International Forum on Satellite Earth Observation for Geohazard Risk Management.
Perspectives concerning Satellite EO and geohazard risk management:
the way forward.
Draft community paper concerning landslide hazards.
Date of Issue
Status
18/04/2012
Draft 1.0
Coordinators of the International Forum:
Philippe Bally (ESA)
Francesco Gaetani (GEO Secretariat)
Lead Authors:
Moretti Sandro, UNIFI (University of Florence), Italy
Cigna Francesca, BGS (British Geological Survey), UK
Contributing Authors:
Bianchini Silvia, UNIFI (University of Florence), Italy
Del Ventisette Chiara, UNIFI (University of Florence), Italy
Guzzetti Fausto, IRPI (Institute for Geo-Hydrological Protection), Italy
Mondini Alessandro, IRPI (Institute for Geo-Hydrological Protection),Italy
Marsh Stuart, BGS (British Geological Survey), UK
Raetzo Hugo, FOEN (Federal Office for the Environment), Switzerland
Raspini Federico, UNIFI (University of Florence), Italy
Schneiderbauer Stefan, EURAC (European Academy), Italy
Zebisch Marc, EURAC, (European Academy), Italy
Gerald Bawden, (USGS), USA
David Norbury (EFG), Belgium
Acknowledgements:
This paper received editing input from Philippe Bally (ESA), Geraint Cooksley (Altamira) and
Andrew Eddy (Athena Global) and Marie-Josée Banwell (Altamira Information).
Page 2/55
Date 18/04/2012
Page 3/55
Date 18/04/2012
CONTENTS
1 SCOPE OF THIS COMMUNITY PAPER ................................................................. 5
2 LANDSLIDE RISK AND EXPOSURE ..................................................................... 6
3 USERS AND THEIR INFORMATION NEEDS WITH REGARDS TO GEOHAZARD
RISKS........................................................................................................................ 11
4 THE EUROPEAN CASE ........................................................................................16
5 THE GLOBAL PERSPECTIVE.............................................................................. 20
6 CURRENT STATE OF SATELLITE EO SERVICES & APPLICATIONS ................... 22
6.1
6.2
7
THE WAY FORWARD ......................................................................................... 36
7.1
7.2
7.3
8
Main EO capacities used or in development ......................................................................... 23
Emerging research .............................................................................................................. 33
Technology & services ......................................................................................................... 38
Science ............................................................................................................................... 40
Users and practitioners ........................................................................................................41
REFERENCES ..................................................................................................... 43
ANNEX1. ....................................................................................................................................... 51
APPENDIX 1. ................................................................................................................................ 55
Page 4/55
Date 18/04/2012
This document is a draft of a series of papers with the scope of reflecting the community
perspective on geohazards. It forms the basis for discussion to be held at the International Forum
on Satellite Earth Observation for Geohazard Risk Management organised by the European
Space Agency on May 21-23, 2012 (www.int-eo-geo-hazard-forum-esa.org). The authors invite
comments and further contributions from the community on this paper so they can be collated
and discussed at the conference. The document will be updated on the basis of this discussion.
Comments should be addressed to esa.conference.bureau@esa.int with the title of the draft paper.
1
Scope of this community paper
This paper presents the perspectives concerning how satellite Earth Observation (EO) can
contribute to geohazard & disaster risk reduction in landslide-prone and landslide-affected areas.
The paper is addressed to both the operational and the scientific users of the landslide community,
and considers the state-of-the-art concerning EO data based landslide research, applications and
services, starting from the situation in Europe and expanding to provide a global perspective.
The current status of the landslide applications based on EO data is tackled through some case
studies and goals achieved over the last decade in a range of activities identified by the contributors
to this document, focussing initially on the European context, and broadened to address global
landslide hazards. The paper examines how to consolidate landslide applications and services to
achieve the benefits expected from their users. Particular reference is made to the forthcoming
availability of large volumes of imagery from new satellite missions and the consequent need for
effective, standardized and widely-accepted methodologies, as well as national and international
capacities for the integration of EO data into every-day practices for landslide risk management;
the final objective is the provision of support for landslide prevention, preparedness and emergency
response, as well as post-emergency and recovery activities, and mitigation strategies.
This paper outlines a 5 to 10-year vision for the landslide community, based on the assessment of
state of the art research and the application of EO for landslide risk management. This document
builds on the 3rd International Geohazards workshop of the Group of Earth Observation (GEO),
held in November 2007 in Frascati, Italy, which recommended “to stimulate an international effort
to monitor and study selected reference sites by establishing open access to relevant datasets
Page 5/55
Date 18/04/2012
according to GEO principles to foster the collaboration between all various partners and endusers”.
Community papers are being elaborated for other types of geohazards for discussion at the
International Forum on Satellite Earth Observation for Geohazard Risk Management (May 21-23,
2012) in Santorini, Greece. Other EO geohazards addressed include seismic hazards, volcanoes,
coastal lowlands and flood defense and inactive mines. The scope and theme of these community
papers are described in Appendix 1.
2
Landslide risk and exposure
Landslides, as a major type of geological hazard, represent one of the natural events that occur
most frequently worldwide after hydro-meteorological events. The occurrence of landslides
depends on complex interactions among a large number of partially interrelated factors, such as
geologic setting, geomorphic features, soil properties, land cover characteristics, and hydrological
and human impacts.
Landslide predisposing or preparatory variables making the slopes susceptible to failure include
soil and rock geo-mechanical properties, slope gradient and aspect, elevation, land cover, lithology
and drainage patterns; triggering or dynamic factors are those initiating landslide movements, and
might be either natural or human-induced, or even any combination of both (Dai and Lee, 2002).
Natural triggers include intense or prolonged rainfall, earthquakes, volcanic eruptions, rapid
snowmelt and permafrost thawing, and slope undercutting by rivers or sea-waves. Other factors
capable of acting as triggers for landslide failures are human activities such as slope excavation and
loading, land use changes (e.g. deforestation), rapid reservoir drawdown, blasting vibrations, and
water leakage from utilities.
Page 6/55
Date 18/04/2012
Figure 1: Recent rainfall induced landslides. Left, rock fall / rock slide in the Tramuntana
range, Majorca, Spain. Right, soil slips / debris flows in NE Sicily, Italy.
Landslides represent a main hazard in mountainous and hilly regions as well as along steep
riverbanks and coastlines, and their impacts depend largely on the area and volume involved, the
motion velocity and intensity, number and distribution of the elements at risk, their vulnerability
and their exposure value.
The global impact of slope failures on the population, structures and infrastructures, the economy
and the environment, remains largely undetermined, but a few examples are given below:

On 14-16 December 1999, massive debris flows initiated by record breaking precipitation
events caused more than 19,000 fatalities along the northern cost of Venezuela, North of
Caracas (Larsen et al., 2001).

In December 2010, heavy rainfall caused hundreds of fatalities and left 1.5 million homeless
in Colombia and Venezuela.

In August 2009, typhoon Morakot brought record breaking cumulative rainfall over central
Taiwan. The precipitation triggered thousands of landslides and debris flows, and caused
massive sediment transport. Mass movements and flooding left 461 people dead and 192
others missing.

A single landslide buried the town of Xiaolin killing an estimated 400 people (Feng 2011,
Tsai et al., 2010).
Page 7/55
Date 18/04/2012
 The 12 May 2008 Great Sichuan 8.0 Earthquake, in Sichuan Province of China, triggered
more than 15,000 landslides, rock falls and debris flows that resulted in 20,000 deaths (Yin
et al., 2009), 29% of the 68,000 fatalities caused by the earthquake.
Data collected by the International Landslide Centre at Durham University, UK, indicate that in
2003
the
death
toll
from
landslides
exceeded
2000
people
globally
(http://www.landslidecentre.org/). This is most probably an underestimate. In Italy, a country for
which a detailed record of landslide and flood mortality exists, in the 52-year period between 1960
and 2011, 789 landslide events have caused 3417 deaths, 15 missing persons and at least 1940
injured people in 522 municipalities (Salvati et al., 2010). These figures show how pervasive is
landslide risk to the population in Italy.
To represent areas prone to landslide risk on a global or continental scale a few attempts have been
made to assess susceptibility, hazard and risk with degrees of accuracy that aren’t always well
established. As shown by the outcomes of recent studies such as those carried out at global scale by
Nadim et al. (2006) and Hong et al. (2007), the risk areas are prevalently concentrated in the
Philippines and Japan and in Central and South America along the Pacific Coast, and in southeastern Asia, with medium to very high degree of hazard (e.g. Figure 2 and Figure 3). Landslideaffected areas are also found in Asia in the Himalayas (India, Nepal), in the European Alps (Italy)
and Balkan regions (Albania, Greece), in the Middle East (Turkey, Georgia, Azerbaijan, Iran), in the
Rocky and Appalachian Mountains (USA and Canada), and in some regions of Africa (Ethiopia,
Kenya, Tanzania, Cameroon).
Page 8/55
Date 18/04/2012
Figure 2: Global landslide hazard mapping: the Global Landslide Hazard Distribution, GDLND. This is
derived from the landslide hotspot map at global scale published by Nadim et al. (2006) based on a
heuristic landslide hazard model considering slope, lithology, soil moisture, precipitation, temperature
and seismicity as input variables. The map was produced in collaboration with the Columbia University
Center for Hazards and Risk Research (CHRR), the Norwegian Geotechnical Institute (NGI), and the
Columbia University Center for International Earth Science and Information Network (CIESIN).
Figure 3: Global landslide susceptibility map derived from surface multi-geospatial data: susceptibility
index data from Hong et al. (2007) plotted with the 2003 and 2007 major landslide events
(Kirschbaum et al., 2009). Hong et al. (2007) used a heuristic approach based on TRMM (Tropical
Rainfall Monitoring Mission) rainfall measures acquired by NASA and JAXA, and a GIS-based
weighted linear combination of different landslide-controlling factors such as slope, soil type and
texture, elevation, land cover and drainage density.
Page 9/55
Date 18/04/2012
There are many factors that can make a region of the world prone to landslides; this includes
topography, soil type and climate; for example, areas with coarse and relatively bare soil types and
rainfall-affected areas are more susceptible to landslide processes. As a consequence, the hazard of
rainfall-induced landslides tends to be much greater in tropical mountainous areas like the
Philippines, Central and South America, and south-eastern Asia, with susceptibility indexes up to 5
(highest susceptibility level).
Other landslide-prone regions resulting from the landslide susceptibility map by Hong et al. (2007)
include the Pacific Rim, the Himalayas and South Asia, the Rocky and Appalachian Mountains, the
Alps, and parts of the Middle East and Africa. India, China, Nepal, Japan, the USA, and Peru
include wide landslide-prone areas as well, with susceptibility indexes between 3 (medium) and 5
as illustrated in Figure 3.
The combination of the landslide susceptibility map with the distribution and vulnerability of the
elements at risk facilitates the understanding of the expected losses due to landslide occurrences
provides an estimation of the number of people exposed to landslides. There are different landslide
susceptibilities that have been produced at a global scale. They do not provide sufficient temporal
perspective or information on magnitude of the expected events, so do not accurately describe the
hazard. They do not account for the distribution and vulnerability of all the elements at risk, hence
they cannot provide statements of risk. Moreover, there is no updated database of landslide
occurrences at a global scale.
Maps of landslide risk at global scale have been generated using the results of the joint World BankColumbia University project on global risk mortality and economic losses (‘Natural Disaster
Hotspots: a global risk analysis’; version 1.0, 2005) that is based on the GDLND map (Figure 2).
Landslide susceptibility has been combined with population data to provide the landslide risk
priority map (Figure 4) that represents the distribution of landslide risk areas for disaster
monitoring and emergency response at global scale.
Page 10/55
Date 18/04/2012
Figure 4: Distribution of areas at higher landslide risk at global scale, based on the combination of the
GDLND and population density. The higher landslide risk lies in the areas along the circum-pacific belt
on the Pacific Coast in central and South America and in Japan-Philippines regions, in Eastern Europe,
in the Middle East, the Himalayas Ranges and South-East Asia, and South-East and central Africa.
3
Users and their information needs with regards to geohazard risks
The EO perspectives on landslide hazard assessment and risk reduction rely on synergic linkages
between the different actors involved in the process of landslide risk management. It is
consequently fundamental to engage a large variety of users from both private and public sectors,
from industry and the scientific community, governmental and research departments, from local to
international levels, and to provide them with easy accessible, continuous, accurate and consistent
information. The citizens represent the ultimate users of landslide risk management activities and
services, as they are those who are affected by risk and can benefit from proper strategies of
landslide risk mitigation, or suffer the consequences of inappropriate policies and actions.
Both operational and scientific users of the landslide community benefit from EO satellite support,
but their needs and requirements depend on their role within the risk management process (Table 1
and Table 2). A main distinction can be made between the activities performed in real (and nearreal) time and those in deferred time. During the ‘real time’ (period measurable in hours, days or
months) the performed emergency activities include urgent, immediate actions such as event nowcasting, containment of effects, counter measures for risk mitigation and restoration of the living
conditions existing before the event itself. On the other hand, the actions of study, forecasting and
prediction aimed at guaranteeing permanent and regular safeguard of human lives and properties
Page 11/55
Date 18/04/2012
at long term scale are carried out during the ‘deferred time’ (period measurable in years or
decades).
Operational users from the landslide community include civil protection agencies, decision
makers and stakeholders. They are often in charge of emergencies related to the occurrence of
ground movements threatening populated areas and, for their role, are asked to manage the
enormous impacts of landslide hazards on society, during both the real and deferred times. While
the population is continuously increasing – especially in developing countries, landslide impacts
are growing, and the knowledge of location, extent, typology, intensity, style and state of activity of
landslide processes is consequently strictly required.
- Civil Protection Authorities: include all the structures and activities put in place by public
administration with the aims of safeguarding the integrity of life, goods, buildings, cultural
heritage and environment from any damage or risk of damage arising from natural
disasters, catastrophes and any other hazardous event. In joint collaboration with the
scientific community, the civil protection agencies coordinate and manage the forecast of
landslide risk scenarios, monitoring and early warning systems, prevention activities aimed
at minimizing damages, relief operations (rescuing people, ensuring early assistance to the
population affected by disasters and overcoming the emergency), as well as training
activities to ensure preparedness of citizens to emergencies. Civil protection emergency
management and support have highly demanding needs; resources (e.g. computing, data,
services, knowledge and expertise) need to be shared in a coordinated and effective way and,
in case of disaster response, privileging timely responses (simple and clear procedures),
frequently updated information, and rapid identification of affected areas. This user group
has crucial importance for successful landslide risk mitigation, as it represents the contact
point
with
local
authorities,
and
provides them
with
direct
suggestions
and
recommendations during landslide-induced emergencies.
-
Policy makers and planners: include a wide range of elected government officials at the
national, regional or local level, politicians, administrations, land use planners and all those
authorities taking part in the selection of the best actions to be performed among several
alternative scenarios. Decision makers are interested in simple long-term effective
information on geohazards, to support their role in hazard mitigation (e.g. through
stabilization and remediation works), risk management (e.g. implementation of land use
planning strategies, regulation and controls driven by clear and firm laws). Their
information needs include identification, mapping and classification of areas with present
or past ground instability, e.g. location, areal extent, volume of displaced material,
Page 12/55
Date 18/04/2012
kinematic behavior and evolution of the phenomenon in space and time is strongly
requested. During and post-emergencies, real-time information includes mainly monitoring
activities (continuous stream of information to remote control stations and alert systems),
residual risk mapping (identification of affected areas and residual risk zonation), analysis
of stability of surrounding areas (selection of safe areas where affected population can be
relocated).
-
Other end-users: include a wide range of end-users of all sectors (e.g. environment,
economic and transport), such as insurance, engineering and construction companies,
infrastructure operators and land owners. They should be considered during the land use
planning phase and decision making processes, in a truly participatory risk management
process.
-
Citizens: are the ultimate beneficiaries of the geohazard-related strategies, and need to be
informed on where, when and to which extent ground instabilities will take place, in the
short-term. Correct and thorough knowledge of a phenomenon is the first step in
understanding and dealing with it properly preventing possible dangers. Hence, one of the
most important duties of the scientific community and responsible authorities is to make
the population aware of proper behaviors to adopt if a landslide occurs, by adopting
awareness and preparedness campaigns (increase the ability to be prepared during
unpredictable events), and establishing simple rules on how to prevent or minimize the
damage induced by landslide phenomena.
Scientific users of the landslide community include universities, geoscience research departments,
environmental agencies, national geological surveys and, generally, those institutes dealing with
slope instability and working on the prediction, monitoring and supervision of the various types of
landslide processes. Their main goals are the collection of satellite EO data validation through on
site measurements and their integration into geotechnical, hydrogeological and deformation
models.
Geological surveys are involved in both education and capacity building activities and actions, as
well as in risk assessment. They deal with long-term monitoring of geohazards, collection and
analysis of data and information related to natural hazards on a daily basis, and represent primary
providers of information products supporting decision makers, local and regional/county
authorities, and population when landslides occur, thus straddling both scientific and operational
roles.
Page 13/55
Date 18/04/2012
The scientific community carries out prediction and prevention research activities for knowledge
development on landslides, and collaborates at both functional and operative levels with the
responsible authorities to develop monitoring, surveillance and alert systems for hydrogeological
risk, mainly in the deferred time and, partially, in near-real time. Its activities include technicalscientific training and assistance for the civil protection agencies and local authorities, in the
framework of simulated events, as well as the development of methodologies for the identification
of landslide triggers, forecasting models, assessment of hydrological thresholds and dangerousness
of landslide processes, definition of operative procedures and protocols for the identification of risk
scenarios in agreement with national and/or local authorities.
Citizens
ACTIONS
ACTIONS
NEEDS
NEEDS
 Easily accessible information
 Updated EO data
 Direct contact with EO segment
ACTIONS
 Prediction and prevention activities
 Protection of environmental resources
 Implementation of early warning systems
ACTIONS
 Emergency management
 Updating risk scenarios
 Relief operations
 Accuracy-based products
 Easily access to scientific information
 Sharing of knowledge
NEEDS
 Real time observation tools
 Clear procedures and methods
 Timely products
 Urban and land use planning
 Risk mitigation strategies
 Clear and firm laws
ACTIONS
 Emergency management
 Relocation
 Long-term information on geohazards
 Timely updated thematic maps
 Large area coverage, simple, effective,
standardized, reliable information
NEEDS




Monitoring and surveillance
Emergency support
Daily bulletins
Daily severity maps
Monitoring data
Rapid mapping (rush mode products)
Residual risk zonation
Detection of safe areas
 Ask for truly participatory risk management processes
 Awareness and preparedness
NEEDS
Other endusers
 Access to scientific information
 Collection of accurate raw EO data
 Feedback about delivered product
NEEDS
Policy makers
and planners




ACTIONS
Civil
Protection
Agencies
 Technical/scientific training and
assistance
 Research on prediction and prevention
 Analysis of past ground movements
 Delivering of EO-based services
NEEDS
Scientific
community
Near-real and Real time
NEEDS
Deferred time
 Simple and standardized advices of
proper behavior in case of events
Table 1: User competences and needs for landslides-related hazard in the deferred, near-real and real
times.
Scientific Users
Page 14/55
Date 18/04/2012
Operational Users
Deferred
time
 Mapping and long-term monitoring
 Typology and kinematics
 Modeling and prediction
 Vulnerability assessment and modelling
 Inventory (location, type, area)
 State and style of activity
 Magnitude (intensity, volume)
 Monitoring of areas at higher risk
 Forecasting
Near-real
and Real
time
 Mapping landslide events and their
consequences
 Statistics of landslide event inventories
 Definition of landslide triggers and related
thresholds
 Event vulnerability assessment and modelling
 Residual risk definition and mapping
 Location of safe areas for relocation of
elements at risk
 Post-event motion assessment
 Residual risk zonation
Table 2: Scientific and operational user needs for landslide hazards.
Taking into account the different objective, tasks and responsibilities of the operational and
scientific landslide community in the deferred, near-real and real times (Table 1 and Table 2), the
information needs of the landslide community can be summarized as follows:
-
Regularly updated landslide maps (susceptibility, hazard and risk maps) and landslide
inventories, including location, type, area, volume, intensity, state and style of activity of
observed phenomena; updated distribution of landslide-affected areas help understanding
ongoing and future instability.
-
Long-term monitoring of areas at higher risk, with regularity and consistency of
observation, to improve the understanding of landslide kinematics and facilitate the
assessment of their future evolution; site-specific information on the instability conditions
are needed to associate the identified motions with causative factors and triggers, and
analyse zones with different susceptibility to landsliding.
-
Post-event motion and damage assessment, mapping of affected areas and identification of
safe zones for relocation of elements at risk; residual hazard and risk zonation.
-
Landslide vulnerability assessment and modelling; forecasting and early warning.
It is clear from the list above that the information required to address these needs and support all
aspects of risk assessment and disaster mitigation should satisfy some requirements in terms of the
spatial and temporal scale of observation.
The spatial scale for landslide phenomena ranges between regional and local, i.e. varies from
studies of landslide mapping over very wide areas (up to a few thousands of square kilometres) to
analysis of single phenomena. Hence, the technologies supporting landslide studies should on one
hand guarantee to cover territories of huge extension, but on the other hand be able to provide
access to detailed information for the understanding of motions occurring over very small areas
Page 15/55
Date 18/04/2012
(e.g. a few square meters) and the characterization of ground deformation with high accuracy.
Temporal scales for landslide hazards are strongly controlled by the intensity of the observed
phenomena and may range between monthly frequencies of observation for extremely and very
slow processes (V < 16 mm/yr, and 16 mm/yr ≤ V < 1.6 m/yr, respectively; Cruden & Varnes 1996),
to daily or hourly observations for more rapid phenomena.
Spatial and temporal scales also vary from phase to phase of the landslide management cycle,
which deals with different needs in terms of frequency and resolution of required information.
More detailed levels of information are required during the emergency and post-emergency phases,
in terms of both spatial and temporal sampling of the observed phenomena. Up to centimetre
resolutions might be required in terms of spatial sampling, and temporal steps of required
acquisitions may be as high as a few minutes during the emergencies.
4
The European case
Landslides affect urban settlements, infrastructures and agricultural and environmentally valuable
land in many hilly and mountainous areas in Europe. Landslides occur in many different geological
and environmental settings across the European territory. Based on the GDLND (Figure 5), most of
the landslide processes in Europe occur in the Italian, Austrian and Swiss territories on the Alpine
belt, as well as in the Pyrenees with medium to high degree of hazard. Medium to very high-risk
areas are also present in Bulgaria, Romania, Serbia, Bosnia, Balkan states (Albania, Greece) and
Turkey. Based on the spatial extent and distribution of the areas at landslide hazard at the global
scale (GDLND), Europe includes more than 30% of these areas and represents a portion of the
global problem.
Page 16/55
Date 18/04/2012
Figure 5: Distribution of areas with higher landslide hazard in Europe derived from the
Global Landslide Hazard Distribution, GDLND (CHRR, NGI and CIESIN, 2005).
Large rock falls, rockslides, rock avalanches and debris flows dominate in the Alps and steep slopes
in other mountain ranges, while slides and flows abound in flysch belts of Slovakia, Czech Republic,
Poland, Italy, Spain, France and other countries. Slides of various types are frequent on cliffs and
steep slopes along the coastline of Southern and Eastern England, as well as along the Bulgarian
Northern Black Sea coastline. Shallow slides and mudflows are widespread in the peat slopes of
Ireland, and slides and lateral spreads affect gentle slopes in quick clays in Sweden and Norway.
Flows and slides also typically occur in clay-rich sediments and sedimentary sequences in Tertiary
basins as well as on riverbanks.
Intense and long-lasting rainfalls represent the most frequent triggers of landslides in continental
Europe. However, rapid snowmelt events and earthquakes are also responsible for many landslides
and a few large landslides, and human activities frequently contribute to many slope failures,
especially in built-up areas. The rapid melting of snow is often overlooked, but it is a primary
trigger of landslides in many areas. Over the last few decades, landslide risk has increased as a
result of population growth in areas at risk and urban expansion; furthermore, climate change and
variations in the precipitation trends in areas at risk will likely change the nature and frequency of
landslide events, in some cases with detrimental effects. Internationally, less developed countries
have seen increased incidence of landslides due to urban expansion and degradation of the natural
environment, particularly deforestation.
Page 17/55
Date 18/04/2012
Geological, morphological and other geo-environmental settings and conditions are greatly variable
in the European territory, and landslide mapping has not been copied homogeneously both across
and within the different EU member states. As a result, there is a lack of harmonisation of mapping
approaches and models, input data, susceptibility, hazard and risk representation levels and scales
in Europe (Hervás, 2007). On the European scale, currently no comprehensive landslide database
exists (EEA 2010). Although many European countries have taken the initiative to create a national
database, the combination of these databases into one continental database is difficult due to
accessibility restrictions and high variability in level of information and resolution (Van Den
Eeckhaut et al. 2011). There is also a significant underestimation of European landslide events
reported in global world databases which is probably due, among the other reasons, to the frequent
occurrence of small and isolated landslide events in Europe (EEA 2010), as opposed to catastrophic
events. For example, a comparison of the hotspot map of Nadim et al. (2006) and landslide hazard
maps of Kirschbaum et al. (2009) shows that both maps attribute high landslide risk or hazard to
the major mountain ranges in the world. Apart from Italy and some Balkan states, no other
European countries are located in the defined hotspot areas, even though many of them face
serious landslide problems (Van Den Eeckhaut et al. 2011).
In the context of the FP7 SafeLand project1 an analysis to identify landslide hazard and risk
hotspots in Europe was carried out and the susceptibility assessment of slide- and flow-type
landslides at European scale was performed employing logistic regression modeling. Moreover, a
landslide dataset was produced combining the extraction of landslide-induced geomorphologic
features from Google Earth imagery and the locations of landslide events reported in about 40
scientific publications all over Europe. In total, the inventory includes more than 1,300 point-wise
landslide locations, which mainly correspond to major active landslides in the Alps, Pyrenees,
Carpathians, and Apennines.
1
http://www.safeland-fp7.eu
Page 18/55
Date 18/04/2012
Figure 6: Landslide inventory at European scale produced in the framework of
the SafeLand project (Van Den Eeckhaut et al. 2011).
Page 19/55
Date 18/04/2012
5
The global perspective
Considerable effort has been put into the global aspects of landslide hazards over the past
decade, ranging from the setting of strategies for mitigation and observation to the sharing of
experiences from different continents. Following the International Strategy and then Decade for
Disaster Reduction in the 1990s and early 2000s, the strategic aspect was picked up first by
CEOS and then brought into sharper focus when they joined with the UN agencies and the
science programmes under the International Council for Science to form the IGOS Partnership.
This was dedicated to the setting of International Global Observing Strategies, one of which was
developed for geohazards that had landslides as one of its four main themes. The IGOS
Geohazards Theme Report, published by ESA in 2004, sets out a long-term (10-years plus)
strategy for the better observation and monitoring of landslides. One of the main strands was to
build a stronger global geohazard community and this has been taken forward for landslides by
the International Consortium on Landslides, principally through the World Landslide Forum.
The IGOS Partnership was an excellent forum for designing strategy but it lacked political
stakeholders and so implementation was on a best-endeavours basis amongst the partners. This
is one of the main reasons for the formal establishment in 2005 of the Group on Earth
Observations (GEO), an intergovernmental organisation with the sole objective of building the
Global Observing System of Systems (GEOSS). Once GEO was established, with almost 100
Member States, and the IGOS Partners had joined it as Participating Organisations, the IGOS
Themes were then integrated into its 10 year Work Plan during 2009. Thus GEO has been
responsible for implementing the strategy through its Disasters Societal Benefit Area (SBA).
This has a series of Tasks or Sub-Tasks focused on different elements of the disaster response
cycle, though not specifically on landslides but rather for multiple hazards.
Within GEO, the community building process has continued through the establishment of its
Geohazards Community of Practice (GHCP). This has set out a Roadmap for the Disasters SBA
based on the four recognised stages of the disaster response cycle; preparedness, early warning,
response and recovery. This helped to shape the new GEO Work Plan for 2012-15 and Tasks in
that plan are now taking forward the implementation of the Roadmap. This Roadmap and Work
Plan provide a potential framework for defining new activities on landslide hazards. There are
also activities related to the ESA and EC initiative on Global Monitoring for Environment and
Page 20/55
Date 18/04/2012
Security that could fit within this framework, including those in the GMES emergency response
core service and within the GMES Downstream FP7 project DORIS.
The Community was recently brought together again by the GHCP with assistance from the
European Science Foundation and the COST Office for a High-Level Conference on Extreme
Geohazards and there will be follow-up meetings during the EGU in April 2012. The scientific
focus of these events will then be complemented by the more applied focus of the International
Forum on Satellite Earth Observation for Geohazard Risk Management in May 2012 at the
Santorini Convention Centre in Greece. This will bring together many of these international
players with staff from some of the key, relevant European, GMES projects. The combination of
excellent science, a strong observing system, applied projects approaching their sustainable,
operational phase and major industrial players will be critical to plan a more consolidated
approach to dealing with landslide hazards globally.
Page 21/55
Date 18/04/2012
6
Current state of satellite EO services & applications
EO satellite technologies are very suitable to support both operational and scientific users in the
process of landslide identification, mapping, characterization and monitoring, thanks to the
capabilities of timely sensing wide areas at relatively low costs, detecting landslide-induced surface
features and land motions, and providing long historical records of acquisitions for many areas of
the world. High-resolution optical and multi-spectral sensors are used to assess fault rupture,
damage assessment, and identify secondary hazards such as triggered landslides. EO geohazard
optical imagery is often used to map and monitor regions that are at the greatest risk and is most
heavily used as a post response tool. Satellite radar is used on a case-by-case basis to further
characterize the risks associated with a given landslide. Increasingly, InSAR techniques are used to
monitor areas at risk on an on-going basis to identify areas at high risk and support mitigation
activities. The access to EO data and the capacity to transform them to pertinent information for
decision makers is critical in order to implement better land use practices and to be efficiently
prepared for crisis management (BRGM, 2007). The main EO capacities are able to address most of
the spatial and temporal observational requirements of the landslide community, by providing data
and related services with spatial and temporal scales of different ranges (from regional to local
scale, from medium to very high resolution, from monthly to daily acquisition frequencies), and
thus fitting with the observational requirements of the community.
The EO capacity is currently used mainly in the framework of the near-real and deferred times and
includes support for the creation and/or updating of landslide inventory maps at a regional scale,
and the characterization and long term monitoring of single unstable slopes at local or single
phenomenon scales. One of the major goals of EO technologies is the capability to provide precise
estimates of ground motions and indicators of landslide activity without generating the necessity to
install any targets on the ground or have any physical contact with the unstable slopes. Emerging
research of the scientific community includes more advanced capacities such as support to
landslide modelling and designing of early warning systems for near-real and real time
applications.
The description of EO capacities concerning landslides is primarily based on the European
experience in the framework of ESA, GMES and ECFP7 projects such as Terrafirma 2003-2012,
SAFER (Services And Applications For Emergency Response) 2009-2011, PanGeo 2011-2013,
DORIS 2010-2013 and SafeLand 2009-2012. The global perspective on landslide-induced hazards
at risk is approached through international initiatives such as those recently carried out by GEO.
Page 22/55
Date 18/04/2012
6.1
Main EO capacities used or in development
Among the available EO satellite technologies, Synthetic Aperture Radar Interferometry (InSAR)
and Persistent Scatterer Interferometry (PSI) have recently demonstrated their suitability for the
detection, monitoring and characterization of extremely to very slow moving landslides, and their
complementarity with on-site measurements, at both regional and local scales (e.g., Strozzi et al,
2005, 2006, 2011, Farina et al. 2006; Colesanti & Wasowski, 2006; Guzzetti et al. 2012).
Preparing landslide maps is important to document the
extent of landslide phenomena in a region, to investigate
the distribution, types, pattern, recurrence and statistics of
slope failures, to determine landslide susceptibility,
hazard, vulnerability and risk, and to study the evolution
of landscapes dominated by mass-wasting processes.
Conventional methods for the production of landslide
maps rely chiefly on the visual interpretation of
stereoscopic aerial photography, aided by field surveys, or
in some cases by field surveys complemented by
stereoscopic aerial photography. These methods are time
consuming and resource intensive (Brabb 1991; Galli et al.
2008; Guzzetti et al. 2000, 2012). New and emerging
techniques based on satellite, airborne, and terrestrial
remote sensing technologies, facilitate the production of
landslide maps, reducing the time and resources required
for their compilation and systematic update (Guzzetti et al.
2012). The several different techniques and methods can
be grouped in two main categories: (i) analysis of surface
morphology,
exploiting
very-high
resolution
digital
elevation models (DEMs), and (ii) interpretation and
Terrafirma
Started
in
2003,
the
project
(www.terrafirma.eu.com)
was
funded
under the GMES Service Element (GSE)
programme of the European Space Agency
in the framework of GMES. Terrafirma has
concentrated on federating public users and
delivering precise terrain deformation
mapping and value adding services to
support civil protection agencies and local
authorities in charge of risk assessment and
mitigation. Terrafirma is a pan-European
ground motion information service focused
on seismic risk, flood defence and costal
lowland subsidence, inactive mines and
hydrogeological risks. As part of the
hydrogeological risk theme, the project built
on the foundations of EO-landslide
methodologies set down by the previous
SLAM project and delivered 16 Landslide
services of Terrafirma concern mountainous
areas affected by slope instability,
approaching three types of service: 9
Landslide Inventory (LSI), 6 Landslide
Monitoring (LSM) and 1 Landslide
Modelling (LSMd) services delivered in
Italy, Switzerland, Spain and Greece (e.g.
Moretti et al. in press). The products are
based on advanced terrain deformation
measurements based on satellite InSAR and
the PSI techniques.
analysis of satellite images, including panchromatic, multispectral and synthetic aperture radar
(SAR) images. Jaboyedoff et al. (2010) and Guzzetti et al. (2012) reviewed the literature on
applications of very-high resolution DEMs obtained by airborne LiDAR surveys for landslide
investigations, and have shown that DEMs and derivative products (e.g., contour maps, shaded
relief images, maps of slope, curvature, measures of surface roughness) are used primarily for the
visual analysis of the topographic surface, and the semi-automatic recognition of morphometric
Page 23/55
Date 18/04/2012
landslide features (Mckean and Roering, 2003; Glenn et al. 2006; Sato et al. 2007, Booth et al.
2009, Passalacqua et al. 2010, Tarolli et al. 2010).
Techniques based on the interpretation of panchromatic, multispectral and synthetic aperture
radar (SAR) images include (Guzzetti et al. 2012): (i) visual (heuristic) interpretation of optical
images, including panchromatic, composite, false-colour, and pan sharpened (“fused”) images (e.g.,
Marcelino et al., 2009; Fiorucci et al., 2011), (ii) analysis of multispectral images, including image
classification methods and semi-automatic detection and mapping of landslides (e.g., Cheng et al.,
2004; Metternicht et al., 2005; Rosin
and Hervás, 2005; Barlow et al., 2006;
Lee and Lee, 2006; Weirich and Blesius,
2007; Martha et al., 2010; Tsai et al.,
2010; Mondini et al., 2011a, 2011b), and
(iii)
analysis
of
SAR
images
(e.g.,
Czuchlewski et al., 2003; Singhroy and
Molch, 2004; Farina et al., 2006;
Lauknes et al., 2010).
Multispectral data of variable spatial and
spectral
resolution
(e.g.,
Quickbird,
IKONOS, SPOT-5, Geoeye, Resourcesat1,
Landsat)
have
been
extensively
exploited for mapping, monitoring and
forecasting
landslides.
Stereoscopic
visual interpretation of pan-sharpened
images (e.g., Nichol et al., 2006, Kouli et
al., 2010) and automatic pixel- and
object-oriented classification methods
(e.g, Martha et al., 2010; Mondini et al.,
2011) have shown great potential for
landslide inventory mapping (Guzzetti et
al., in press).
Change detection methods based on
temporal variations of landscape spectral
properties
Page 24/55
Date 18/04/2012
of
bi-temporal
image
PREVIEW
In the framework of EC FP7, PREVIEW was an Integrated Project
that provided new techniques to enhance information services for
risk management and to better protect European citizens against
environmental risks reducing their consequences.
Within the landslides thematic domain of PREVIEW project, the
Landslides Platform aimed at supporting the risk management
phases (prevention, forecasting, alert and crisis) at a regional and
global scale, through the integration of satellite EO data (i.e. ERS,
ENVISAT, Quickbird) and in situ data. The Landslides Platform
aimed to create an operational early warning system based on the
coupling of pre-existing thematic information (hazard and
geomorphologic maps, monitoring data) with “real time” data from
various sources and technologies (e.g. meteorological data).The
major objectives of the proposed system consisted in the integration
of several information sources and existent pre-operational tools
for estimating landslide risk and performing hazard zonation with a
particular focus on remote sensing techniques and in-situ
monitoring, and in the prediction and early warning of shallow
rapid slope movements through integrated new technologies of
meteo-hydrological modelling, radar-meteo techniques, geophysical
methods and GIS tools.
The specific services and related products provided by the Platform
are two types: Monitoring of slow-moving landslides (Service 1) and
Prediction of shallow rapid slope movements (Service 2). Service 1
concerned the monitoring of deep-seated slow-moving landslides
over large areas, by application of interferometric techniques; it
provided pre-crisis information to end-users by integrating several
information sources and existent pre-operational tools, including
interferometric data. This Service was based on the integration of
Earth Observation data processing through PSP-DIFSAR and in
situ measurements (i.e. Electrical Resistivity Tomography ERT,
displacement maps from Ground-Based radar) to monitor terrain
motions for the mitigation of landslide hazards; the final product of
slow-moving landslides monitoring service are updated landslides
inventory maps in the sample areas: Bormio and Valufurva in
Lombardia region (Italy), Vagnharad and Sundsvall (Sweden).
Service 2 dealt with the use of hydro-meteo modelling, radar
remote sensing and geophysical methods for forecasting and early
warning for local rapid slope movements, in some sample sites Italy
(i.e. Armea basin, in Liguria Region and Island of Ischia in
Campania Region).
acquisitions (pre- and post- landslide event) are particularly effective for updating landslide
extensions (e.g., Fiorucci et al., 2011). Methods based on correlation of high-quality optical images
showed good performances to quantify ground displacements and monitoring landslide activity and
represent useful tools for understanding slope failure mechanisms. (e.g., Delacourt et al., 2007;
Leprince et a., 2008). Furthermore, imaging spectroscopy techniques are essential for retrieving
hydrological and geomorphological diagnostic features, such as soil properties, land use, rainfall
fields, that are used as inputs in many landslide predictive models (e.g., van Westen et al., 2008).
Benefitting from the experience gained in Terrafirma and PREVIEW, the landslide thematic
platform of the EC FP7 SAFER (Services and Applications For Emergency Response) project2
delivered 10 similar services combining ancillary and thematic maps with EO satellite VHR, HR
and MR radar and optical data, for the detection, mapping, monitoring and/or forecasting of
landslide processes, at national to local scale, and delivering four typologies of EO based services in
Italy, Austria, Slovakia and Taiwan: 4 Landslide Inventory Mapping (LIM), 3 Landslide Monitoring
(LM), 2 Rapid Landslide Mapping (RLM) and 1 Real-time Shallow Landslide Forecasting (RSLF).
A step forward is undertaken by the EC FP7 DORIS project3 2010-2013, a downstream service for
the detection, mapping, monitoring and forecasting of ground deformations in seven test areas with
different physiographic and environmental settings, in Hungary, Italy, Poland, Spain, and
Switzerland. The services extensively exploit EO satellite C-, L- and X-band SAR archives to provide
unprecedented, very long time-series of ground deformations for landslide and subsidence areas.
Further, the time series of ground deformation are combined with ground-based InSAR data, GPS
measurements, and geophysical probing, for an improved monitoring and forecasting of ground
deformations, improving the ability of environmental and civil protection authorities to manage the
risks posed by ground deformations.
2
3
http://safer.emergencyresponse.eu/
http://www.doris-project.eu/
Page 25/55
Date 18/04/2012
Figure 7. Satellite EO based landslide services and applications at European scale, overlapped
onto the landslide hazard map of the GDLND (CHRR, NGI and CIESIN, 2005).
Satellite EO based applications are already mature in some countries such as Italy, Switzerland and
Spain, as demonstrated by many national initiatives carried out in the last years. A detailed
description of these activities is provided in Annex 1.
The main objectives and achievements of the above mentioned applications for the creation or
updating of landslide maps at regional scale, and the long term monitoring of unstable slopes at
local scale, are summarized below:
a)
Mapping and inventory: EO based landslide mapping and inventory applications and
services provide information on the spatial distribution of mass movements and generally operate
at regional scale. They integrate satellite-based ground deformation measurements into preexisting landslide inventories produced with field surveys conventional geomorphologic tools,
stereoscopic photo-interpretation of multi-temporal aerial and/or satellite optical imagery,
thematic, geological and topographic data, and. EO satellite data offer a cost-effective means to
identify indicators of slope instability, in the form of terrain features and landforms identified
through interpretation of optical imagery, as well as ground displacement estimates provided by
InSAR and PSI technologies.
The final goal of these applications is the creation or the improvement of landslide inventory maps,
through the delivery of qualitative (e.g. state of activity) and quantitative (e.g. intensity)
information of each mapped phenomenon and the detection and mapping of those phenomena not
previously identified through conventional investigations. Landslide services and applications like
Page 26/55
Date 18/04/2012
those of Terrafirma LSI, SAFER LIM and RLM, SLAM, PREVIEW, SAR.net, SAR.net2 and DORIS
have shown how the exploitation of EO data can reply to most of the users’ needs for landslide
identification and mapping, through the rapid detection of unstable areas and the identification of
their spatial extension and temporal evolution to support the emergency management process,
especially in the deferred time.
Example of successful applications are those carried out in Italy in the Calabria and Abruzzo
Regions (Bianchini et al. 2012, in press), South Tyrol, Reno valley and Biferno basin (Righini et al
2012), or those in Switzerland in Cantons of Valais, Graubuenden, Ticino, Vaud, Bern and Uri
(Raetzo et al., unpublished), as well as those in Lower Austria, in the Xiaolin village in Taiwan, in
Spain in the Central Pyrenees and the Gulf of Corinth in Greece. Further examples are those from
the SUDOE-DO-SMS project for Guéthary in France, and Bassin de Tremp and Valle de Tena in
Spain.
Within the Service for Landslide Monitoring (SLAM) project, as far as landslide inventories are
concerned, the terrain motion survey mappedr the whole hydrographic basin, or for a significant
portion of it. In Italy, the service developed for the whole extent of Arno river basin (about 9,000
sq. km), and on an area of about 1,200 sq. km inside the Liri-Garigliano-Volturno (LGV) rivers
basin (Italy).
In Switzerland, the service covered the geographic areas of Oberwallis and Grindelwald (Figure 8
modified from: Righini G, Pancioli V, Casagli N (2011) Updating landslide inventory maps using
Castemauro
Castemauro
Persistent Scatterer Interferometry).4
Figure 8: Example of mapping and inventory applications using EO satellite data: case of
Biferno basin (Italy (mod. from: Righini et al., 2011).
4
Int J Remote Sens. doi: 10.1080/01431161.2011.605087
Page 27/55
Date 18/04/2012
b)
Monitoring and characterization: EO satellite based landslide monitoring applications
analyze the temporal evolution of landslide-induced ground motions by exploiting ground motion
information provided by InSAR and PSI techniques. These data can support the geological and
kinematic interpretation of the slope instability affecting the observed areas, especially in built-up
and densely urbanized slopes, where landslide indicators are difficult to recognize due to the
presence of the urban fabric. Local-scale long-term monitoring of displacements induced by
specific slope movements, using EO satellite data integrated and compared with the available
conventional ground-based instruments networks (e.g. topographic levelling, inclinometers,
extensometers, GPS), allows the analysis of the temporal variability of landslide motions and
kinematics.
Besides the use of PSI technologies, conventional InSAR allows analysis not only of motion
velocities exceeding the limitation of the PSI approaches (i.e. few tens of cm/yr), but also
deformation trends significantly differing from the deformation model (e.g. linear) used during the
multi-temporal PSI processing (e.g. non-linear and/or accelerated motion). A supplementary
advantage of InSAR is the spatial coverage and the ability to detect the landslide limits with lower
costs than with PSI. But InSAR analyses are very demanding and experience is needed to face 3Dproblems, atmosphere deformations or phase unwrapping.
Landslide services and applications like those of Terrafirma LSM, SAFER LM, SLAM, PREVIEW,
DORIS and MORFEO have shown that the exploitation of EO data can answer most of the users’
needs for landslide monitoring thanks to the capability of resolving the temporal variability of
ground deformation and the possibility of reconstructing the history of displacement of landslideaffected areas, to recognize any precursors to landslide failures or identify any variability of the
motion behaviour due to triggering factors such as prolonged or intense rainfalls, thus supporting
the risk management process during both the near-real and the deferred times.
Example of successful applications are those carried out in Italy for the landslides of Ancona,
Frazzanò, Gaggio Montano, Gorgoglione, Santo Stefano d’Aveto and Mont-de-la-Saxe, or in
Switzerland for the Lumnez landslide (Raetzo et al. 2006, 2007), and in Slovakia for the area of
Spis Castle and Lubietova. Applications from the Italian initiatives include studies for Assisi, Bindo,
Abruzzo from the MORFEO project.
Page 28/55
Date 18/04/2012
Within the SLAM project, the Landslide displacement monitoring was designed to accurately
quantify the displacement rate of areas or points inside the area of interest. The outputs could be
used to monitor the areas characterized with high hydrogeological risk, and to test the effectiveness
of structural interventions. This product has been tested at different sites on Arno and LiriGarigliano-Volturno (LGV) river basins. In the former, two test sites (Chiusi della Verna and
Terranova Bacciolini areas, Italy) were studied. The LGV basin was studied only, choosing two test
sites located in the Avellino Province. For Switzerland, this product was tested on six test sites
chosen among Grindelwald, Grubenglet-scher, Grächen, La Frasse, Montagnon and Lumnez.
Further applications are those from the ongoing DORIS project carried out for Assisi and
Collazone, and the area of Mt. Nebrodi in Italy, for Rácalmás and Dunaszekcső in Hungary, for the
Tramuntana Range in Spain, and for Zermatt and St. Moritz-Engadine Valley in Switzerland. Add
Further Examples Here & Respective References.
Figure 9: Example of monitoring applications using EO satellite data: case of Lumnez, Switzerland (modified
after Raetzo et al. 2006).
The two main SAR products generated for the landslide hazard are the landslide inventory and the
landslide monitoring product. For both these products a significant multi-annual stack of
interferometric observations is required. Both ascending and descending image geometries are
required with the same high priority to achieve coverage on slopes with different expositions. PSI
type processing is complemented by individual differential interferograms using preferably pairs
Page 29/55
Date 18/04/2012
with very short spatial baselines (<< 50m) and different time intervals (days, weeks, months, years)
to map landslides above the tree-line on other sparsely vegetated slopes.
In the United States, the US Geological Survey (USGS) conducts research science on landslide
hazards across the United States.5 Their research program relies on in situ field instrumentation
combined with space-borne and airborne optical imagery, ground-based and airborne LiDAR, with
very limited use of satellite radar and airborne UAVSAR (NASA). High-resolution optical satellite
EO data are periodically analyzed on a case-by-case approach to primarily map landslides following
a large storm or earthquake. The objective is to collect an inventory of landslides (snapshot in
time) to understand the geomorphic response and hazard potential of large storms or earthquakes
thereby contributing to the development of predictive models for future events.
For example, the USGS used space-borne and aerial imagery over Haiti following the 2010
earthquake to map the extent of landslides, thereby providing a geomorphic constraint on the
distribution of ground shaking in a country with few near-field seismic stations. Satellite EO
imagery is directly used for wildfire burn severity mapping; an assessment of what percentage of
the unburned vegetation remains follow a fire, a key factor in debris-flow risk modeling of the fire
area.
The limited use of satellite EO data by USGS for routine monitoring of landslides is a function of
the extent of the country, the spatial resolution of many of the data sources, heavily vegetated
landslides, and finite resources to acquire and process the imagery at the level needed for a
comprehensive monitoring program. Traditional C-band satellite InSAR data are of limited value
for landslide detection and monitoring, particularly from a hazard assessment standpoint. This is
especially true in areas with significant vegetation or where the landslide slip rates require routine
tasking on every orbit to minimize signal decorrelation and detection thresholds. Alternatively, PSI
processing approaches have been shown to be very effective at monitoring landslides, but require a
significant data archive that is currently lacking for most of the landslide prone regions in the
United States.
The USGS Landslide Hazards Program will increasingly be using remote-sensing technologies for
landslide research over the next 5 to 10 years, with a focus on fully characterizing the number of
active landslide across the United States and assessing the risk they pose. Until there is sufficient
5
http://landslides.usgs.gov
Page 30/55
Date 18/04/2012
SAR data archive for the United States to exploit more sophisticated PSI process approaches,
routine monitoring of the nations landslides will be limited in scope.
The ability to reliably and economically acquiring SAR data from sources such as TerraSAR-X,
TanDEM-X, RADARSAT-2, and COSMO SkyMED would greatly facilitate their use in assessing
landslide hazards in the US.
This includes rapid tasking, processing, and delivery during crisis
response, and a reduction in the cost of the data. Ideally, geohazard EO data are needed to
comprehensively assess the national landslide hazards and sufficient SAR imagery needs to be
collected and analyzed for the top 5 to 10 percent of the landslides that pose the greatest risk.
Page 31/55
Date 18/04/2012
From this the following requirements concerning the acquisition strategy arise:
A) space-borne SAR:
HR SAR (i) for landslide inventory and landslide hazard purposes: continuous observations
descending and ascending repeat coverage (at least 2 images per month in interferometric
mode), the focus is to extend and guarantee observations over mountainous and hilly
terrain with priority areas defined in section 2. Narrow orbital tubes are required to get
overall short spatial baselines and many pairs with very short spatial baselines. For
Sentinel-1 all ascending and all descending orbits should be considered. Single (HH or VV)
polarization would be sufficient.
VHR SAR: [(i) for hazard inventory purposes: continuous observations descending and
ascending repeat coverage (at least 2 images per month in interferometric mode). The
demonstration that this is also possible with VHR SAR is given by the COSMO-SkyMed
constellation which achieves over Italy full interferometric coverage with 16 day repeat
intervals in both ascending and descending orbits. (ii) for hazard monitoring purposes on
hotspots (e.g.. most critical landslides): continuous observations over one selected area in
all descending and ascending repeat orbits (e.g. TerraSAR-X every 11 days) means that no
data are then available for areas outside of this swath. If possible (e.g. using COSMOSkyMed constellation of 4 satellites) a full spatial coverage with continuous observations
descending and ascending repeat coverage (at least 2 images per month in interferometric
mode) is required. If this is not possible (e.g. TerraSAR-X) the requirement is to pre-select
for both ascending and descending geometry a set of modes which achieve full spatial
coverage over the landslide areas and then to acquire as much interferometric data in these
modes as possible.
B) space-borne Optical:
HR Optical/VHR Optical: (i) for landslide inventory and landslide hazard purposes to
provide background reference imagery: archive image (no more than 10-years old),
panchromatic or true colour composite. (ii) for hazard inventory purposes (e.g. historical
hazard mapping): VHR optical (no more than 1-year old), better than 5m resolution,
panchromatic or true colour composite, stereo pair 1 year apart useful for deliniation; (iii)
for hazard monitoring purposes (including early warning and response): repeat data VHR
Page 32/55
Date 18/04/2012
optical with frequency of monitoring frequency (better than 5m resolution, panchromatic or
multispectral).
6.2
Emerging research
a)
Modelling
The main objective of the LandSlide Modelling (LSMd) service is to develop and validate a
methodology combining (i) space measurement of past displacement derived from InSAR and/or
PSI analyses, (ii) conventional in situ investigations and (iii) geotechnical modeling, to characterize
and predict the risk associated to slope instability under heavy rain, and to support the design of
appropriate mitigation measures.
Figure 8: The location of the measured PSI displacements at Kerasia village
and the sections studied. White diamonds represent boring locations. In
the lower left box typical damage at a house.Example of modelling
applications using EO satellite data: case of Kerasia, Greece (Moretti S.,
Raspini F., Cigna F., Cooksley G., Banwell M.J., Wegmuller U., Strozzi T.,
Raetzo H., Stamatopoulos K. Landslide inventory, monitoring and
modelling using Persistent Scatterer Interferometry in the framework of
the European project Terrafirma. The 11th International & 2nd North
American Symposium on Landslides).
Page 33/55
Date 18/04/2012
b)
Early Warning Systems & Forecasting
For rainfall-induced landslides, early warning systems (EWS)exploit the empirical observation that
a minimum amount of precipitation is necessary to trigger landslides (Reichenbach et al. 1998,
Wieczorek 1996, Guzzetti et al. 2007). Regional to national warning systems based on empirical
rainfall thresholds and systematic rainfall measurements or forecasts, are – or have been –
operational, e.g., in Hong Kong (Premchitt et al, 1994), the San Francisco Bay region (Keefer et al.
1987), Rio de Janeiro (D’Orsi et al, 1997), Nagasaki (Iwamoto, 1990), Jamaica (Ahmad, 2003), the
Piedmont region (Aleotti, 2004), and the Yangtze River (International Early Warning Programme,
2005). The National Aeronautics and Space Administration (NASA) has developed a global system
to forecast the possible occurrence of landslides (and floods) based on near real-time rainfall
estimates obtained through a Multi-satellite Precipitation Analysis (TMPA) system (Hong et al.
2006).
In Italy, since October 2009, the Italian National Department of Civil Protection, uses a prototype
national landslide warning system based on two main components (Brunetti et al. 2009): (i) a set of
empirical thresholds for the possible occurrence of rainfall-induced landslides, and (ii) an ensemble
of small scale, national (synoptic) landslide hazard and risk zonations. The warning system
compares rainfall measurements (obtained from a national network of more than 1950 rain gauges)
and quantitative rainfall forecasts (an output of Limited-Area Meteorological models) with
empirical rainfall thresholds, to inform “where” and “when” landslides are expected in a given
region. Hazard and risk zonations are used to establish if the expected slope failures occur in areas
that are considered highly susceptible to landslides, or where landslide risk to the population is
severe or significant (Brunetti et al. 2009).
Within the SAFER project, the Real-time Shallow Landslide Forecasting (RSLF) service
represented a thematic assessment map at national scale in Italy consisting of a near real-time
prediction of the occurrence of shallow landslides on a large scale. The rainfall threshold method
tested in the Emilia-Romagna region was applied to the whole Italian territory. The DPC divided
the whole national area into 46 alert zones and supplied the historical rainfall data with 3,000 rain
gauges located in the whole national territory.
In order to select the most appropriate rain gauges for every alert zone, the length of the available
data series and their central position in the area were considered. For every zone a representative
rain-gauge was identified, for which the thresholds were built considering the cumulated rainfall
from 1 to 245 days. In this way three rainfall thresholds could be defined for each cumulative
Page 34/55
Date 18/04/2012
period, corresponding to three different alert levels. To compare the real data with these thresholds
value the estimated rainfall volume for the next 24 hours provided by DPC, and the rainfall
measured by the rain-gauges in the previous 244 days were considered. Everyday a cumulative
rainfall value was obtained and compared to the thresholds, to forecast an alert level 24 hours early.
When the higher threshold value was exceeded, the second level of the operational chain was
automatically launched. The second level of the Real-time Shallow Landslide forecasting was
updated and modified to operate on a national scale. It was based on the slope stability simulator
HIRESSS, that can operate at high spatial resolution on a large scale running in supercomputer
facilities. The reliability of the second level of the forecasting system was clearly dependent on
quality and precision of the national scale geotechnical data.
Figure 9. Example of a map
obtained by the comparison
between the predicted and
observed cumulative rainfall
and the thresholds.
Page 35/55
Date 18/04/2012
7
The way forward
There are four fundamental questions that concern the use of Satellite EO to support the seismic
hazard risk management community:
• What objectives does this community need to achieve over the next 5 to 10 years?
• What factors can accelerate the realization of these objectives? (looking at technology &
services and looking at science);
• Is the international community ready to collectively address the challenges associated with
these objectives? (role of mandated organisations, role of international organisations, role
of industry? are new partnerships needed?);
• What about other users not using Satellite EO? (globally, many users are not aware or able
or cannot afford space technologies).
The strategy to be undertaken by the landslide community in the next 5-10 years builds upon the
history of achievements, goals and criticalities brought to light through the experimentations,
applications and services carried out by the community in the last decade. Today, EO
technologies already play a strong role as support for the hazard and risk applications concerning
landslide processes, ranging from landslide mapping at the regional scale, monitoring of single
slopes, to even modelling of landslide motions and correlation with triggering factors.
As analysed in section 5, from the European perspective many national and international projects
and initiatives allowed covering a very good portion of the European priority; in the last 10 years
EO-based landslide applications covered more than 50 areas of interest with inventory, monitoring
and modelling services, which are distributed mainly in Italy, Switzerland, Greece, Spain, Slovakia,
Hungary. To a rough approximation, their areas of interest totally cover about 70-80,000 km2,
which correspond to about 35-40% of the landslide hazard priorities shown by the GDLND in 2005.
From the point of view of the spatial coverage of the landslide services, it is clear that in the next
years there will be a significant need for the landslide community to focus on those areas that were
not investigated so far, which – for Europe scale – are mainly located in Austria, Bulgaria,
Romania, Serbia, Bosnia, Albania and Turkey (Cf. figure 7, in section 6 above).
From the European perspective, these areas are reasonably considerable as the upcoming target
areas for the landslide community, and by focusing on them it is believed that an additional 25-30%
of the European priority will be addressed in the next decade. As for the EO-based applications
Page 36/55
Date 18/04/2012
under development, the whole community will certainly work on the strengthening of the
methodological approaches for EO-based landslide modelling and early-warning, which - in turn will be clearly supported by the copious availability of EO data at MR, HR and VHR from both
optical and radar sensors of the different space agencies.
Bearing these objectives in mind, the priorities and requirements to further improve EO-based
applications supporting landslide management, and to make them more consolidated and accepted
by the user communities, needs the engagement of all the actors of the community - both
operational and scientific – and of the EO data providers as well. The acceleration of the landslide
objectives therefore involve technological, scientific and operational factors, as well as national and
international organizations with a mandate in landslide hazard and risk management, such as civil
protection authorities and geological surveys.
On the one hand, the space agencies such as ESA and CSA will need to guarantee continuity of
acquisitions over the areas at higher landslide priority, by employing also improved spatial and
temporal resolutions, wider area coverage, and sustainable costs and delivery times of the EO
products (see Section 7.1). Further advances on technology will, on the other hand, include the
reinforcement of the computing capacity to support large volumes of EO data and broadening of
the use of wide area processing strategies, which will be aided by a careful validation of its
performances for the different temporal and spatial scales of the landside applications performed
by the scientific community (see Section 7.2).
From the European perspective, the next decade will also face the operational implementation of
the GMES Initial Operations (GIO) in 2011-2013 and the fully operational services starting from
2014, which will certainly enhance the communication between the scientific and operational
community of users, by bridging the gap between science and operational applications, through the
provision of mature landslide mapping and monitoring services, in the thematic and emergency
support frameworks (see Section 7.3).
By distinguishing between technological, scientific and operational priorities, the agenda of the
landslide community for the next decade is discussed in detail in the three sections below, with a
perspective primarily focused on Europe. This is expected to be complemented and enhanced by
non-European community members with outside-Europe observational, technological and
scientific requirements.
Page 37/55
Date 18/04/2012
Factors that can accelerate the realization of these objectives can be grouped in three categories:
technology and services, science, users.
7.1
Technology & services
Taking into account the feedback received in the last years from the users of EO-based services of
EU and ESA projects such as Terrafirma, SAFER and PREVIEW, and in light of the results of recent
experiments carried out by the scientific community, the EO based technologies, applications and
services can be considered already quite mature.
To preserve their maturity and worth, EO data and EO-based services and applications need to
address specific observational requirements to be able to support the identification, mapping and
monitoring of landslide processes; they can be summarized as follows:
-
Continuity and consistency of acquisition of EO optical and radar data, to guarantee the
availability of image stacks and archives in the coming years, and allow the comparison of
recent and past scenarios of landslide evolution and consequences. As for the radar
imagery, the requirement involves the need of geometrical consistency (i.e. acquisition
parameters of each radar stack must be kept identical for the whole set of images), to
guarantee their suitability to be processed through conventional and multi-temporal
interferometric approaches.
-
Wide geographical coverage of EO imagery acquired during the background planning, to
give the community the chances of activating EO based studies of deformation processes in
areas affected by landslide events not previously monitored with either EO data or on site
instrumentation.
-
Improved temporal resolutions (shorter revisit times) and regularity of acquisitions to
enlarge the range of applicability of EO-derived motion services to landslides faster than a
few tens of centimetres per year (e.g. current limitation of most InSAR and PSI-based
landslide products, due to monthly acquisition frequencies of C band data), and to
guarantee proper and systematic temporal sampling of the observed phenomena. The
Sentinel-1 mission is expected to largely increase the contribution of SAR based
observation of landslides to support historical hazard mapping and operational
monitoring. This is primarily due to its systematic observation capabilities with High
Resolution and large swath with a high temporal sampling (12 days, 6 with two platforms).
Recent experimentations with COSMO-SkyMed X band data at weekly repeat cycles have
shown the unmatched precision and level of details achievable with EO data acquired with
Page 38/55
Date 18/04/2012
improved temporal resolutions, a crucial requirement for landslide monitoring and earlywarning practices.
-
Improved spatial resolutions of EO radar data, guaranteeing High Resolution acquisitions
for the priority areas at highest landslide risk, to enhance the capability and scale of
applicability of the derived motion services and include – among the phenomena that can
be monitored - those evolving at local- to single slope-scales. Landslide services based on
VHR radar data have allowed the estimation of ground motions with scales and level of
details up to 5-10 times higher than MR and HR data based products, with significantly
improved capability of detecting and mapping landslide-induced deformation.
-
Availability of dual-mode SAR acquisitions (i.e. ascending and descending) for hilly and
mountainous areas, where the landslide products are strongly influenced by the visibility
of the slopes and their orientations; dual-mode datasets allow to better constrain landslide
motions (by combining the velocity estimated along the two geometries), and increase the
number of slopes monitored within the observed areas (by increasing the chances of
detecting slopes with different orientation and steepness)
-
Sustainable costs of EO data and derived products, to enhance the affordability and ease of
acquisition of EO imagery and their derivatives, and increase the efficiency-to-costs ratio
of the EO-based landslide products.
-
Timeliness of the EO data access/distribution, to guarantee suitability of EO based
landslide products in the framework of the emergency response practices.
Most of these requirements and challenges will be fully addressed by the upcoming European
Radar Observatory (ERO) of ESA, that will ensure the continuity of SAR C band ERS 1/2 and
ENVISAT ASAR data with the launch of the Sentinel-1 constellation (first Sentinel-1 satellite, 1A,
due to be launched in mid 2013, and the second satellite, 1B, to be launched in 2015). The
constellation will guarantee improved and more regular coverage compared to their predecessors,
and provide imagery for the GMES user community over Europe, Canada and main shipping
routes, delivered within an hour of acquisition, thus fitting most of the technological needs of the
community. The three Canadian satellites of the RADARSAT constellation (RCM), to be launched
beginning in 2014, will also address these needs by ensuring C-band SAR data continuity after
RADARSAT-2, improved operational use and enhanced revisit times (e.g. 4 day cycles), a wide
range of spatial resolutions (from 100 m up to 3 m) and daily access to 95% of the world to
Canadian and international users. As for the optical imagery, the Sentinel-2 pair will also
substantially contribute to respond to these needs, by systematically acquiring HR data globally
Page 39/55
Date 18/04/2012
and guaranteeing continuity of SPOT and Landsat data by providing optical acquisitions in the
visible, near infrared and short infrared bands.
In addition to the above observational requirements, which have been specifically analyzed by
GMES in the last years and will be faced through the introduction of Sentinel-1 and Sentinel-2
constellations of ESA, the priorities concerning EO data and technology to guarantee further
advances and continuity of the EO-based landslide services in the next years include:
-
Reinforce computing capacity and capability to fully support large data volumes such as
those that will be available once the satellites of the Sentinel-1/2 and RADARSAT
constellations will be fully operating.
-
Make broader use of wide area processing strategies of satellite SAR imagery, as those
employed in the framework of the nationwide PS processing of ERS1/2 and ENVISAT data
for the Extraordinary Plan of Environmental Remote Sensing (EPRS-E) of the Italian
Ministry of Environment and Territory of the Sea (METS), or the WAP (Wide Area
Processing) strategy promoted by the Terrafirma project.
-
Implementing the Emergency Management Services (EMS) in the framework of the GMES
Initial Operations (GIO) 2011-2013 plan, and the GMES fully operational services starting
from 2014, with main focus on emergency-response (rather than risk assessment); similar
services have been provided during the GMES built-up phase in 2009-2011 by the FP7
SAFER project through the landslide mapping, monitoring and forecasting services, in the
thematic and emergency support frameworks.
7.2
Science
Accounting for the remarkable improvements achieved in the last two decades with the progressive
development of EO technologies and their integration into landslide related research and
applications, and considering also the upcoming advances that will be achieved with the thorough
exploitation of new EO satellites and derived data, further efforts are still needed from the scientific
community to make EO-based landslide services more consolidated. In particular, some of the
scientific objectives and strategies that will be undertaken by the scientific community include:
-
Development and further enhancement of the emerging techniques for EO-based landslide
modelling and early warning purposes;
Page 40/55
Date 18/04/2012
-
Validation and assessment of the performances of wide area processing strategies (e.g. the
Terrafirma WAP and the EPRE-E data) for landslide hazard and risk studies, considering
real, near-real and deferred time applications; while these processing strategies present
challenges in alpine environments, they are well suited to built-up and urban areas;
-
Preparedness for the near-future exploitation of EO radar data from the ESA ERO, i.e. the
Sentinel 1 constellation;
-
Standardization of the methodologies employed for the implementation of EO-based
landslide services, creating guidelines for the interpretation of EO data and their
derivatives (e.g. PSI products) aimed at landslide mapping, monitoring and modelling. A
step forward to this objective is currently undertaken for instance by the EU FP7 project
PanGeo, by creating a standardized procedure to be followed by the Geological Surveys for
the interpretation of PSI products for the identification and mapping of geohazards
affecting urban areas in Europe6, and trying to make it compliant with the INSPIRE
directive7; applicability of this procedure to landslide mapping will certainly need to be
further improved to better address the specific needs and requirements of the landslide
community of operational and scientific users.
-
Improve the communication to the end-user by bridging the gap between science and
operational application. Particularly at local and regional level the limited knowledge
about the potentials and capacities of EO data may hinder further usage of satellite based
information. An improved communication also includes the information about the
constraints of the EO data based technologies. In order to convince the end-user to apply
EO data based services we need to build trust that we only achieve by openly discuss also
the limits of the EO data approaches.
7.3
Users and practitioners
Considering the already high level of maturity of the EO-based technology and derived products for
landslide hazards and risk, one of the main targets for the landslide community is to further act on
the level of acceptance and understanding of these technologies in the end user community.
Although EO technologies are already accepted and widely employed by the operational landslide
communities of some countries such as Italy and Switzerland, the landslide experts need to focus
6
7
http://www.pangeoproject.eu
Directive 2007/2/EC - OJ L 108 of 25.4.2007
Page 41/55
Date 18/04/2012
further on the strategies to make EO-based landslide services accepted by the community of users
and practitioners of other European and non-European countries. Thus, future perspectives of the
landslide experts for the next decade include the following objectives:
-
Improve the accessibility of EO based landslide products, attract new end users and
enhance their understanding and knowledge on EO technologies and their potentials in
support of hazard and risk management; the PanGeo project is actively contributing to this
purpose, by providing free and open-access geohazard information services for more than
50 towns of Europe, and encouraging the European geological surveys, decision-makers,
regulators and civil protection agencies to systematically assess geohazards with the
support of technologies based on EO radar data.
-
Enhance the acceptance of EO data based products in the end user community of Europe
and worldwide, by demonstrating their complementarity with on site surveys,
conventional and ground-based monitoring techniques.
-
Stimulate further integration of EO data based products into every-day practices in the
framework of landslide risk management, to support all phases of the disaster
management cycle, from prevention, preparedness and emergency response, to postemergency and recovery activities (relocation of elements at risk and reconstruction
planning) and mitigation strategies; using the examples of successful applications from the
pre-operational GMES emergency support landslide services of the SAFER project
activated in response to emergency situations in Europe (e.g. Rapid Landslide Mapping in
L’Aquila, after the 2009 earthquake), extend emergency support services to other
countries.
Page 42/55
Date 18/04/2012
8
References
Ahmad R (2003) Developing early warning systems in Jamaica: rainfall thresholds for hydrological
hazards. National Disaster Management Conf., Ocho Rios, St Ann, Jamaica, 9–10 September 2003.
http://www.mona.uwi.edu/uds/rainhazards_files/frame.htm.
Aleotti P (2004) A warning system for rainfall-induced shallow failures. Eng Geol 73: 247–265
Barlow, J., Franklin, S., Martin, Y., 2006. High spatial resolution satellite imagery, DEM
derivatives, and image segmentation for the detection of mass wasting processes. Photogrammetric
Engineering and Remote Sensing 72, 687–692.
Bianchini, S., Cigna, F., Casagli, N.: Improving landslide inventory with Persistent Scatterers in
Calabria, Italy. In: Proceedings of The Second World Landslide Forum, WLF2 2011, Putting science
into practice, Rome, Italy, 3-9 October 2011, in press.
Bianchini, S., Cigna, F., Righini, G., Proietti, C., Casagli, N.: Landslide HotSpot Mapping by means
of Persistent Scatterer Interferometry. Environmental Earth Sciences, doi: 10.1007/s12665-0121559-5. 1-18, 2012.
Booth, A.M., Roering, J.J., Perron, J.T., 2009. Automated landslide mapping using spectral
analysis and high-resolution topographic data: Puget Sound lowlands, Washington, and Portland
Hills, Oregon. Geomorphology 109, 132–147. doi:10.1016/ j.geomorph.2009.02.027.
Brabb, E.E., 1991. The world landslide problem. Episodes 14 (1), 52–61.
BRGM (Bureau de Recherche Géologiques et Minières) 2007. IGOS Geohazards Theme Report,
BRGM/RP-55739-FR, August 2007.
Brunetti, M. T., S. Peruccacci, M. Rossi, F. Guzzetti, P. Reichenbach, F. Ardizzone, M. Cardinali, A.
C. Mondini, P. Salvati, and G. Tonelli (2009), A prototype system to forecast rainfall induced
landslides in Italy, The First Italian Workshop on Landslides.
Cheng, K.S., Wei, C., Chang, S.C., 2004. Locating landslides using multi-temporal satellite images.
Advances in Space Research 33 (3), 96–301.
Page 43/55
Date 18/04/2012
Colesanti C., Wasowski, J.: Investigating landslides with space-borne Synthetic Aperture Radar
(SAR) Interferometry. Engineering Geology, 88, 173-199, 2006.
Columbia University Center for Hazards and Risk Research (CHRR), the Norwegian Geotechnical
Institute (NGI), and the Columbia University Center for International Earth Science and
Information Network (CIESIN); International Bank for Reconstruction and Development/The
World Bank, Global Landslide Mortality Risks and Distribution, Global Landslides Total Economic
Loss Risk Deciles, Version 1.0, 2005, in: Dilley, M., R.S. Chen, U. Deichmann, A.L. Lerner-Lam, M.
Arnold, J. Agwe, P. Buys, O. Kjekstad, B. Lyon, and G. Yetman, Natural Disaster Hotspots: A Global
Risk Analysis, Version 1.0, Disaster Risk Management Series, No. 5, The World Bank, March 2005
Cruden, D.M., Varnes, D.J.: Landslide types and processes. In: Turner AK, Schuster RL (eds)
Landslides: Investigation and Mitigation, Sp. Rep. 247, Transportation Re-search Board, National
research Council. National Academy Press, Washington DC: 36-75, 1996.
Czuchlewski, K.R., Weissel, J.K., Kim, Y., 2003. Polarimetric synthetic aperture radar study of the
Tsaoling landslide generated by the 1999 Chi-Chi earthquake, Taiwan. Journal of Geophysical
Research 108 (F1), 7.1–7.11.
D’Orsi R, D’Avila C, Ortigao JAR, Dias A, Moraes L, Santos MD (1997) Rio-Watch: the Rio de
Janeiro landslide watch system. In: Proc. 2nd PSL Pan-AM Symposium on Landslides, vol. 1. Rio
de Janeiro, 21–30.
Dai, F.C., Lee, C.F.: Landslide characteristics and slope instability modeling using GIS, Lantau
Island, Hong Kong. Geomorphology, 42, 213-228, 2002.
Delacourt, C., Allemand, P., Berthier, E., Raucoules, D., Casson, B., Grandjean, P., Pambrun, C.,
Varel, E., 2007. Remote-sensing techniques for analyzing landslide kinematics: a review. Bulletin
de la Société géologique de France 178(2), 89-100.
EEA (European Environment Agency): Mapping the impacts of natural hazards and technological
accidents in Europe: an overview of the last decade. EEA technical report 13/2010. Copenhagen,
Denmark. doi:10.2800/62638, 2010.
Farina, P., Colombo, D., Fumagalli, A., Marks, F., Moretti, S.: Permanent Scatterers for landslide
investigations: outcomes from the ESA-SLAM project. Engineering Geology 88: 200-217, 2006.
Page 44/55
Date 18/04/2012
Feng, Z. (2011), The seismic signatures of the 2009 Shiaolin landslide in Taiwan, Natural Hazards
and Earth System Sciences, 11(5), 1559–1569, doi:10.5194/nhess-11-1559-2011.
Fiorucci, F., Cardinali, M., Carlà, R., Rossi, M., Mondini, A.C., Santurri, L., Ardizzone, F., Guzzetti,
F., 2011. Seasonal landslides mapping and estimation of landslide mobilization rates using aerial
and satellite images. Geomorphology 129 (1–2), 59–70. doi:10.1016/j.geomorph.2011.01.013.
Galli, M., F. Ardizzone, M. Cardinali, F. Guzzetti, and P. Reichenbach (2008), Comparing landslide
inventory maps, Geomorphology, 94(3-4), 268–289.
Glenn, N.F., Streuker, D.R., Chadwick, D.J., Thackray, G.D., Dorsch, S.J., 2006. Analysis of Lidar
derived topographic information for characterizing and differentiating land- slide morphology and
activity. Geomorphology 73, 131–148.
Guzzetti, F., A. C. Mondini, M. Cardinali, F. Fiorucci, M. Santangelo, and K.-T. Chang (2012),
Landslide inventory maps: New tools for an old problem, Earth Science Reviews, 112(1-2), 1–25,
doi:10.1016/j.earscirev.2012.02.001.
Guzzetti, F., M. Cardinali, P. Reichenbach, and A. Carrara (2000), Comparing landslide maps: A
case study in the upper Tiber River Basin, central Italy, Environmental Management, 25(3), 247–
263.
Guzzetti, F., S. Peruccacci, M. Rossi, and C. P. Stark (2007), Rainfall thresholds for the initiation of
landslides in central and southern Europe, Meteorology and Atmospheric Physics, 98(3), 239–267.
Hervás, J. (Ed.): Guidelines for Mapping Areas at Risk of Landslides in Europe. Proc. Experts
Meeting, JRC, Ispra, Italy, 23-24 October 2007. JRC Report EUR 23093 EN, Office for Official
Publications of the European Communities, Luxembourg, 53 pp, 2007.
Hong, Y., Adler, R., and Huffman, G.: Use of Satellite Remote Sensing Data in the Mapping of
Global Landslide Susceptibility, J. Nat. Hazards, 43, 245–256, 2007.
Page 45/55
Date 18/04/2012
Hong, Y., R. F. Adler, and G. J. Huffman (2006) Evaluation of the potential of NASA multi-satellite
precipitation analysis in global landslide hazard assessment, Geophysical Research Letters,
33(L22402).
ISDR (International Strategy for Disaster Reduction): Global Assessment Report on Disaster Risk
Reduction, 2011.
Iwamoto M (1990) Standard amount of rainfall for warning from debris disaster. In: Proc. 6th Int.
Conf. and Field Workshop on Landslides ALPS 90 (Cancelli A, ed), vol. 79b. Milano: Ricerca
Scientifica ed Educazione Permanente, 77–88.
Jaboyedoff, M., Oppikofer, T., Abellàn, A., Derron, M.H., Loye, A., Metzger, R., Pedrazzini, A.,
2010.
Use
of
LIDAR
in
landslide
investigations:
a
review.
Natural
Hazards
1–24.
doi:10.1007/s11069-010-9634-2.
Keefer DK, Wilson RC, Mark RK, Brabb EE, Brown WM- III, Ellen SD, Harp EL, Wieczorek GF,
Alger CS, Zatkin RS (1987) Real-time landslide warning during heavy rainfall. Science 238: 921–
925.
Kirschbaum, D. B., Adler, R., Hong, Y., and Lerner-Lam, A.: Evaluation of a preliminary satellitebased landslide hazard algorithm using global landslide inventories. Natural Hazards and Earth
System Science, 9, 673-686, 2009.
Kouli, M., Loupasakis, C., Soupios, P., Vallianatos, F., 2010. Landslide hazard zonation in high risk
areas of Rethymno Prefecture, Crete Island, Greece. Natural Hazards 52, 599-621.
Larsen, M.C., Wieczorek, G.F., Eaton, L.S., Morgan, B.A., and Torres-Sierra, H., 2001, Natural
hazards on alluvial fans; The Venezuela debris flow and flash flood disaster: U.S. Geological Survey
Fact Sheet 103–01.
Lauknes, T.R., Piyush Shanker, A., Dehls, J.F., Zebker, H.A., Henderson, I.H.C., Larsen, Y., 2010.
Detailed rockslide mapping in northern Norway with small baseline and persistent scatterer
interferometric SAR time series methods. Remote Sensing of Environment 114, 2097–2109.
doi:10.1016/j.rse.2010.04.015.
Page 46/55
Date 18/04/2012
Lee, S., Lee, M.-J., 2006. Detecting landslide location using KOMPSAT 1 and its application to
landslide-susceptibility mapping at the Gangneung area, Korea. Advances in Space Research 38
(10), 2261–2271.
Leprince, S., Berthier, E., Ayoub, F., Delacourt, C., Avouac, J.-P., 2008. Monitoring earth surface
dynamics with optical imagery. Eos Transactions AGU, 89(1).
Marcelino, E.V., Formaggio, A.R., Maeda, E.E., 2009. Landslide inventory using image fusion
techniques in Brazil. International Journal of Applied Earth Observation and Geoinformation 11,
181–191.
Martha, T.R., Kerle, N., Jetten, V., van Westen, C., Vinod Kumar, K., 2010. Characterising spectral,
spatial and morphometric properties of landslides for semi-automatic detection using objectoriented methods. Geomorphology 116, 24–36.
McKean, J., Roering, J., 2003. Objective landslide detection and surface morphology mapping
using high-resolution airborne laser altimetry. Geomorphology 57 (3–4), 331–351.
Metternicht, G., Hurni, L., Gogu, R., 2005. Remote sensing of landslides: an analysis of the
potential contribution to geo-spatial systems for hazard assessment in mountain environments.
Remote Sensing of Environment 98, 284–303.
Mondini, A.C., Chang, K.T., Yin, H.Y., 2011a. Combining multiple change detection indices for
mapping
landslides
triggered
by
typhoons.
Geomorphology
134
(3–4),
440–451.
doi:10.1016/j.geomorph.2011.07.021.
Mondini, A.C., Guzzetti, F., Reichenbach, P., Rossi, M., Cardinali, M., Ardizzone, F., 2011b. Semiautomatic recognition and mapping of rainfall induced shallow landslides using satellite optical
images. Remote Sensing of Environment 115, 1743–1757. doi:10.1016/j.rse.2011.03.006.
Moretti, S., Cigna, F., Raspini, F., Cooksley, G., Banwell, M.J., Raetzo, H., Herrera, G., Notti, D.,
Davalillo, J.C.: Use of Persistent Scatterer InSAR within Terrafirma Landslide Services. In:
Proceedings of The Second World Landslide Forum, WLF2 2011, Putting science into practice,
Rome, Italy, 3-9 October 2011, in press.
Page 47/55
Date 18/04/2012
Nadim, F., Kjekstad, O., Peduzzi, P., Herold, C., and Jaedicke, C.: Global landslide and avalanche
hotspots, Landslides, 3, 159– 173, 2006.
Nichol, J.E., Shaker, A., Wong, M.S., 2006. Application of high-resolution stereo satellite images to
detailed landslide hazard assessment. Geomorphology 76, 68– 75.
Passalacqua, P., Tarolli, P., Foufoula-Georgiou, E., 2010. Testing space-scale methodologies for
automatic geomorphic feature extraction from LiDAR in a complex mountainous landscape. Water
Resources Research 46, W11535. doi:10.1029/ 2009WR008812.
Premchitt J, Brand EW, Chen PYM (1994) Rain-induced landslides in Hong Kong, 1972–1992. Asia
Engineer, June, 43–51.
Raetzo H., Wegmüller U., Strozzi T., Marks F., Farina P.: Monitoring of Lumnez landslide with ERS
and ENVISAT SAR data. Proc. ENVISAT Symposium, 2007.
Raetzo, H., Lateltin, O., Bollinger, D., and Tripet, J.P.: Hazard assessment in Switzerland – codes of
practice for mass movements, Journal of IAEG, 2002.
Raetzo, H., Wegmüller, U., Strozzi, T., Marks, F., Farina, P., Monitoring of Lumnez landslide with
ERS and ENVISAT SAR data, ESA Fringe Conference, Montreux, 2006.
Reichenbach, P., M. Cardinali, P. De Vita, and F. Guzzetti (1998), Regional hydrological thresholds
for landslides and floods in the Tiber River Basin (central Italy), Environmental Geology, 35, 146–
159.
Righini, G., Pancioli, V., Casagli, N.: Updating landslide inventory maps using Persistent Scatterer
Interferometry (PSI). International Journal of Remote Sensing, 33(7), 2068-2096, 2012.
Rosin, P.L., Hervás, J., 2005. Remote sensing image thresholding methods for determining
landslide activity. International Journal of Remote Sensing 26 (6), 1075–1092.
Salvati, P., C. Bianchi, M. Rossi, and F. Guzzetti (2010), Societal landslide and flood risk in Italy,
Natural Hazards and Earth System Sciences, 10, 465–483.
Page 48/55
Date 18/04/2012
Sato, H.P., Yagi, H., Moarai, M., Iwahashi, J., Sekiguchi, T., 2007. Airborne lidar data measurement
and landform classification mapping in Tomarino-tai landslide area, Shirakami Mountains, Japan.
In: Sassa, K., Fukuoka, H., Wang, F., Wang, G. (Eds.), Progress in Landslide Science. Springer,
Berlin, pp. 237–249.
Singhroy, V., Molch, K., 2004. Characterizing and monitoring rockslides from SAR techniques.
Advances in Space Research 33 (3), 290–295.
Strozzi T., P. Farina, A. Corsini, C. Ambrosi, M. Thüring, J. Zilger, A. Wiesmann, U. Wegmüller,
and C. Werner, Survey and monitoring of landslide displacements by means of L-band satellite SAR
interferometry, Landslides, Vol 2, Nr. 3, pp 193-201, Oct. 2005.
Strozzi T., U. Wegmüller, H.R. Keusen, K. Graf, and A. Wiesmann, Analysis of the terrain
displacement along a funicular by SAR interferometry, IEEE Geoscience and Remote Sensing
Letters, Vol. 3, No. 1, pp. 15-18, 2006.
Strozzi T., R. Delaloye, D. Poffet, J. Hansmann and S. Loew, Surface subsidence and uplift above a
headrace tunnel in metamorphic basement rocks of the Swiss Alps as detected by satellite SAR
interferometry, Remote Sensing of Environment, Volume 115, Issue 6, Pages 1353-1360,
doi:10.1016/j.rse.2011.02.001, 15 June 2011.
Tarolli, P., Sofia, G., Dalla Fontana, G., 2010. Geomorphic features extraction from high resolution
topography: landslide crowns and bank erosion. Natural Hazards. doi:10.1007/s11069-010-9695-2.
Tsai, F., Hwang, J.-H., Chen, L.-C., Lin, T.-H., 2010. Post-disaster assessment of landslides in
southern Taiwan after 2009 Typhoon Morakot using remote sensing and spatial analysis. Natural
Hazards and Earth System Sciences 10, 2179–2190. doi:10.5194/nhess-10-2179-2010.
Van Den Eeckhaut M., Hervás J., Jaedicke C., Malet J. P., Montanarella L., Nadim F.: Statistical
modelling of Europe-wide landslide susceptibility using limited landslide inventory data.
Landslides. doi:10.1007/s10346-011-0299-z, 2011.
Van Westen, C.J., Castellanos, E., Kuriakose, S.L., 2008. Spatial data for landslide susceptibility,
hazard, and vulnerability assessment: an overview. Engineering Geology 102, 112–131.
Weirich, F., Blesius, L., 2007. Comparison of satellite and air photo based landslide susceptibility
maps. Geomorphology 87 (4), 352–364.
Page 49/55
Date 18/04/2012
Wieczorek GF (1996) Landslide triggering mechanisms. In: Landslides: investigation and
mitigation (Turner AK, Schuster RL, eds). Washington DC: Transportation Re- search Board,
National Research Council, special report, pp 76–90
Yin, Y., F. Wang, and P. Sun (2009), Landslide hazards triggered by the 2008 Wenchuan
earthquake, Sichuan, China, Landslides, 6(2), 139–152, doi:10.1007/s10346-009-0148-5.
Page 50/55
Date 18/04/2012
ANNEX1.
National Landslide Initiatives in a European Context
Italy:
Recent national initiatives funded by the Italian National Civil Protection Department
(Dipartimento di Protezione Civile,DPC) have been carried out to enhance the acceptance of EO
based applications and services for landslide risk.
At the national level since 2005 DPC carried out two research projects: SAR.net in 2005-2008 and
SAR.net2 since 2009 and still on going. The objectives of the SAR.net projects are:

the updating of an information system multiblog MIG (Multi-risk Information Gateway)
aimed at developing a network of technical and scientific resources in support to DPC
activities;

the implementation of the data analysis derived from the National system of satellite
surveillance which make use of remote sensing data for mapping, monitoring and analysis
of risk scenario related to mass movements;

the definition of operational procedures for the rapid assessment of landslide risk for
emergency situation management;

monitoring of the deformation of the Stromboli volcano by means of GB-InSAR;

the development of a national early warning system based on meteorological now casting
and real time prediction of rapid shallow landslides.
Other national projects, such as the MORFEO (MOnitoraggio del Rischio da Frana mediante dati
EO, Monitoring landslide risk exploiting EO data) 2008-2010 (http://www.morfeoproject.it)
funded by the Italian Space Agency (ASI), provided valuable examples on how applications based
on EO and non-EO data can operatively support the Italian Civil Protection Department in the
process of mapping, prevention and management of landslide risk during emergencies. The project
delivered landslide-monitoring services for test sites in three Regions in Italy.
In reference to the European scenario, in particular concerning to international project related to
ground mass movement risk management, the DPC has been involved in many international
projects, still ongoing or recently closed, as: PREVIEW, (FP6) SAFER, and DORIS (FP7) and
Terrafirma (funded by ESA). Thanks also to these projects, European countries have reached
Page 51/55
Date 18/04/2012
consistent advances in the study of ground movements and have obtained a primary position in the
related risk mitigation activities. In this framework, the previous projects, dealing with multiple
ground deformations, chiefly mass movements and land subsidence, are aimed at improving the
understanding of the complex phenomena that cause ground deformations, including mass
movements and land subsidence, and to foster the current capabilities of Civil Protection
Authorities (CPA) and services, at different administration and operational levels, to manage the
risks posed by different ground deformations. Landslide mapping and monitoring EO systems have
been interested by several and relevant improvements over the past few years and many of these
new developments have already been consolidated in operative scenarios. The exploitation of new
technologies has been used to advance knowledge and to prepare products and tools useful to
different actors, including CPAs, private companies and public organization in charge with
landslide risk mitigation activities. In Europe, and in many other parts of the world, there is a clear
need for (i) high-quality data on landslide vulnerability, (ii) organized information and tools for
modelling and forecasting the impact that landslide events can have on a territory, and (iii) specific
modelling tools to support the preparedness and the recovering and reconstruction phases of a
landslide crisis management cycle. CPA takes advantage from the EO-based project and initiatives
aimed to restrict the existing gap among techniques/sensors developers, data/signal processing,
monitoring/mapping methods and operational applications within early warning procedures and
risk mitigation activities in the civil protection emergency cycle.
When a disaster occurs, the emergency phase is managed first at the national (or lower, i.e. regional
or municipal) level, and can be managed at higher level when and where requested (subsidiarity
principle). After an event, CPA organizations could be prepared to cope effectively with the damage
caused by a landslide event, and to have clear strategies for recovery and reconstruction actions
based on sound, state-of-the-art scientific understanding of landslide phenomena and their spatial
and temporal evolution.
Contact with the world of scientific research is increasingly close as well as the use of new
technological systems (such as EO data and technique) to estimate the foreseen damages and to
monitor vulnerable zones in case of extreme events.
In this framework Centers of Competence, CTS
following a specific request from the Civil
Protection department design an operational service to support CPAs dealing with landslide
mitigation and preparedness, and for post-event recovery and reconstruction activities. CFSE are
expected to deliver products (including methods, tools, and specifications) to users operating at
different administrative and organizational levels, from local to continental levels. Each Center of
competence involved in the emergency phase provides services, information, data, processed data
and technical-scientific contributions in specific fields.
Page 52/55
Date 18/04/2012
After the occurrence of a landslide event-related emergency, in the framework of the post-disaster
activities leaded by the National Department of Civil Protection, several activities were performed
by CTS for geo-hazards, aiming at supporting regional authorities in the emergency management
and hydro-geological risk assessment. In particular CPA benefits from the EO data and technical
ability to detect and map landslides (measuring rate of movement and/or extent of the occurred
phenomenon), and to assess and forecast the hazard of landslides of different types of element at
risk (structures, infrastructures and landscapes).
Switzerland:
In Switzerland the natural hazard division of the Federal Office for the Environment (FOEN)
started in 2006 with the exploitation of satellite based SAR-data in order to have a Monitoring and
Early Warning System (MS/EWS) for landslides and other mass movements (e.g. rock glaciers,
subsidence). Within the data model the process type and the velocity (activity) according to the
Swiss guideline for mass movements are defined (Raetzo et al. 2002). The interpretation on the
spatial distribution and on the dynamic of the landslides is mainly based on InSAR and PSI
techniques. For InSAR analyses ERS (since 1991), JERS, ENVISAT, ALOS, TerraSAR-X and
COSMO-SkyMed data of ascending and descending modes are used. While PSI technique is mainly
based on ERS and ENVISAT data due to the spatial and temporal coverage. A large set of geodetic
data are used for cross validation and in areas with missing SAR data.
The combined use of EO and ground measurements increases the database quality and the number
of the monitored landslides in a so called “InSAR map”. Orthophotos, geological and topographic
data are also integrated in the GIS. The landslide inventory is validated by a large number of
partners coming from cantonal and national institutions, private companies and universities.
Hazard maps according to the Federal Laws take into account available InSAR maps. As large parts
of the country were not investigated before satellite based SAR is a substantial contribution for
detection and mapping of slope instabilities. Landslide areas within important elements at risk can
be monitored by conventional methods (e.g. inclinometers, GPS) and satellite based SAR. As EO
data cover large areas and high resolution SAR data have limited availability a national strategy for
monitoring and Early Warning System is required. Based on the successful contributions of the
Terrafirma services, InSAR and PSI data are integrated in the Swiss landslide risk management.
Page 53/55
Date 18/04/2012
Spain:
As for Spain, the project DO-SMS 2009-2011, part of the Territorial Cooperation Program SUDOE
between France and Spain (dosms.get.obs-mip.fr/cosiweb), aimed at developing tools for ground
deformation monitoring and sustainable land management for natural hazards, and delivered 3
landslide mapping services in Spain and France.
At the European scale, the landslide services and applications exploiting EO satellite technologies
are distributed mainly in Italy, Switzerland, Greece, Spain, Slovakia, Hungary.
Page 54/55
Date 18/04/2012
APPENDIX 1.
Scope of the community papers prepared for the May 2012 International
Forum on Satellite Earth Observation for Geohazard Risk Management
Different geohazard risk communities represented at the Forum are working on community papers
with the scope to describe the state-of-the-art concerning research & applications and define a 5 to
10 year vision on the use of data from new and planned Satellite EO missions. The types of
geohazard addressed cover primarily Solid Earth & natural geohazards and the following themes
are identified:
I) Seismic hazards
II) Coastal lowland subsidence & flood defence
III) Landslides
IV) Volcanoes
V) Other hydro-geological risks (such as groundwater management & inactive mines)
Focus is on the different phases of risk management, in particular hazard identification,
quantification and monitoring for prevention and preparedness. The contribution of Satellite EO is
considered primarily concerning hazard mapping and risk assessment. Other publications
concerning Satellite EO for exposure or asset mapping are available such as for instance the report
‘Using high resolution satellite data for the identification of urban natural disaster risk’ (Uwe
Deichmann, Daniele Ehrlich & al.) prepared in association with EC/JRC and published by the
Global Facility for Disaster Reduction and Recovery (GFDRR).
Draft community papers will be distributed before the event, will be discussed at the Forum and
will be available for on line for download and open review after the event (www.int-eo-geo-hazardforum-esa.org) with the aim of publishing final papers to be released by the conference coorganizers. This will be used to produce a scientific and technical publication.
Following the International Forum on Satellite Earth Observation for Geohazard Risk
Management, a scientific and technical report will be released as a joint ESA-GEO publication.
Page 55/55
Date 18/04/2012