not contain specications about the degree of hazard or the intensity or probability of an event. The map will be provided to local and regional planning authorities for water, trafc, and forest management. It helps the planner identify hot spots and make decisions concerning measures of protection. On the other hand, it also shows areas not endangered and free for planning. In critical cases, the hazard map has to disclose the requirement for further analysis. In this cases a detailed expertise analysis has to decide if measures are technically feasible, economically reasonable and under sustainable aspects really necessary.

To help potential users interpret the

hazard map, the results are presented to all authorities. Furthermore, an intensive cooperation with the Bavarian Environment Agency is offered. In addition, a limited version of the hazard map is published on the Internet (www.bis.bayern.de). But the Alpine part of Bavaria is not the only region affected by geological hazards. The Alpine foothills and the Swabian-Franconian Jurassic-mountains are affected as well. For the mid-term, the goal is to develop hazard maps for the whole of Bavaria.

KIENHOLZ, H., ERISMANN, TH., FIEBIGER, G. & MANI, P. (1993): Naturgefahren: Prozesse, Kartographische Darstellung und Manahmen. In: Tagungsbericht zum 48. Deutschen Geographentag in Basel, 293 312, Stuttgart. KRUMMENACHER, B., PFEIFER, R., TOBLER, D., KEUSEN, H. R., LINIGER, M. & ZINGGELER, A. (2005): Modellierung von Stein- und Blockschlag; Berechnung der Trajektorien auf Prolen und im 3-D Raum unter Bercksichtigung von Waldbestand und Hindernissen. anlsslich Fan-Forum ETH Zrich am 18.02.2005, 9 p., Zollikofen. LIED, K. (1977): Rockfall problems in Norway. In: Istituto Sperimentale Modelli e Strutture (ISMES), 90: 51-53, Bergamo. LIENER, S., (2000): Zur Feststofieferung in Wildbchen. Geographisches Institut Universitt Bern. Geographica Bernensia G64, Bern. MAYER, K. & VON POSCHINGER, A. VON (2005): Final Report and Guidelines: Mitigation of Hydro-Geological Risk in Alpine Catchments, CatchRisk. Work Package 2: Landslide hazard assessment (Rockfall modelling). Program Interreg IIIb Alpine Space. MAYER, K., PATULA, S., KRAPP, M., LEPPIG, B., THOM, P., POSCHINGER, A. VON (2010): Danger Map for the Bavarian Alps. Z. dt. Ges. Geowiss., 161/2, p. 119-128, 10 gs. Stuttgart, June 2010

MEISSL, G. (1998): Modellierung der Reichweite von Felsstrzen. In: Innsbrucker Geographische Studien, 28: 249 p., Innsbruck (Selbstverl. des Instituts fr Geographie der Universitt Innsbruck). ONOFRI, R. & CANDIAN, C. (1979): Indagine sui limiti di massima invasione dei blocchi franati durante il sisma del Friuli del 1976. Regione Autonoma Friuli-Venezia Giulia e Universit degli Studi di Trieste, 41 p., Trieste (Cluet Publisher). PERSONENKREIS GEOGEFAHREN (2008): Geogene Naturgefahren in Deutschland Empfehlungen der Staatlichen Geologischen Dienste (SGD) zur Erstellung von Gefahrenhinweiskarten; not published. SELBY, M.J. (1993): Hillslope Materials and Processes, Oxford University Press, Oxford. UMWELTBUNDESAMT [eds.] (2008): Klimaauswirkungen und Anpassung in Deutschland Phase 1: Erstellung regionaler Klimaszenarien fr Deutschland. http://www.umweltdaten.de/ publikationen/fpdf-l/3513.pdf WADGE, G., WISLOCKI, A.P. & PEARSON, E.J. (1993): Spatial analysis in GIS for natural hazard assessment. In: Goodchild, M.F., Parks B.O. & Steyaert, L.T. (Hrsg.) Environmental modelling with GIS: 332-338, New York, Oxford. WIECZOREK, F. G., MORRISSEY, M. M., IOVINE, G. & GODT, J. (1999): Rockfall Potential in the Yosemite Valley, California. In: U.S. Geological Survey Open-File Report 99-0578, http://pubs.usgs.gov/of/1999/ofr-990578/.

Anschrift der Verfasser / Authors addresses: Karl Mayer Bavarian Environment Agency (LfU) (Ofce Munich) Lazarettstrae 67 80636 Munich GERMANY Andreas von Poschinger Bavarian Environment Agency (LfU) (Ofce Munich) Lazarettstrae 67 80636 Munich GERMANY

Literatur / References: BUNDESAMT


FR RAUMENTWICKLUNG, BUNDESAMT FR WASSER UND GEOLOGIE, BUNDESAMT FR UMWELT, WALD UND LANDSCHAFT (BUWAL) [eds.] (2005): Empfehlungen Raumplanung und Naturgefahren. 50 p., Bern. EVANS, S. G. & HUNGR, O. (1993): The assessment of rock fall hazards at the base of talus slopes. Canadian Geotechnical Journal, 30 (4): 620-636, Ottawa (Nat. Res. Council of Canada). HEGG, C. & KIENHOLZ, H. (1995): Deterministic paths of gravity-driven slope processes: The Vector Tree Model. In: Carrara, A. & Guzzetti, F. (eds.): Geographical Information Systems in Assessing Natural Hazards, 79 92, Dordrecht.

Seite 129

Hazard assessment and mapping of mass-movements in the EU

Seite 130

DIDIER RICHARD

Standards and Methods of Hazard Assessment for Rapid Mass Movements in France Standards und Methoden der Gefhrdungsanalyse fr schnelle Massenbewegungen in Frankreich
Summary: Hazard assessment is required for different purposes and is carried out through expertise assessments at different levels, using various approaches. Hazard assessment and mapping methods are standardized at least for their use in the frame of land-use planning in what is called the plan for the prevention of natural hazards (plan de prvention des risques naturels prvisibles, PPR). This is one of the main instruments used by the French national authorities for preventing natural hazards while taking them into account in land use development. Within this procedure, a general methodological guidelines document and other documents specic to the different types of hazards specify the conditions and clarify the method and approach proposed to draw up the PPR. One of these documents is dedicated to mass movement hazards. In this procedure, the hazard map is an intermediate step in elaborating the risk map, i.e. the regulations stemming from the PPR (together with the associated regulations). Various types of information available and databases can be used for hazard assessment and hazard mapping, based on an inventory of phenomena and a back-analysis of current and past events. Hazard assessment must characterize a given hazard in terms of intensity and frequency of occurrence. For mass movements, specic approaches are proposed, given the specic characteristics of these phenomena. Zusammenfassung: Gefahrenbeurteilungen sind fr verschiedene Zwecke erforderlich und werden in Form von fachlichen Gutachten auf unterschiedlichen Ebenen anhand verschiedener Anstze vorgenommen. Gefhrdungsbeurteilung und Kartierungsmethoden sind zumindest fr die Verwen-

dung im Rahmen der Flchennutzungsplanung standardisiert: Der Plan fr die Verhinderung von Naturgefahren (plan de prvention des risques naturels prvisibles, PPR) ist eines der wichtigsten Mittel der franzsischen nationalen Behrden fr die Vermeidung natrlicher Gefahren und ndet in der Flchennutzungsplanung Bercksichtigung. Im Rahmen dieses Verfahrens beschreiben allgemeine methodologische Richtlinien und andere, fr die verschiedenen Arten von Gefahren spezische Dokumente die Bedingungen und geben Aufschluss ber die empfohlenen Methoden und Anstze zum Erstellen des PPR. Eines dieser Dokumente befasst sich mit den durch Massenbewegungen verursachten Gefahren. In diesem Verfahren ist der Gefahrenzonenplan ein Zwischenschritt in der Erstellung des Risikoplans, d.h., die Vorgaben stammen vom PPR (gemeinsam mit den zugehrigen Bestimmungen). Fr die Erstellung von Gefhrdungsanalysen und die Gefahrenzonenplanung (Gefahrenkartierung) stehen beruhend auf einem Bestand von Phnomenen und einer Analyse aktueller und vergangener Ereignisse verschiedene Arten von Informationen und Datenbanken zur Verfgung. Gefhrdungsanalysen mssen eine gegebene Gefahr in Bezug auf die Intensitt und Hugkeit des Auftretens beschreiben. Fr Massenbewegungen sind spezische Anstze empfohlen, welche die spezischen Merkmale dieser Erscheinungen bercksichtigen.
of territorial coherence at an inter-urban scale and Introduction Hazard assessment of rapid mass movements is required for different purposes than for other natural phenomena. Depending on the objectives, this must be carried out at different scales. Hazard assessment can also take different forms, but most often its nal outcome is a hazard map. Different types of expertise from various experts and approaches contribute to hazard assessment. Therefore, establishing standardized approaches, methods and tools is demanding. The eld of landuse planning, however, integrates standardized hazard assessment and mapping methods. Hazards mapping and land-use planning Natural hazards must be taken into account in landuse planning documents. These are mainly schemes local urban planning at the community scale. Typically, urban planning procedures and decisions, under the jurisdiction of national or local authorities, must integrate natural hazards. The plan for prevention of natural hazards (plan de prvention des risques naturels prvisibles - PPR) established by the law of February 2, 1995, is now one of the national authoritys main instruments for preventing natural hazards. The PPR is a specic procedure designed to take into account natural hazards in land-use development. The PPR is elaborated under the authority of the departments prefect, which approves it after formal consultation with municipalities and a public inquiry. The PPR involves the local and regional authorities concerned from the very rst steps of its preparation (Fig. 1). It can cover one or several types of hazards and one or several municipalities.

Seite 131

Hazard assessment and mapping of mass-movements in the EU

Seite 132

craft, commercial or industrial activity, for their completion, use or exploitation and requirements of any kind can be used, up to total prohibition. The PPR may also dene general preventive, protection and safety measures that must be
Fig. 1: PPR elaboration scheme (Source: V. Boudires; 2008) Abb. 1: Programm zur Ausarbeitung eines PPR (Quelle: V. Boudires; 2008)

are exposed to various phenomena stemming from the instability of slopes and cliffs (collapses, rock falls, landslides). Mass movements are demonstrations of the gravitational movement of ground masses destabilized under the inuence of natural solicitations (snow melting, abnormally heavy rainfall, an earthquake, etc.) or human activities (excavation, vibration, deforestation, exploitation of materials or groundwater, etc.). They vary greatly in form, resulting from the multiplicity of triggering mechanisms (erosion, dissolution, deformation and collapse under static or dynamic load), themselves related to the complexity of the geotechnical behaviour of the materials (geologic structure, geometry of the fracture networks, groundwater characteristics, etc.) According to the velocity of movement, two groups can be distinguished: Slow movements, for which the deformation is progressive and can be accompanied by collapse but in principle without sudden acceleration: Ground subsidence consecutive to changes in natural or articial subterranean cavities (quarries or mines); Compaction by shrinkage of clayey grounds and by consolidation of certain compressible grounds (muck, peat); Creep of plastic materials on low slopes; Landslides, i.e. a mass movement along a at, curved or complex discontinuity surface of cohesive grounds (marls and clays); Shrinkage or swelling of certain clayey materials depending on their moisture content. Rapid movements which can be split into two groups, according to the propagation mode of materials:

taken into account by communities as well as individuals. This option particularly concerns measures relating to the safety of persons and the organization of rescue operations as well as all general measures that are not specically related to a particular project. Finally, the PPR may take an interest in existing structures as well as new projects. However, for property construction that has been allowed in the past, only limited improvements whose cost is less than 10% of the market or

For areas exposed to greater hazards, the PPR is a document which informs the public on zones that expose populations and property to hazards. It regulates land use, taking into account natural hazards identied in this zone and goals of nonaggravation of risks. This regulation extends from authorising construction under certain conditions to prohibiting construction in cases where the foreseeable intensity of hazard or the nonaggravation of existing risks warrants such action. This guides development choices on less exposed land in order to reduce harm and damage to persons and property. The PPR is designed for urban planning communities and government and is incumbent on everybody: individuals, companies, authorities, especially when delivering building permits. It must therefore be annexed to the local urban planning plan when such a document exists. The basis for the regulation of projects in the perimeter of a PPR is to discontinue development in areas with the greatest hazard and, therefore, to prohibit land development and construction. This principle must be strictly

applied when the safety of persons is involved. In other cases, this principle remains particularly warranted by the cost of preventive measures to reduce the vulnerability of future constructions and the cost of compensation in cases of disaster, nanced by society. However, since the prevention objectives are then based on economic considerations, it is possible to discuss the limits of prohibitions and requirements with local actors, elected ofcials and economic and consumer representatives without departing from this principle. Adjustments can be accepted when the situation does not allow alternatives. For example in urban centres, where requirements to reduce the vulnerability of projects and preventive, protection and safety measures allowing the organization of emergency services will be set up. The PPR may operate in zones that are directly at risk, but also in other zones that are not in order to avoid aggravating existing risks or causing new ones. It regulates projects for new installations. It may prohibit or impose requirements on any type of construction, structure, development or any farming, forestry,

estimated value of the property can be required. As a complement to the PPR the central tool of the French national authorities natural hazards prevention action other procedures and tools are designed to provide preventive information that must be provided to inhabitants possibly exposed to hazards (information tools: DDRM, DCS, DICRIM, IAL, etc.) as well as measures relating to the safety of persons and the organization of rescue operations that must be taken into account by communities and private individuals (safety measures plan: PCS). These procedures are mandatory for the municipalities with an existing PPR. Danger studies are also mandatory for certain classes of hydraulic works (new regulations for dams and dikes). Adequate hazard assessment (and mapping) is of course also necessary for all these prevention tools. Rapid mass movements Approximately 7,000 French municipalities are threatened by mass movements, one-third of which can be highly dangerous for the population. Most of these towns, located in mountain regions,

Seite 133

Hazard assessment and mapping of mass-movements in the EU

Seite 134

The rst group includes: Subsidence resulting from the sudden collapse of the top of natural or articial subterranean cavities, without damping by the surface layers; Rock falls resulting from the mechanical alteration of fractured cliffs or rocky scarps (volumes ranging from 1 dm3 to 10 or 10 m );
4 5 3

Standards and methods In Frances administrative and institutional

sinking, collapse, rock falls, landslides, and associated mud ows, but it excludes debris ows in general. The general guide, published in August 1997, presents the PPR, species how it should be drawn up and tries to answer the numerous questions that may arise for their implementation. The other guidelines, such as the one dedicated to mass movements, clarify the method and approach proposed for the various types of risks. The general methodology establishes that the PPR is composed of: a presentation report explaining the analysis of the phenomena considered and the study of their impacts on people and existing or future property. This report explains the choices made for prevention, stating the principles the PPR is based on and commenting the regulations adopted. a regulatory map at a scale generally between 1:10,000 and 1:5,000, which delineates areas controlled by the PPR. These are risk-prone areas but also areas where development could aggravate the risks or produce new sources of risk. regulations applied to each of these areas.

The regulations dene the conditions required for carrying out projects, prevention, protection and safety measures that must be taken by individuals or communities, but also measures applicable to existing property and activities. The regulatory zoning of the PPR is based on risk assessment, which depends on the analysis of the natural phenomena that may occur and of their possible consequences in terms of land use and public safety. This analysis includes four preliminary stages: Determination of the risk basin and the study perimeter; Knowledge of the historic and active natural phenomena: inventory and description; Hazard qualication: characterization of natural phenomena which can arise within the study perimeter; Evaluation of the socioeconomic and human stakes subjected to these hazards. The elaboration of the PPR generally begins with the historical analysis of the main natural phenomena that have affected the studied territory. This analysis, possibly supplemented by expert advice on potential hazards, results

organization, certain activities and policies remain the jurisdiction of centralised authorities, such as the policy for natural risk prevention, overseen by the Ministry of the Environment. This is probably one of the most signicant differences compared with other Alpine countries. One consequence is the willingness to maintain a minimum homogeneity and coherence at the national level and in the way different types of natural hazards are treated. Within the framework of this common procedure, a general methodological guidelines document has been published, followed by others specic to the different types of hazards: oods, forest res, earthquakes, snow avalanches (to be approved), torrential oods (to be approved) One of these guideline documents is dedicated to geological hazards, including subsidence,

Some rock slides. The second group includes: Debris ows, which result from the transport of materials or viscous or uid mixtures in the bed of mountain streams; Mud ows, which generally result from the evolution of landslide fronts. Their propagation mode is intermediate between mass movement and uid or viscous transport.

Fig. 2: The PPR methodological guidelines collection Abb. 2: Die Sammlung methodologischer Richtlinien fr einen PPR Fig. 3: Positioning of the hazard map within the general procedure of PPR elaboration Abb. 3: Positionierung des Gefahrenzonenplans in der allgemeinen Ausarbeitungsphase eines PPR

Seite 135

Hazard assessment and mapping of mass-movements in the EU

Seite 136

in a hazard map that evaluates the scope of predictable phenomena. This map, including an analysis of the territory outcomes carried out in consultation with the various local partners, is the basis for reection during the elaboration of the PPR. Combining the levels of hazard and outcomes allows dening risk zones. Therefore, in this procedure the hazard map is an intermediate step necessary to elaborate the risk map, i.e. the real regulatory outcome of the PPR (together with the associated regulations). The study of phenomena by risk basin produces the hazard map, which is combined with the identication of elements at risk in drawing up the risk map. Data and information The rst step in elaborating hazard maps consists of collecting all available data and information that can be exploited for hazard assessment. Priority is given to the qualitative general studies and to the back-analysis of past events. The general studies are conducted based on existing data, the back-analysis of past or current events

and eld surveys. Priority must be given to these elements, as stipulated by article 3 of the decree of October 5th, 1995, which species that the elaboration of PPR takes into account the current state of knowledge. The main information sources are: Municipal archives (technical documents, deliberations, miscellaneous documents, petitions, general reports or accident reports, etc.); Parochial archives; Departmental sources (archive and quarry services, miscellaneous diagnoses, etc.); Engineering consulting rm documents (geotechnical and geological reports, civil engineering studies and reports, eld visit reports, etc.); General and research documents (scientic papers, geological guides, monographs, PhD theses, etc.); Field surveys and eye witness accounts; Existing databases and maps, aerial photographs. Historical and existing studies as well as eld investigations are collected for the study of the
Fig. 5: Geological maps and databases (www. brgm.fr) Abb. 5: Geologische Karten und Datenbanken (www. brgm.fr)

Study of phenomena by risk basin Historical and existing studies, eld investigation Informative map of natural phenomena Hazard map

Identication of elements at risk Available maps and data bases

Regulatory documents
Fig. 4: The rst step of hazard mapping Abb. 4: Der erste Schritt der Gefahrenzonenplanung Fig. 6: Example of a ZERMOS map Abb. 6: Beispiel eines ZERMOSPlans

Elements at risk appreciation

Risk Prevention Plan (PPR)

Necessary information and consultation

Risk management
Annexation as servitude in the PLU

Seite 137

Hazard assessment and mapping of mass-movements in the EU

Seite 138

Intensity level Low


Fig. 7: The BDMVT, French database of mass movements (www. bdmvt.net) Abb. 7: BDMVT franzsische Datenbank fr Massenbewegungen (www. bdmvt.net)

Coutermeasures importance level Can be nanced by an individual owner Can be nanced by a limited group of owners Concerns a spatial area larger than the individual ownership scale and/or very higth cost and/or technically difcult No possible technical countermeasure Only a few cases in France (Schilienne, la Clapire...)

Medium High

Major

Fig. 8: Example of relationships proposed between the importance of countermeasures and intensity level Abb. 8: Beispiel der empfohlenen Beziehungen zwischen der Bedeutung der Gegenmanahmen und der Intensittsstufe

implement. Different classes of intensity can be identied if these measures remain within the domain of an individual owner or a group of owners or if they require community intervention and investment (Fig. 8). Geological hazard qualication is based on

In most cases, the occurrence probability is not a true probability, but is simply a scale of relative susceptibility, relying on elements such as slope angle, lithology, fracturing of the rock mass, presence of water, etc. The hazard is graded by combining the time occurrence and the intensity, typically in a 2D table (Fig. 10). There is no general specication for this stage of the hazard evaluation, but presenting the key of the hazard evaluation is strongly recommended. and In the presence of substantial human socioeconomic danger, methods and

phenomena step. Maps and databases are available for this work: geological maps at a 1:50,000 scale, covering France (Fig. 5 - www.brgm.fr); a few Zermos maps (Fig. 6) of zones exposed to soil movement hazards, a combination of susceptibility levels and geomorphologic features, which are quite old and not exhaustive; a French database of mass movements (Fig. 7 - www.bdmvt.net); and an events database of the RTM services that will soon be on line. Hazard assessment Hazard evaluation includes three components: the intensity of mass movements, the time of occurrence and the spatial extension. Once translated into regulatory zoning, the information contained in this map will be used to manage and plan land development and construction works. Hazards are thus qualied in terms of intensity. Considering the variety of mass movements,

it is difcult to directly translate their physical characteristics in terms of intensity, except by dening as many hazards as movement types, which would make the hazard zoning document difcult to read. It is therefore necessary to refer to more global criteria so they can be compared and their use for regulatory zoning facilitated. Different methods are possible to assess a representative intensity level for all phenomena: As for earthquakes, intensity can be translated in terms of potential for damage, using parameters such as the volume of soil or rock involved, the depth of the failure surface, the nal displacement, the kinetic energy, etc. However, damage potential depends not only on the physical phenomenon, but also on the vulnerability of buildings, which introduces a bias. Intensity can be assessed according to the importance and the cost of protection measures that would be necessary to

qualitative criteria, such as the observed or expected damage or impacts or the cost range of possible countermeasures for the intensity evaluation. The frequency of events is estimated on the basis of the historical events identied on the site. The reference hazard is the most severe potential events considered by the expert as likely to occur in a 100-year period (or more frequently if human lives are concerned), or the most severe historical event identied on an equivalent site. The probabilistic approach based on a frequency analysis is possible only for some phenomena such as rock falls. This assumes that sufcient data are available, which is actually rare. As most mass movements are not repetitive processes, contrary to earthquakes or oods, it is necessary to consider a probability of occurrence of an event qualitatively over a given period (e.g. 50 or 100 years), without reference to numerical values. For instance, three levels or probabilities may be used: low, medium and high.

tools specifying the spatial extension of the phenomena, thus reducing uncertainty, can be used: run-out modelling for rock falls, geophysics surveys delineating underground mines, etc. In case of rock falls and related phenomena, hazard evaluation includes both the stability analysis of rock masses and run-out distance evaluation. Numerical tools are increasingly used to estimate the maximal run-out distance, but the reliability of the results is highly dependent on the experience of the engineering geologist. Generally, the topographic basis used is the IGN (National Geographic Institute) 1:25,000 map, enlarged to 1:10,000. In presence of

Seite 139

Hazard assessment and mapping of mass-movements in the EU

Seite 140

Conclusion Methods assessing hazards for rapid mass movements are still mostly empirical and rely on the experience of the engineering geologist. The PPR guidelines give a general framework and general principles for hazard assessment and mapping. Precise rules are not yet available at the national level. The geological analysis remains the basis of hazard evaluation, but numerical tools as GIS and computer simulation are also used. The main requirement is that the method used should be explained.

Acknowledgements Jean-Louis Durville, Conseil gnral de

l'environnement et du dveloppement durable. Alison Evans, Service de Restauration des Terrains en Montagne de Haute-Savoie. The person to contact for more information on this policy within the French Ministry of Sustainabledevelopment, is Franois Hdou (Francois. HEDOU@developpement-durable.gouv.fr).

Literatur / References: Anschrift des Verfassers / Authors address: Didier Richard


Fig. 9: Decision process for assessing the reference hazard Abb. 9: Entscheidungsprozess zur Bewertung der Bezugsgefhrdung
RISK PREVENTION FRENCH WEBPORTAL: www.prim.net RISK MAPPING: http://cartorisque.prim.net/ WEBSITE OF THE FRENCH MINISTRY IN CHARGE OF RISK PREVENTION POLICY: http://www.developpement-durable.gouv.fr/ FRENCH MASS MOVEMENTS DATABASE: http://www.bdmvt.net/ BRGM (bureau de recherches gologiques et minires) Website: http:// www.brgm.fr/ LCPC (1999) L'utilisation de la photo-interprtation dans l'tablissement des plans de prvention des risques lis aux mouvements de terrain. Collection Environnement, 128 p. LCPC (2000) Caractrisation et cartographie de l'ala d aux mouvements de terrain. Collection Environnement, 91 p. MINISTRE DE L'AMNAGEMENT DU TERRITOIRE (1999). Plans de prvention des risques naturels (PPR). Risques de mouvements de terrain. La Documentation franaise, 71 p.

Cemagref Unit de Recherche rosion torrentielle, neige et avalanches BP 76 F 38402 Saint-Martin-dHres Cedex or Schilienne (Isre), involving more than 10 million cubic metres of material, ad hoc methods of hazard assessment have been set up, including the monitoring of movement and various computer simulations. Tel. : +33 4 76 76 27 73 mail : didier.richard@cemagref.fr

substantial damage potential or if the precision of the study and the amount of available data allow it, it is possible to map the hazards on a 1:5,000-scale map. As far as very large mass movements are concerned, such as La Clapire (Alpes-Maritimes)

Probability of occurrence Intensity level


Determining factors identied on the site are diffuse, poorly determined.

Low

Many determining factors are identied on the site. Some factors unlisted can appear with time.

Medium

Some nonidentied determining factors on the site. The intensity of the factors is high.

High

Rock Falls < 1 dm3 Rock Falls < 100 m3 Collapses > 100 m3

Low

Very low to low hazard Very low to low hazard /

Very low to low hazard Medium hazard High hazard

/ High hazard High hazard

Medium High

Abb. 10: Beispiel fr die Erstellung einer bersichtstabelle ber Steinschlaggefahr (von CETE du sud-ouest) Fig 10: Example of hazard table determination for rock fall hazard (from CETE du sud-ouest)

Seite 141

Hazard assessment and mapping of mass-movements in the EU

Seite 142

Introduction With Law 19/2005, the Parliament of Catalonia approved the creation of the Geological Institute of Catalonia (IGC) assigned to the Ministry of Land Planning and Public Infrastructures (DPTOP) of the Catalonian Government. One of the functions of the IGC is to study and assess geological hazards, including avalanches, to propose measures to develop hazard forecast, prevention and mitigation and to give support to other agencies competent in land and urban planning, and in emergency management. Therefore, the IGC is in charge of making ofcial hazard maps for such a nality. These maps comply with the Catalan Urban Law PERE OLLER, MARTA GONZLEZ, JORDI PINYOL, JORDI MARTURI, PERE MARTNEZ (1/2005) which indicates that building is not allowed in those places where a risk exists. and The high density of urban development infrastructures in Catalonia for requires As information planning.

Geological Hazard Prevention Map of Catalonia 1:25,000 (MPRGC25M) The most important mapping plan is the Geological Hazard Prevention Map of Catalonia 1:25,000 (MPRGC25M). This project started in 2007. The MPRGC includes the representation of evidence, phenomena, susceptibility and natural hazards of geological processes. These are the processes generated by external geodynamics (such as slope, torrent, snow, coastal and ood dynamics) and internal (seismic) geodynamics. The information is displayed by different maps on each published sheet. The main map is presented on a scale of 1:25,000, and includes landslide, avalanche and ood hazard. The hazard level is qualitatively classied as high (red), medium (orange) and low (yellow). The methods used to analyze hazards basically consist of geomorphological, spatial and statistical analysis. Several complementary maps on a 1:100,000 scale show hazards caused individually by different phenomena in order to facilitate the

Geohazards Mapping in Catalonia Kartierung von geologischen Gefahren in Katalonien


Summary: This paper presents the different lines of work being undertaken by the Geological Institute of Catalonia (IGC) on geological hazard mapping. It describes the different map series, scales of representation, methodologies and its expected use. Keywords: hazard mapping, geohazards, Catalonia. Zusammenfassung: Diese Abhandlung bietet einen berblick ber die verschiedenen Aktivitten des Geologischen Instituts Katalonien (IGC) fr die Kartierung geologischer Gefahren. Sie beschreibt die unterschiedlichen Kartenserien, den Umfang der Darstellungen, die angewandte Methodik und den erwarteten Gebrauch der Karten. Schlsselwrter: Gefahrenkartierung, Geogefahren, Katalonien.

geo-thematic

a component of the Geoworks of the IGC, the strategic programme aimed at acquiring, elaborating, integrating and disseminating the basic geological, pedological and geothematic information concerning the whole of the territory in scales suitable for land and urban planning. Geo-hazard mapping is an essential part of this information. Despite some tests carried out with wide land recovery (Mountain Regions Hazard Map 1:50,000 [DGPAT, 1985], Risk Prevention Map of Catalonia 1:50,000 [ICC, 2003]), at present the work is done mainly on two scales: land planning scale (1:25,000), and urban planning scale (1:5,000 or more detailed). These scales imply different approaches and methods to obtain hazard parameters used for such a purpose. The maps are generated in the framework of a mapping plan or as the nal product of a specic hazard report. These different types of hazard mapping products are explained below.

Fig. 1: First published sheet, Vilamitjana (65-23), in 2010. Abb. 1: Das erste verffentlichte Blatt, Vilamitjana (65-23), 2010.

Seite 143

Hazard assessment and mapping of mass-movements in the EU

Seite 144

reading of the sheet and understanding of the mapped phenomena. Two additional maps for ooding and seismic hazards, represented on a 1:50,000 scale, are added to the sheet. The map is to provides government and individuals with an overview of the territory with respect to geological hazards, identifying areas where it is advisable to carry out detailed studies in case of action planning. At the same time, a database is being implemented. It will incorporate all the information obtained from these maps. In the future it will become the Geological Hazard Information System of Catalonia (SIRGC). The procedure followed in the main map consists of three steps: 1.Catalogue of phenomena and evidences 2.Susceptibility determination 3.Hazard determination The catalogue of phenomena and evidence is the base of the further susceptibility and hazard analysis. It consists of a geomorphologic approach and it comprises the following phases: 1. Bibliographic and cartographic search: the information available in archives and databases is collected. 2. Photointerpretation: carried out on vertical aerial photos of ights from different years (1957, 1977, 1985, 2003, etc.). The observation of the topography and the vegetation allows the identication of areas with signs of instability coming from the identication and characterization of events that occurred recently or in the past, and from activity indicators. 3. Field survey: checking and contrasting on the eld, the elements identied in the previous phases. Field analysis allows a better approach and understanding, and therefore identifying signs and phenomena are not observable through the photointerpretation.

4. Population inquiries: the goal of this stage is to complement the information obtained in the earlier stages, especially in aspects such as the intensity and frequency. It is done through a survey to witnesses who live and/or work in the study areas. In a second step, areas susceptible to be affected by the phenomena are identied from the starting zone to the maximum extent determinable at the scale of work. Their limits are drawn taking into account the catalogue of phenomena, geomorphological indicators of activity, and from the identication of favourable lithologies and morphologies of the terrain. This phase includes the completion of GIS and statistical analysis to support the determination of the starting and run-out zone. It can be extensively applied with satisfactory results with regard to the scale and purpose of the work. Finally, hazard is estimated on the basis of the analysis of the magnitude and frequency (or activity) of the observed or potential phenomena. Susceptibility areas are classied according to the hazard matrix represented in Fig. 2. Hazard zones are represented as follows: areas where no hazard was detected (white), zones with low hazard (yellow), medium hazard zones (orange), and areas with high hazard (red). In order to obtain an equivalent hazard for each phenomena, an effort was made to

equate the parameters that dene them. The same frequency/activity values were used for all phenomena, but magnitude values were adapted to each of them. Each hazard level contains some considerations for prevention (Fig. 3). These considerations inform about the need for further detailed studies and advise about the use of corrective measures.

An epigraph is assigned, to identify the hazard level and the phenomena that causes it, especially in overlapping areas (Fig. 5). This epigraph consists of two characters, the rst in capital letters, indicates the value of hazard (A for high hazard, M for medium hazard and B for low hazard), and the second, in lower-case, indicates the type of phenomena (e for large landslides, s for landslides, d for rockfalls, x for ows, a for avalanches and f for subsidence and collapses). The higher the overlapping is, the longer the epigraph will be.

Fig. 3: Prevention recommendations. Abb. 3: Empfohlene Prventivmanahmen.

Hazard

from

each

phenomena

is

analyzed individually. The main challenge of the map is to easily present the overlapping hazard of different phenomena. A methodology identifying that this overlap exists has been established with this objective in mind. It indicates what the maximum overlapped hazard is (Fig. 4), but in any case, without obtaining new hazard values.
Fig. 5: Example of multi-hazard representation. Abb. 5: Beispiel von Mehrfachrisiken.

Fig. 6: Main map 1:25000, which includes landslides, avalanches, sinking and ooding according to geomorphologic criteria. Fig. 2: Hazard matrix (based on Altimir et al, 2001). Abb. 2: Gefahrenmatrix (auf der Grundlage von Altimir et al, 2001). Fig. 4: Multi-hazard representation. Abb. 4: Darstellung von Mehrfachrisiken. Abb. 6: Hauptkarte 1:25000; sie veranschaulicht die Gefahren hinsichtlich Bergstrze, Lawinen, Absenkung und Hochwasser nach geomorphologischen Kriterien.

Seite 145

Hazard assessment and mapping of mass-movements in the EU

Seite 146

Complementary maps Complementary maps represent the hazard established for each individual phenomena at 1:100,000 scale. The purpose of these maps is to facilitate the interpretation of the main map. Depending on the type of phenomena identied in the main map, the number of complementary maps can vary from 1 to 6.

The nal map (Fig. 8) also represents the values of the basic seismic acceleration of the compulsory "Norma de Construccin Sismorresistente Espaola" (NCSE-02) for a placement in rock, and the intensity of the seismic emergency plan (SISMICAT).

Fig. 10: Flooding hazard map 1:100,000 based on hydraulic modeling. Abb. 10: Hochwasser-Gefahrenzonenkarte 1:100.000 auf der Grundlage hydraulischer Modellierung.

Fig. 12: First published Avalanche Paths Map, Val dAran Nord, in 1996. Abb. 12: Erste verffentlichte Lawinenzugkarte Val dAran Nord, 1996.

The termination of the MZA allows a rst global


Fig. 8: Seismic hazard map 1:100,000. Abb. 8: Seismische Gefahrenzonenkarte, 1:100.000. Fig. 7: Complementary map of surface landslide hazard. Abb. 7: Komplementrkarte ber Erdrutschrisiken. Fig. 11: Flooding hazard map symbology.

vision of the avalanche hazard distribution in this region. The area potentially affected by avalanches covers 1,257 km2. That is at 3.91% of the Catalan country, and considering the Pyrenean territory, it affects 36%. At present, all the avalanche information is stored in the avalanche database of Catalonia (BDAC). New events, coming from avalanche observation, are added to this database. The information is available via the Internet at: http://www.icc.cat/msbdac/. A second mapping plan, already nished, is the Avalanche Paths Map (MZA). It was begun in 1996 and nished in 2006. An extent of 5,092 km2 was surveyed. During this process 17,518 avalanche paths were mapped. This is a susceptibility map on a scale of 1:25,000, useful for land planning in the Pyrenean areas. The methodology is based on the French Carte de Localisation des Phnomnes dAvalanches (Pietri, 1993). On this map, the avalanche paths, mapped from terrain analysis (photointerpretation and eld work), are represented in orange, and the inventory information (witness surveys, historical documents, eld surveys and dendrochronology) is represented in violet. Hazard maps for urban planning At present, for all the municipalities that want to increase their building limits, the procedure is rst of all to make a preliminary hazard map on a 1:5,000 scale. This element is, in fact, just a map of yes or no, which states if a hazard exists or not. If the municipality decides not to develop in hazardous areas, the process nishes. In the case that the municipality wants to build in the hazardzone areas, more detailed studies have to be completed. These studies include complex data collection, usually via drilling specic boreholes, other geotechnical work, and advanced modelling.

Seismic hazard map This map was obtained from the map of seismic areas for a return period of 500 years, for a middle ground, and considering the effects of soil amplication. To take into account the amplication
Fig. 9: Seismic hazard map symbology. Abb. 9: Symbologie seismische Gefahrenzonenkarte.

Abb. 11: Symbologie Hochwasser-Gefahrenzonenkarte.

Avalanche Paths Map (MZA)

of the seismic motion due to soft ground, a geotechnical classication of lithologies from the Geological Map of Catalonia 1:25,000 into 4 types was carried out: R (hard rock), A (compact rocks), B (semi-compacted material) and C (non cohesive material). This classication is based on the speed of the S-wave through them (Fleta et al., 1998). The proposed amplications were assigned to each group of lithologies. For types R and A no additions of any degree of intensity were made, but for types B and C, there was an addition of 0.5 degrees of intensity.

Flooding hazard map The ooding hazard map at 1:50,000 scale shows the limits of the hydraulic modeling for periods of 50, 100 and 500 years provided by the Catalan Water Agency (ACA). A ooding map according to geomorphologic criteria was done in those streams were hydraulic modeling was not performed.

Seite 147

Hazard assessment and mapping of mass-movements in the EU

Seite 148

Anschrift der Verfasser / Authors addresses: Pere Oller, Marta Gonzlez, Jordi Pinyol, Jordi Marturi, Pere Martnez Institut Geolgic de Catalunya C/ Balmes 209/211 08006 Barcelona

Fig. 13: Interface of the avalanche data server Abb. 13: Benutzeroberche des Lawinendatenservers

Literatur / References:
PIETRI, C., 1993: Rnovation de la carte de localisation probable des avalanches. Revue de Gographie Alpine n1. P. 85-97. AGNCIA CATALANA DE LAIGUA (Departament de Medi Ambient i Habitatge). Directrius de planicaci i gesti de lespai uvial. Guia tcnica. 45 pp. ALTIMIR, J.; COPONS, R.; AMIG, J.; COROMINAS, J.; TORREBADELLA, J. AND VILAPLANA, J.M. (2001): Zonicaci del territori segons el grau de perillositat desllavissades al Principat dAndorra. Actes de les 1es Jornades del CRECIT. 13 I 14 de setembre de 2001. P. 119-132. FLETA, J., ESTRUCH, I. I GOULA, X. (1998). Geotechnical characterization for the regional assesment of seismic risk in Catalonia. Proceedings 4th Meeting of the Environmental and Engineering Geophysical Society, pg. 699-702. Barcelona, setembre 1998. NCSE-02 (2002). Norma de Construccin Sismorresistente Espaola. Parte General y de Edicacin, Comisin Permanente de Normas Sismorresistentes, Real Decreto 997/2002 del 27 de septiembre de 2002, Boletn Ocial del Estado n 244, viernes 11 de octubre de 2002. Ministerio de Fomento. P. 35898-35987.

The phenomena taken into account are landslides, rock falls, sinking and snow avalanches. In these maps, the hazard mapping is obtained from frequency/intensity analysis. Advanced modelling analysis is performed in order to obtain the most accurate results, and to support the observational data and expert criteria. Up to the present day, there is no standard methodology. The current challenge for the IGC is to prepare guidelines for such a goal in order to guarantee the standards of quality and homogeneity. There are preliminary studies of a hazard mapping plan 1:5,000 for snow avalanches. In this map terrain is classied into high hazard (red), medium hazard (blue) and low hazard (yellow). Urban planning implications regarding hazard have not been dened yet. An analysis of the MZA, supported by the statistical model, resulted in the identication of 24 urban areas to be mapped. The mapping methodology includes terrain analysis, avalanche inventory, nivometeorological analysis and numerical modelling to complete the information.

Seite 149

Hazard assessment and mapping of mass-movements in the EU

Seite 150

CLAIRE FOSTER, MATTHEW HARRISON, HELEN J. REEVES

Standards and Methods of Hazard Assessment for Mass Movements in Great Britain Standards und Methoden der Gefahrenbewertung von Massenbewegungen in Grobritannien
Summary: With less extreme topography and limited tectonic activity, Great Britain experiences a different landslide regime than countries in many other parts of the world e.g. Italy and France. Glacial modication of the landscape during the Pleistocene, followed by severe periglacial conditions have led to the presence of high numbers of ancient or relict landslides. Debris ows and rock falls common to higher relief areas of Europe occur but are less likely to interfere with development and population centres. Despite the often subdued nature of landslides in Great Britain, numerous high prole events in recent years have highlighted the continued need to produce useable, applied landslide information. The British Geological Survey has developed a national landslide susceptibility map which can be used to highlight potential areas of instability. It has been possible to create the national susceptibility map (GeoSure) because of the existence of vast data archives collected by the survey such as the National Landslide Database, National Geotechnical Database and digital geological maps. This susceptibility map has been extensively used by the insurance industry and has also been adopted for a number of externally funded projects targeting specic problems. Keywords British Geological Survey, Landslides, GeoSure, National Landslide Database

Zusammenfassung: Aufgrund einer weniger extremen Topographie und der beschrnkten tektonischen Aktivitt des Landes unterscheiden sich Auftreten und Verlauf von Erdrutschen in Grobritannien von denen in vielen anderen Lndern der Welt, z.B. Italien und Frankreich. Glaziale Vernderungen der Landschaft whrend des Pleistozns, denen schwierige periglaziale Bedingungen folgten, haben eine hohe Anzahl von vorzeitlichen oder relikten Bergstrzen verursacht. Die fr hhere Entlastungszonen in Europa typischen Muren und Felsstrze treten zwar auf, doch ihre Wahrscheinlichkeit, Entwicklungs- und Bevlkerungszentren zu beschdigen, ist gering. Trotz des hug geringen Ausmaes von Erdrutschen in Grobritannien heben zahlreiche bekannte Ereignisse der letzten Jahre nach wie vor die Notwendigkeit hervor, anwendbare Informationen ber Rutschungen zu erstellen. Vom British Geological Survey (BGS) wurde eine nationale Gefahrenhinweiskarte fr Rutschungen entwickelt, anhand derer potentielle Bereiche von Instabilitt aufgezeigt werden knnen. Die Erstellung der nationalen Gefahrenhinweiskarte (GeoSure) war auf der Grundlage umfangreicher Datenarchive mglich, die vom BGS zum Beispiel auf der Grundlage der National Landslide Database, der National Geotechnical Database und von digitalen geologischen Karten angelegt wurden. Diese Gefahrenhinweiskarte ndet beispielsweise in der Versicherungsbranche Anwendung und wurde fr eine Reihe extern nanzierter Projekte bernommen, die auf bestimmte Probleme abzielen. Schlsselwrter British Geological Survey, Rutschungen, GeoSure, National Landslide Database
planners. This view led to national assessments Background on landslide research and planning in Great Britain Prior to the 1966 Aberfan disaster, which led to the deaths of 144 people, landsliding was not widely considered to be particularly extensive or problematic in Great Britain (GB). In the years following the disaster, a limited amount of research into landslide distribution and mechanisms was undertaken but failed to lead to a structured regulatory framework for managing landslide risk. The Aberfan landslide and costly disruptions to infrastructure projects in the 1960/70s (Skempton & Weeks 1976 and Early & Skempton 1972) strengthened the view that the extent of ground instability was neither well understood nor managed by developers or of landslides being carried out in the 1980s and 1990s on which the current national policy is largely based. These assessments provided the basis for planning policies and guidance that, to some degree, continue to control development on or around unstable ground. However, limited resources since this initial push to understand the problem meant that these initiatives have failed to develop into an effective, integrated, national response to deal with landslides in GB. The current systems, which are neither centralized nor legally binding, comprise a system of planning regulations (Town and Country Panning Act 1990), guidance notes, operational regulations and building codes (Building Regulations, 2006). With the exception of the Building Regulations, none of these legal statutes specically mention

Seite 151

Hazard assessment and mapping of mass-movements in the EU

Seite 152

landslides. The majority of the legislation can be interpreted as placing responsibility with the developer, utility operator or landowner to ensure landslides are not an issue. The main source of regulatory information regarding slope instability issues is contained within Planning Policy Guidance Note 14 (PPG14) and its associated Annex (Anon 1990, 1994). The Annex sets out the procedure for landslide recognition and hazard assessment and emphasises the need to consider ground instability throughout the whole development process from land-use planning, through design to construction. These documents provide recommendations that slope instability be considered in any planning decision. If landsliding is a known issue, a developer must provide evidence that any development activity will not exacerbate landslide activity and that any building will be safe. However, PPG14 is not legally compulsory and only recommends that the local planning authorities should endeavour to make use of any relevant expertise when assessing whether a planning application may be affected by ground instability. The guidance notes do not specically refer to geological or geotechnical expertise but details of some information sources of are provided, including BGS data. Despite this, there is no legal compulsion for a planning authority to understand the extent or nature of landslide hazards within their area of concern and, thus, include them in planning decisions. Building regulations put further emphasis on the role of the developer to control the impact of instability requiring that The building shall be constructed so that ground movement caused by. land-slip or subsidence (other than subsidence arising from shrinkage), in so far as the risk can be reasonably foreseen, will not impair the stability of any part of the building. (Anon. 2004). The current PPG14 predates the era of

GIS and advises that citizens consult geological maps and the now defunct Department of the Environment Landslide Database. These sources of information have been superseded by the BGSs GeoSure and continually updated National Landslide Database. Despite the availability of these resources, national guidance has never been updated to take this into account. Despite the advances in landslide mapping and hazard mapping, there is still no legal compulsion to use or consider it within a planning application in GB. Development of landslide susceptibility maps and databases in GB BGS began to map geological hazards digitally in the mid 1990s. These early steps have paved the way for the development of much more detailed hazard maps that cover the whole of Great Britain and are complimented by detailed landslide mapping and an extensive National Landslide Database (NLD). The rst systematic assessment of hazards was triggered by the insurance industry after it identied a need to better understand geological hazards. Insurance losses caused by ground movements (including subsidence) between 1989 and 1991 reached around 12bn following a particularly dry period and, as a result, a digital geohazard information system (GHASP GeoHAzard Susceptibility Package) was developed by the BGS. This rst decision support system (DSS) gave a weighted averaged result for each of the 10000 postcode sectors in GB and came to be used by around 35% of the Industry (Culshaw & Kelk, 1994). Since the development of GHASP, improvements in GIS technology and the availability of digital topographical and geological mapping for 98% of GB have led to advances in the methods used to map geohazard potential.

The BGS has since developed a Geographical Information System (GIS)-based system (GeoSure) to assess the principal geological hazards across the country (Foster et al. 2008, Walsby 2007, 2008). One output is a GIS layer that provides ratings of the susceptibility of the country to landsliding on a rating scale of A (low or nil) to E (signicant), which has been simplied for Fig. 1. Importantly, a high susceptibility score does not necessarily mean that a landslide has happened in the past or will do so in the future, but where a landslide hazard is most likely to occur if the slope conditions are adversely altered by a change in one or more of the factors controlling slope instability (Fig. 1). GeoSure is produced at 1:50,000 scale and can be integrated to show the spatial distribution of landslide susceptibility in relation to buildings and infrastructure. According to the dataset, 350,000 households in the UK, representing 1% of all housing stock, are in areas considered to have a 'signicant' landslide susceptibility (Rated E). GeoSure works by modelling the causative factors of landsliding: lithology, slope angle and discontinuities being of prime importance. This has been made possible through the use of GIS due to its ability to spatially display and manipulate data (Soeters & Van Westen, 1996). The GeoSure methodology uses a heuristic approach to assess and classify the propensity of a geological formation to fail as well as to score the relevant causative factors. The BGS holds large amounts of information about the lithological nature of the rocks and soils within Great Britain. The National Geotechnical Physical Properties database contains information on the geographical distribution of physical properties (such as strength) of a wide range of rocks and soils present in GB. This information is vitally important in determining the propensity of a material to fail. The scores assigned to each lithology are based on material strength, permeability and known susceptibility to instability. Discontinuities

were assessed as an important causative factor as they reect the mass strength of a material, its susceptibility to failure and its ability to allow water to penetrate a rock mass. Scores were dened in line with those used in the British Standard 5930: Field Description of Rocks and Soils (British Standards Institute 1990) and by Bieniawski (1989). Analysis of known landslides showed that slope angle is one of the major controlling factors and this was derived from the NEXTMap digital terrain model of Britain at a 5m resolution. The scores for all the causative factors at each grid cell are combined in an algorithm to give an overall score based on the relative susceptibility to landsliding. The method is exible enough to allow alteration (nationally or locally) of the algorithm in the future and include other factors such as the presence and nature of supercial deposits.

Fig. 1: GeoSure layer showing the potential for landslide hazard Abb. 1: GeoSure-Schicht veranschaulicht das Potential von Rutschungsgefhrdungen.

Seite 153

Hazard assessment and mapping of mass-movements in the EU

Seite 154

Another important tool to both inform and assess landslide susceptibility in GB is the National Landslide Database (NLD). Landslide databases are commonplace in Europe but there is variability in their complexity and amount of further work carried out to further enhance or update the datasets. Assessing an areas susceptibility to landsliding requires knowledge of the distribution of existing failures and also an understanding of the causative factors and their spatial distribution. This type of information is only available from a detailed database of past events from which one can draw out relevant information which may inform the user of where landslides may occur in the future. The National Landslide Database is the most comprehensive source of information on recorded landslides in GB and currently holds records of over 15,000 landslide events (Fig. 2). Each of the 15,000+ landslide records can hold information on over 35 attributes including location, dimensions, landslide type, trigger mechanism, damage caused, slope angle, slope aspect, material, movement date, vegetation, hydrogeology, age, development and a full bibliographic reference. A fully digital workow has been developed at BGS to enable capture of landslide information. The rst stage of the process involves using digital aerial photograph interpretation software (SocetSet) to capture digital landslide polygons which can then be altered through eld checking using BGSSIGMA mobile technology (Jordan 2009; Jordan et al. 2005). BGSSIGMAmobile is the BGS digital eld data capture system running on rugged tablet PCs with integrated GPS units, and is used extensively for all geological mapping activities within the British Geological Survey (Jordan et al., 2008). When collecting landslide information, either for the NLD or for digital maps, internationally recognised standards have been followed where appropriate. The database
Fig. 2: Distribution of landslide database points from the National Landslide GIS database. OS topography Crown Copyright. All rights reserved. Abb. 2: Verteilung der Rutschungs-Datenbankpunkte von der National Landslide GIS Datenbank. OS Topographie Crown Copyright. Alle Rechte vorbehalten.

'style of activity.' Whilst the NLD follows the style of activity denitions, it has simplied the state of activity terms dened by Varnes (1978) into active, inactive and stabilised whilst also adding descriptions on the state of development (Advanced, degraded, incipient). Whilst activity state and style have been described in the WP/ WLI denitions (WP/WLI, 1993), age has been somewhat neglected. Data for modern landslides observed either at the time of the event or through comparison of aerial photographs and geological mapping, is included in the NLD. To record cause, the NLD has incorporated both triggering and preparatory factors, limited to those most likely to be identiable and relevant in GB. The denitions are based upon the WP/WLI (1990). Further adaptations of landslide susceptibility maps in Great Britain Following the creation BGS has of the Geosure within a

are most likely to occur in the future. An initial study determined ve main components which should be considered when determining the hazard potential of debris ows affecting the road network: 1. Availability of debris material 2. Hydrogeological conditions 3. Land use 4. Proximity of stream channels 5. Slope angle It was considered that information regarding each of these could be extracted from existing digital datasets. The resulting interpreted data were combined to produce a working model of debris ow hazard that could be validated by comparing with known events (Fig. 2). The A85 debris ow event in 2004 is shown alongside the modelled susceptibility layer, existing drainage channels are shown as particularly susceptible to failure through debris ows. Whilst the assessment of debris ows highlights areas where they may occur in the future, it does not attempt to model the run-out of such failures.

methodology,

worked

consortium including the Transport Research Laboratory (TRL) and the Scottish Executive to create a digital hazard layer specically for debris ows. This work was triggered in August 2004 following a period of intense rainfall which led to two debris ows trapping 57 motorists on the A85 trunk road in Scotland. As a consequence of this event and others during the same period, the Scottish Executive commissioned a study to assess the potential impact of further debris ows on the transport network of Scotland (Winter et al., 2005). BGS was involved in the provision of a GIS layer highlighting slopes susceptible to debris ows. Debris ows, one of the ve main types of landslides, have a specic set of preparatory criteria which differs from translational and rotational slides. This modied assessment sought to digitally capture this set of criteria and create a layer showing areas where debris ows

dictionaries internationally

have

been

produced terminology.

using For

recognised

Future Developments Currently, work is ongoing to validate the current methodology against statistical methods such as bivariate statistical analysis and probabilistic methods. The GeoSure method is based upon expert knowledge and a heuristic approach which is being tested against more statistic-based approaches to assess its validity. Naranjo et al., (1994) consider statistical methods to be the most appropriate method for mapping regional landslide susceptibility because the technique is objective, reproducible and easily updateable. Bivariate analysis for instance relies upon the availability of landslide occurrence and causal parameter maps, which are compared against

landslide type, the dictionary denitions follow the conventions set out by Varnes (1978), the EPOCH project (Flageollet, J.C., 1993) and the WP/WLI (1990). Age and activity of a landslide are important factors to record within a landslide inventory. Temporal landslide data is as important to understanding the geomorphic evolution of an area as the spatial distribution of slides. However, it is extremely difcult to date ancient landslide events with any degree of accuracy and, as such, the ages assigned to landslides only provide an arbitrary indication of age. The WP/WLI (1990) regrouped the Varnes (1978) denitions on age and activity under the following headings: 'state of activity,' 'distribution of activity' and

Seite 155

Hazard assessment and mapping of mass-movements in the EU

Seite 156

distributed data and causal factor information contained in the National Landslide Database of Great Britain, assesses the landslide susceptibility in Great Britain. It uses a heuristic approach to model the causative factors that cause these events. It assesses and classies the propensity of a geological formation to fail as well as to score the relevant causative factors (e.g. slope angle). By using these methodologies and datasets, a national assessment of the potential hazard to landsliding mass movement events in Great Britain can therefore be undertaken. Anschrift der Verfasser / Authors addresses: Dr. Helen J. Reeves Head of Science Land Use
Fig. 3a: Extract from the debris ow susceptibility layer along with b: the Glen Ogle debris ow of 2004. Abb. 3a: Ausschnitt der Gefahrenhinweiskarte fr Muren, gemeinsam mit b: dem Murgang in Glen Ogle, 2004.

BIENIAWSKI Z T (1989). Engineering Rock Mass Classications. Wiley Interscience, New York, 272 p BRITISH STANDARDS INSTITUTE. (1990). BS 5930. The Code of practice for site investigations. HMSO, London, 206 p EARLY, K.R. & SKEMPTON, A. 1972. Investigation of the landslide at Walton's Wood, Staffordshire. Quarterly Journal of Engineering Geology, 5, 19-41. FLAGEOLLET, J. C. (Ed) 1993. Temporal occurrence and forecasting of landslides in the. European Community. EPOCH (European Community Programme). FOSTER, C, GIBSON, AD & WILDMAN, G (2008). The new national landslide database and landslide hazards assessment of Great Britain. In: Sassa, K, Fukuoka, H & Nagai, H + 35 others (eds), Proceedings of the First World Landslide Forum, United Nations University, Tokyo. The International Promotion Committee of the International Programme on Landslides (IPL), Tokyo, Parallel Session Volume, 203-206. JORDAN, C. J., 2009. BGSSIGMAmobile; the BGS Digital Field Mapping System in Action. Digital Mapping Techniques 2009 Proceedings, May 1013, Morgantown, West Virginia, USA, Vol. U.S. Geological Survey Openle Report. JORDAN, C. J., BEE, E. J., SMITH, N. A., LAWLEY, R. S., FORD, J., HOWARD, A. S., AND LAXTON, J. L., 2005. The development of digital eld data collection systems to full the British Geological Survey mapping requirements. GIS and Spatial Analysis: Annual Conference of the International Association for Mathematical Geology, Toronto, Canada, York University, 886-891. NARANJO, J.L., VAN WESTEN, C.J. AND SOETERS, R. 1994. Evaluating the use of training areas in bivariate statistical landslide hazard analysis: a case study in Colombia. International Institute for Aerial Survey and Earth Sciences. 3 : 292300 SKEMPTON, A. & WEEKS, A. 1976 The Quaternary history of the Lower Greensand escarpment and Weald Clay vale near Sevenoaks, Kent. Philosophical Transactions of the Royal Society, A, 283, 493-526. SOETERS, R. & VAN WESTEN, C.J. 1996. Slope instability recognition, analysis and zonation. In: Transportation Research Board Special Report 247, National Research Council, National Academy Press, Washington, D. C., 129-177. SUZEN, M.L. AND DOYURAN, V. 2004. A comparison of the GIS based landslide susceptibility assessment methods: multivariate versus bivariate. Environmental Geology, 45, 665- 679. THE BUILDING AND APPROVED INSPECTORS REGULATIONS (Amendment). 2006. HMSO. TOWN AND COUNTRY PLANNING ACT. 1990. HMSO. VARNES D. J.: Slope movement types and processes. In: Schuster R. L. & Krizek R. J. Ed., Landslides, analysis and control. Transportation Research Board Sp. Rep. No. 176, Nat. Acad. oi Sciences, pp. 1133, 1978. WALSBY, JC (2007). Geohazard information to meet the needs of the British public and government policy. Quaternary International, 171/172: 179-185. WALSBY, JC (2008). GeoSure; a bridge between geology and decision-makers. In: Liverman, D.G.E., Pereira, CPG & Marker, B (eds.) Communicating environmental geoscience. Geological Society, London, Special Publications, 305: 81-87. WINTER, M. G., MACGREGOR, F & SHACKMAN, L (Eds) 2005. Scottish Road Network Landslides Study. The Scottish Executive. Edinburgh. WP/ WLI. 1993. A suggested method for describing the activity of a landslide. Bulletin of the International Association of Engineering Geology, No. 47, 53-57. WP/ WLI. (International Geotechnical Societies UNESCO Working Party on World Landslide Inventory) 1990. A suggested method for reporting a landslide. Bulletin of the International Association of Engineering Geology, No. 41, 5-12.

Planning & Development British Geological Survey, Kingsley Dunham Centre, in the future by numerical methods for smaller, regional studies. Further adaptations to the GeoSure methodology, similar to those used to assess debris ows, are planned for the future. Rock fall hazard could be another type of mass movement that is investigated using the heuristic GeoSure approach applying different causal factors and scoring algorithms. Conclusion In Great Britain, landsliding does not have a structured regulatory framework, but historical events, such as the Aberfan disaster and Scottish debris ow events (Winter et al, 2005), have highlighted the importance of understanding the distribution and mechanisms that cause landslide mass movement events in Great Britain. The BGS GeoSure methodology, using spatially Literatur / References:
ALEOTTI, P., AND CHOWDHURY, R. 1999. Landslide hazard assessment: Summary review and new perspectives. Bulletin Engineering Geology and Environment, Vol. 58, pp. 2144. ANON. (1990). Planning Policy Guidance 14: Development on Unstable Land. Department of the Environment, Welsh Ofce. Her Majesty's Stationery Ofce, London. ANON. (1994). Planning Policy Guidance 14 (Annex 1): Development on Unstable Land: Landslides and Planning. Department of the Environment, Welsh Ofce. Her Majesty's Stationery Ofce, London. Anon. (2004). The Building Regulations 2000 (Structure), Approved Document A, 2004 Edition. Ofce of the Deputy Prime Minister. Her Majesty's Stationery Ofce, London. CULSHAW, MG & KELK, B (1994). A national geo-hazard information system for the UK insurance industry - the development of a commercial product in a geological survey environment. In: Proceedings of the 1st European Congress on Regional Geological Cartography and Information Systems, Bologna, Italy. 4, Paper 111, 3p.

each other to create a weighted value for each parameter determined by calculating the landslide density (Aleotti and Chowdhury, 1999 and Szen and Doyuran, 2004). Results from an initial pilot study suggest that, in small areas, where detailed landslide mapping exists, bivariate (conditional probability) and probabilistic approaches are able to more accurately predict landslide susceptibility than GeoSure. However, this approach only works where landslides have been mapped. This technique cannot be used where no landslide mapping has been undertaken. Another issue with the conditional probability technique is that it relies on the assumption that all the parameters are mutually exclusive. The value of the heuristic approach is its ability to highlight areas where there are no known landslides but where there is existing knowledge on the underlying causative factors. The heuristic approach is able to produce national scale assessments which could be rened

Keyworth, Nottingham. United Kingdom, NG12 5GG. Direct Tel:- +44 (0)115 936 3381 Mobile:- +44 (0)7989301144 Fax:- +44 (0)115 936 3385 E-mail:- hjre@bgs.ac.uk

Seite 157

Hazard assessment and mapping of mass-movements in the EU

Seite 158

KARL MAYER, BERNHARD LOCHNER

International Comparison: Summary of the Expert Hearing in Bolzano on 17 March 2010 Internationaler Vergleich: Zusammenfassung des Expert Hearings in Bozen vom 17. Mrz 2010
Summary: The AdaptAlp work package 5 Expert Hearing on March 17th, 2010 in Bolzano was attended by 28 experts from eight countries. It was dedicated to the goals of action 5.1: The creation of a multilingual glossary on landslides and especially the elaboration of minimum requirements for hazard mapping. Beside a short presentation on the progress and the further approach of the multilingual glossary, the state of the art in hazard mapping for each involved region was presented by several people responsible. Based on these presentations, which build the basis for the further approach, short abstracts were composed for each region. These short descriptions can be seen inside the ofcial Hearings report published on the AdaptAlp Homepage (www.adaptalp.org). In a further step, based on these abstracts and the presentations, two tables were created. On the one hand, all used maps were grouped according to different types and on the other hand diverse characteristics of maps were summarized and compared at the country level. With these matrices, similarities and differences between the involved regions become visible and a least common denominator could be elaborated. These denominators should be discussed at the next meeting (December 2010) and, as a result, a compilation of minimum requirements to the creation of Danger, Hazard and Risk maps will be published.

Zusammenfassung: Das AdaptAlp Workpackage 5 Expert Hearing am 17. Mrz 2010 in Bozen wurde von 28 Experten aus acht Lndern besucht und widmete sich inhaltlich vollstndig den Zielen von Action 5.1: Der Aufbau eines mehrsprachigen Glossars zu Hangbewegungen und insbesondere die Erarbeitung von Mindestanforderungen zur Erstellung von Gefahrenkarten. Neben einer kurzen Vorstellung des Projektfortschrittes und der weiteren Vorgehensweise hinsichtlich der Erarbeitung eines mehrsprachigen Glossars wurde von Vertretern aus allen beteiligten Lndern der jeweilige State oft the Art bezglich Gefahrenkartierung vorgestellt. Ausgehend von diesen Prsentationen, welche die Grundlage fr das weitere Vorgehen bilden, wurden im Anschluss an das Treffen Kurzzusammenfassungen fr jede Region verfasst, welche innerhalb eines Gesamtberichtes auf der AdaptAlp Homepage (www.adaptalp.org) einzusehen sind. In einem weiteren Schritt wurden auf Basis dieser Beitrge zwei Tabellen erstellt, welche einerseits alle verwendeten Karten strukturiert nach verschiedenen Typen und andererseits unterschiedliche Charakteristiken von Karten zusammenfassen und auf Lnderebene vergleichen. Mithilfe dieser Matrizen werden Gemeinsamkeiten und Unterschiede zwischen den beteiligten Regionen sichtbar und ein kleinster gemeinsamer Nenner kann erarbeitet und in einem nchsten Meeting (Dezember 2010) xiert werden. Ergebnis dieses Vorgehens und des Projektteiles wird eine Zusammenstellung von Mindestanforderungen zur Erstellung von Gefahrenhinweiskarten und Gefahrenkarten sein.
processes, a large variety of maps and methods 1. Introduction In dealing with geological and hazards spatial today, are used in the different European countries to prevent natural disasters. Hazard Exactly this variety, which reaches Zone Plans (Gefahrenzonenplan), geotechnical (active) (passive) from simple danger mappings to legally binding should be shown inside this part of the AdaptAlp project. However main goal of work package 5 (WP 5) is not only the description of this variety, but a development of a least common denominator which includes the minimum requirements for the creation of Danger, Hazard and Risk maps. This article focuses on the AdaptAlp Expert Hearing from 17 March 2010 take place in Bolzano and which dedicates the contents of work package 5. In the following sections, the main goals of this meeting and the contributions from the involved experts were shown. In the nal chapter, rst basic approaches concerning a possible synthesis out of the big variety of hazard planning methods is pointed out.

measures come to implementation to minimize risk. Because of a time limitation of active measures (e.g. protective walls) and the decrease of space for permanent settlings, spatial planning gets more and more important. Due to avalanche catastrophes in the 1950s which were affecting large parts of the Alps, in 1954 in the Swiss municipal Gadmen, the rst Avalanche-ZonePlan was passed. This was the rst time a natural hazard was considered in spatial planning (cf. Glade a. Felgentreff 2008, p 160f). Nowadays, almost 60 years later, hazard mapping is a central part in risk management. Countless types of Danger, Hazard and Risk maps are produced for all kinds of risks. With regard to natural hazards, especially geological

Seite 159

Hazard assessment and mapping of mass-movements in the EU

Seite 160

2. Main goals of the Expert Hearing The topics of the expert hearing are all about the goals of the AdaptAlp Work package 5 Hazard Mapping: Hazard zones are designated areas threatened by natural risks such as avalanches, landslides or ooding. The formulation of these hazard zones is an important aspect of spatial planning. AdaptAlp will evaluate, harmonise and improve different methods of hazard zone planning applied in the Alpine area. Focus will be on a comparison of methods for mapping geological and water risks in the individual countries. A glossary will facilitate transdisciplinary and translingual cooperation as well as support the harmonisation of the various methods. In selected model regions, methods to adapt risk analysis to the impact of climate change will be tested. This should support the development of hazard zone planning towards a climate change adaptation strategy. The results will be summarized in a synthesis report (www.adaptalp.org). The ofcial description of WP 5 shows two main parts (goals), which are worked out in Action 5.1 under the leadership of the Bavarian Environment Agency (LfU) in collaboration with the alpS Centre for Natural Hazard and Risk Management in Innsbruck and with the inputs from the international experts of the project partners. The two main goals are the elaboration of a multilingual glossary to landslides and the development of minimum standards to create danger, hazard and risk maps. As announced in the introduction, the main focus of the hearing in Bolzano lies on the elaboration of basics for the denition of minimum standards for hazard mapping. Therefore the progress of the glossary was only

addressed inside a short presentation at the beginning of this meeting. The rest of this one-day session was dedicated to the contents of hazard mapping. Due to this and the fact that the glossary part is already described in detail within chapter 2.6 of this publication, this article only refers to the hazard mapping part. 3. Hazard mapping in the Alpine regions At the beginning of this chapter, it is important to clarify that, because of the scheduled timing of the project, at this time no nal results can be presented. Nevertheless, the theoretical approach and the already achieved marks can be shown. In general the course of action in getting a synthesis to hazard mapping is structured in three steps. First step is the evaluation of the state of the art in hazard mapping in each country involved. Exactly this point was the intention and the main goal of the hearing in Bolzano. Two main questions remained to be answered: What kinds of danger, hazard and risk maps are ofcially applied in each country? Which standards are these maps based on? To answer these questions, each participant gave a short overview of the ofcial used danger, hazard and risk maps and also information on the creation of such maps were given in short presentations. The second step will be the harmonisation of the different methods used in several countries. Therefore similarities should be worked out and the least common denominator in the methods of hazard mapping should be found. This second step is to be discussed in detail in the next workshop at the end of 2010. The nal part will be the creation of a report, which includes the results of this

harmonisation. Within the hearing in Bolzano, the plenum discussed the possible commitment of such a report for each country. However the title of the project contained the term minimum standards, which rather sounds like a legal term, the involved experts decided to switch to word standards with requirements. So this legal character is avoided and the nal report will include a part with minimum requirements to the creation of danger, hazard and risk maps. 4. Short summary from the expert-contributions in Bolzano In the following sections, the state of the art presentations from several experts in Bolzano are shown in short summaries for each country. 4.1 Germany In Germany, geogenic natural hazards, such as mass movements, karstication, large scale ooding, as well as building ground that is affected by subsidence and uplift, shall in future be recorded, assessed and spatially represented using a common minimum standard. An important component for developing danger maps is the construction and evaluation of landslide inventories (e.g. landslide or sinkhole inventories). The recorded data in the inventories should have a minimal nationwide standards and are divided into: Main data on the topic area mass movements and subrosion / karst with information about the spatial positioning, about determination of coordinates, etc. Commonly shared technical data of the subject area mass movements and subrosion / karst with information about the date of origin, about the land use and about damage, etc.

Specic technical data of the subject area mass movement and subrosion / karst Surface data concerning subsidence and uplift Regarding landslides, slide, fall, ow and subrosion processes are recorded in the inventories. Methods lasting from eld studies to computerized modelling are used for the creation of these danger maps. In Germany, danger maps serve as a rst estimation of possible natural hazards caused by certain geological conditions and should serve as a planning reference for possible investigations of individual objects where necessary. On the danger map, the areas in which natural hazards are possible are not delineated precisely and local conditions (e.g. prevention schemes, topographic peculiarities) are not taken into consideration in every case. Because of these reasons, it is recommended adding the following annotations for each subject area: The following map was created for a 1:25,000 scale and is not precise. It serves as a rst estimation of possible engineering geological hazards and cannot replace a geotechnical survey. Areas within the immediate vicinity of danger elds can also be affected. The intensity and probability of a possible event cannot be extracted from the map. 4.2 Austria At this time there is no regulatory framework or technical norm concerning mass movements in Austria. Only the course of actions concerning oods, avalanches and debris ows are regulated by law. This includes the generation of hazard zoning maps (Gefahrenzonenplan). These are generated by the Austrian Service for Torrent and Avalanche Control (Forsttechnischer Dienst fr Wildbach- und Lawinenverbauung, WLV).

Seite 161

Hazard assessment and mapping of mass-movements in the EU

Seite 162

As there are no legal instructions or standards in Austria about if or how to deal with the evaluation of mass movements, the federal states are all following a different course of action. The status of available data is very different in the individual states. In some of the federal states almost no data is available, others have a lot of data but not digitally available. And then there are states that can rely on a lot of digitally available data and are working on generating landslide susceptibility maps. 4.3 Italy (Piemonte, Emilia-Romagna, Province Bolzano) In Italy the national law (high level, n. 445/1908) and Royal Decree R.D. (n. 3267/1923) were the rst public regulations on land use planning. At the beginning of 70s the land use management was transferred to regions. The national Law n. 183/1989 introduced land use planning at a basin scale. The government sets the standards and general aims without xing a methodology to analyse and evaluate the dangers, hazards and risks related to natural phenomena. The same law designated the Autorit di Bacino (Basin Authority) whose main goal is to draw up the Basin Plan, a tool for planning actions and rules for conservation and protection of the territory. One of the available tools produced by ARPA Piemonte is the Italian Landslides Inventory (IFFI). It is a national program of landslides inventory, sponsored by national authorities and made locally by the regions. It is the rst try of an inventory based on common graphical legend and glossary. The Emilia-Romagna Landslide Inventory Map (LIM) reports over 70,000 landslides, while the historical data base contains about 6,600 landslide events. LIM may be considered as an

elementary form of a hazard map and, based on this, enforce rules and obligations addressing landslide hazard reduction: only existing hamlets and villages can extend on dormant landslides; on active ones, all new construction is forbidden. Otherwise, the use of a purely descriptive terminology (active, dormant), restricts the usability of this map, being often obsolete, and is therefore a frequent bone of contention. In the federal state law from 11 August 1997, the base for the approval of guidelines to the creation of hazard plans (Gefahrenzonenplne) for South Tyrol was laid. Also the role of municipalities was dened to carry out the planning within three years. Finally, the approval of plans and the role of coinvolved partners are also part of this law. The scale of this legal binding hazard plan (Gefahrenzonenplan) in South Tyrol tends to the working level of detail for the analyzed area. In settlements, a 1:5,000 scale and in other regions a 1:10,000 scale is used and landslides, hydrological hazards and avalanches are analyzed.

for developing these maps are outlined in the federal guideline where a three step procedure is proposed: 1) Firstly, an indispensable prerequisite for the landslide hazard identication is obtaining information about past slope failure events: the maps of phenomena and the registration of events (database). 2) Secondly, hazard assessment implies the determination of magnitude or intensity over time. Five classes of hazard are determined in Switzerland: high danger (red zone), moderate danger (blue zone), low danger (yellow zone), residual danger (yellow-white zone) and no danger (white zone). 3) Based on the hazard maps and risk analysis, three kinds of measures can be then taken (third step): planning measures, technical measures and organizational measures. 4.5 France

documents is dedicated to geological hazards, which includes subsidence, sinking, collapse, rock falls, landslides, and associated mud ows, but excludes debris ows. 4.6 England Up until 1966, the UK Government were not interested in Geohazards, they were more interested in nding oil and gas to help the UK economy develop and expand. After the Aberfan disaster (where 144 people, 116 of them children), the UK government were much more interested and funded a number of research projects to look at the UKs geohazards. An inventory is the rst step in building an understanding of the occurrence of geohazards. Currently BGS maintains two main shallow geohazard databases: the National Landslide and Karst Database (www.bgs.ac.uk). These inventories provide the basis for analysing the spatial distribution of the geohazard and their causal factors. From this understanding susceptibility can be assessed. In 2002, BGS developed a nationwide susceptibility assessment of deterministic geohazards such as landslides, skrink-swell, etc. called GeoSure (http://www.bgs. ac.uk/products/geosure/). 4.7 Spain (Catalonia) The Parliament of Catalonia approved, with Law 19/2005, the creation of the Geological Institute of Catalonia (IGC), assigned to the Ministry of Land Planning and Public Infrastructures (DPTOP) of the Catalonian Government. The most important mapping plan is the Geological Hazard Prevention Map of Catalonia 1:25,000 (MPRGC25M). As a component of the Geoworks of the IGC, the strategic program

4.4 Switzerland The plan for prevention of natural hazards (plan Switzerland is a hazard-prone country exposed to many mass movements, but also to oods and snow avalanches. Active and dormant landslides take some 6% of the national surface. Most of the landslides are very slow or slow reaching some millimetres to centimetres of displacement per year. Sudden slope movements with velocities up to 40 m/s are also observed (e.g. rock avalanches). The federal laws came into force in 1991 and are based on an integrated approach to protect people and property from natural hazards. The nontechnical, preventive measures are of particular importance: land-use planning, zoning, building codes. The reference documents in Switzerland are the natural hazard maps. The techniques de prvention des risques naturels prvisibles PPR) established by the law of 2 February 1995 is the central tool of the French State's action in preventing natural hazards. The elaboration of the PPR is conducted under the authority of the prefect of the department, which approves it after formal consultation of municipalities and a public inquiry. The PPR is achieved by involving local and regional concerned authorities from the beginning of its preparation. It can handle only one type of hazard or more and cover one or several municipalities. In the frame of this common procedure, a general methodological guidelines document has been published. One of these guideline

Seite 163

Hazard assessment and mapping of mass-movements in the EU

Seite 164

Comparison of different maps and their scales Austria Level basic Type of map Geomorphologic map Geotechnical map Engineering geological map Level of attention 1:5,000 to 1:2,000 and 1:25,000 to 1:50,000 1:10,0001:25,000 1:10,000-1:50,000 (M1), 1:2,0001:10,000 (M2), >1:50,000 1:5,000-1:2,000 or bigger (M3) GBA and Krnten large scale 1:5,000-1:50,000 1:5,000 (landslides) WLV Germany Bayern Switzerland CH variable scales 1:200,000 1:250,000 Municipal 1:25,000- 1:10,000 1:100,000 1:25,000 Slovenia Slovenia Arpa Piemonte 1:10,000 Italy South Tyrol 1:5,000 Emilia Romagna 1:10,000 France France Spain Catalonia 1:10,000 UK UK variable

Inventory map inventory

1:25,000 to 1:50,000

1:10,000

1:10,000

1:10,000

Multi-temporal inventory map 1:10,000-1:50,000 (M1), 1:2,0001:10,000 (M2), 1:5,000-1:2,000 or bigger (M3) 1:25,000 1:200,000 (K, regional), 1:50,000 (St., local) K, Bleiberg: 1:10,000 1:2,000-1:10,000 1:5,000-1:2,000 or more not smaller than 1:50,000, usually 1:2,000 to 1:5,000 1:10,000 1:25,000 1:10,000 1:5,000; 1:10,000 1:5,000; 1:10,000 1:5,000; 1:10,000 1:250,000 1:5,000; 1:10,000 1:5,000 1:10,0001:25,000 1:25,000 1:5,000 1:1,000 variable scales 1:10,0001:25,000 1:50,000 and bigger 1:10,0001:50,000 yes 1:25,000 (2000) 1:5,000 (2009) 1:25,000 1:50,000

Map of phenomena

1:50,000 and bigger

1:10,000

1:5,000 or 1:10,000

suscepti-bility

Map of area of activity Landslide susceptibility map, danger map (Gefahrenhinweiskarte) hazard index map Hazard map

1:10,000

1:25,000

1:10,000-1:50,000

1:250,000

1:10,000

hazard

Detailed Study (Detailstudie) Hazard zone map (Gefahrenzonenkarte) Hazard zone map of the development plan Map of potential damage

risk

Vulnerability map Risk zoning map, risk map

Fig. 1: Comparison of different maps and their scales Abb. 1: Vergleich unterschiedlicher Karten und deren Mastab

Seite 165

Comparison of information collected for different inventories


Characteristics Austria GBA Inventory x national scale regional scale study/ detailed scale geometry (width, length...) Basic information where when what why who reported when Landslide conditions activity ( number of events...) slope position approx. original slope positional accuracy site description depth to bedrock depth to basal failure plane slope aspect slope Geology in general Geology, specied geologic unit tectonic unit lithology stratigraphy bedding attitude weathering geotechnical properties (rock, debris) geotechnical parameters (shear,) x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x
Hazard assessment and mapping of mass-movements in the EU

Ger By x x x x x x x x x x x x x CH SLO EmRo AP ST F UK Catalan

CH

SLO

Italy

UK

ES

Seite 166

K x

rock mass structure joints joint spacing discontinuities structural contributions Land cover Land use Hydrogeology Relationship to rainfall Classication of mass movements (not specied) Classication type rate of movement material water content Causes Trigger Precursory signs (ssures,) Silent witnesses Rock fall: shadow angle Rock fall: (geometric) slope gradient Damage "Hazard" to infrastructure Remedial measures Costs of rem. Measures Costs of investigation Method used to gather info (eld survey, aerial photo-interpretation,) Degree of precision of information Certainty/ reliability of information Investigations, reports, documentation, references included Bibliography included Fig. 2: Comparison of characteristics and information collected for different inventories and maps Abb. 2: Vergleich von Charakteristiken und eingehende Informationen fr unterschiedliche Inventare und Karte

x x

x x

x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x Seite 167 x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x

Hazard assessment and mapping of mass-movements in the EU

Seite 168

aimed to acquiring, elaborating, integrating and disseminating the basic geological, pedological and geothematic information concerning the whole of the territory in the suitable scales for the land and urban planning. This project started in 2007. In the MPRGC, evidence, phenomena, susceptibility and natural hazards of geological processes are represented. These processes are generated by external geodynamics (such as slope, torrent, snow, coastal and ood dynamics) and internal (seismic) geodynamics. The information is displayed by different maps on each published sheet. The main map is presented on a scale of 1:25,000, and includes landslide, avalanche and ood hazard. Hazard level is qualitatively classied as high (red), medium (orange) and low (yellow). The methods used to analyze hazards basically consist of geomorphologic, spatial and statistical analysis. 4.8 Slovenia Legislation, planning and prevention measures are not satisfying in the eld of landslides in Slovenia and the primary activities are still focused on remediation instead on the prevention measures. The updated Act on Spatial planning from 2007, governing natural disasters also discusses problems with mass movements, but a common methodology and procedures to prevent geologyrelated natural disasters does not exist yet. At the moment for Slovenia, a landslide susceptibility map (scale 1:250,000) and a debris-ow susceptibility map (scale 1:250,000) is elaborated by the Geological Survey of Slovenia. In addition to this, a probabilistic model of slope mass movement susceptibility for the Bovec municipality in north-western Slovenia was developed based on the expert geohazard map at scale 1:25,000 and several other relevant inuence factors.

5. Conclusion As mentioned in the introduction of this article, the state of the art in hazard mapping in the involved countries isnt in balance. This fact was also conrmed inside the Expert Hearing in Bolzano. To solve this problem, in a rst step the big variety of maps applied in the several regions was summarized in one table (see Fig. 1). This chart builds the basis for further actions concerning the creation of minimum requirements. It is structured into different levels and the associated type of maps. The levels lasting from basic (e.g. geomorphologic maps) over inventories (e.g. inventory map), susceptibility (e.g. susceptibility map) and hazard (e.g. hazard map) to risk (e.g. risk map). Furthermore, a matrix (see Fig. 2) with specied characteristics and information collected for different maps was created out of the great wealth of information given at the hearing in Bolzano. In particular, this table should help to nd accordances between the different approaches. All the characteristics used in any involved country (e.g. inventory) form the basis for the denition of minimum requirements to hazard mapping. Finally, out of these two matrices a recommendation will be created and, based thereon, the nal minimum requirements should be xed in the next workshop on December 2010 in Munich. The nal report on the whole project will include a chapter with the decided minimum requirements to the creation of Danger, Hazard and Risk maps.

Anschrift der Verfasser / Authors addresses: Karl Mayer Bavarian Environment Agency (LfU) (Ofce Munich) Lazarettstrae 67 80636 Munich GERMANY Bernhard Lochner alpS Centre for Natural Hazard and Risk Management Grabenweg 3 6020 Innsbruck - AUSTRIAText

Literatur / References:
CRUDEN, D.M. & VARNES, D.J. (1996): Landslide types and processes. In A. Keith Turner & Robert L. Schuster (eds), Landslide investigation and mitigation: 36-75. Transportation Research Board, special report 247. Washington: National Academy Press. FELGENTREFF, C. & GLADE, T. (Hrsg.) (2008): Naturrisiken und Sozialkatastrophen. Spektrum Akademischer Verlag, Heidelberg, 454 S. KOMAC, M. (2005): Probabilistic model of slope mass movement susceptibility - a case study of Bovec municipality, Slovenia. Geologija, 48/2, 311-340. KOMAC, M. & RIBII, M. (2006): Landslide susceptibility map of Slovenia at scale 1:250.000. Geologija, 49/2, 295-309. KOMAC, M., KUMELJ, . & RIBII, M. (2009): Debris-ow susceptibility model of Slovenia at scale 1: 250,000. Slovenia. Geologija, 52/1, 87-104. MAYER, K. & POSCHINGER, A. von (2005): Final Report and Guidelines: Mitigation of Hydro-Geological Risk in Alpine Catchments, CatchRisk. Work Package 2: Landslide hazard assessment (Rockfall modelling). Program Interreg IIIb Alpine Space. MAYER, K., Patula, S., Krapp, M., Leppig, B., Thom, P., Poschinger, A. von (2010): Danger Map for the Bavarian Alps. Z. dt. Ges. Geowiss., 161/2, p. 119-128, 10 gs. Stuttgart, June 2010 RAETZO, H., LATELTIN, O., TRIPET, J.P., BOLLINGER, D. (2002): Hazard assessment in Switzerland codes of practice for mass movements. Bull. of Engineering Geology and the Environment 61(3): 263-268. RIBII, M., KOMAC, M., MIKO, M., FAJFAR, D., RAVNIK, D., GVOZDANOVI, T., KOMEL, P., MIKLAVI, L. & KOSMATIN FRAS, M. (2006): Novelacija in nadgradnja informacijskega sistema o zemeljskih plazovih in vkljuitev v bazo GIS_UJME : konno poroilo. Ljubljana: Fakulteta za gradbenitvo in geodezijo (in Slovene).

Seite 169

Seite 170

AdaptAlp
DI Maria Patek, MBA Bundesministerium fr Land- und Forstwirtschaft, Umwelt und Wasserwirtschaft Abteilung IV/5 Marxergasse 3 1030 Wien Tel.: 01/711 00 - 7334 Fax: 01/71100 - 7399 E-Mail: die.wildbach@lebensministerium.at