Catálogo de publicaciones - libros

Disaster management takes current information technology to its limits due to the very large amount of relevant data and tight response times. This is in particular true for satellite data which needs to be retrieved from different archives, processed an interpreted. Typical issues are efficiency of database access, integration of multiple heterogeneous sites and quality of image interpretation. Unfortunately experiences from many different projects and systems have shown that these issues cannot be considered independently from each other. Therefore this paper proposes a concept addressing the mentioned issues by providing an integrated solution. For this purpose the paper focuses on content based searches for satellite images in a distributed environment.

Australia has a three-tiered hierarchal model of government. A single Federal government, eight State/Territory governments and approximately seven hundred municipal councils make up the three tiers. Each of these tiers, and the separate jurisdictions within the tiers, can have their own standards and arrangements for managing information useful for Emergency Management (EM). Other information resources are held by private organizations. The business drivers for a coordinated national approach to ‘data collection, research and analysis…’ was identified by the Council of Australian Governments (COAG) review and documented in their report ‘Natural Disasters in Australia — Reforming mitigation, relief, and recovery arrangements’ in 2001 and released in August 2002. Representatives of all tiers of governments were signatories to this report. Later in 2001 the events in New York on September 11 reinforced the business drivers for access to data that transcends jurisdictional boundaries, as did the 2003 bushfires in Canberra. Against this backdrop there are several projects that are addressing the infrastructure and data requirements at the state/territory level. The ‘LIST’ in Tasmania. ‘VicMap’ in Victoria, the ‘EICU’ project in NSW, the ‘SIS’ project in Queensland, the ‘SLIP’ project in Western Australia and the ESA CAD system in the ACT are examples of spatial information Infrastructure initiatives that partially support EM at the jurisdictional level.

At the national level the Australian & New Zealand Land Information Council (ANZLIC) proposed a national Distributed Spatial Data Library in 2003. Previous attempts to create centralized repositories have failed but maturing web services and the ability to produce hard-copy maps on-demand have moved this concept to a practical reality. Underpinning the distributed library is the development of a community ‘All Hazards’ Data Taxonomy/Model for the EM community. The majority of the state jurisdictions provided input to the taxonomy, while additional expertise in the modeling and socio-economic domains were provided by Geoscience Australia (GA). The data identified by the taxonomy is sourced from varied and complex sources and formatted into a simplified, coherent form suitable for Emergency Management. The benefits of sharing data through a standardized framework are being progressively demonstrated to organizations through the ability to provide early warning of threats to their assets and services, while ensuring they maintain control of their data. There are still many hurdles to overcome before an infrastructure to support a Distributed Spatial Data Library can be realized. These hurdles can be broadly categorized as technological and cultural. The technological hurdles are no longer a significant barrier as bandwidth steadily increases, and major GIS systems support web service based data integration. It is arguably the cultural hurdles that are the most difficult. The process of consultation and review used in creating the ‘All Hazards’ taxonomy has created a realization among the jurisdictions of the benefits of closer ties and co-operation in data sharing and delivery arrangements. There is still some distance to travel but the implementation of an Australian Distributed Spatial Data Library for Emergency Management is moving closer to reality.

Dynamic chaining of geographic information services (geo-services) is emerging as a viable framework for evolving flexible geo-information systems, integrating heterogeneous geographically dispersed geoinformation systems, and for providing on-demand access to geographic information in many application domains and location-based services. Alongside functionality, quality of service (QoS) is basic to successful chaining of dispaprate geo-services. This paper explores QoS provisioning in the context of geo-service chaining for disaster management. The paper presents a QoS model for disaster management and illustrates how userlevel QoS requirements can be supported in a QoS-aware geo-service architecture.

This paper presents and discusses an approach for web-based assessment and decision support using multi-criteria evaluation methodology to be integrated into an interoperable service infrastructure. The research is imbedded in the EU-funded BALANCE project, which tries to assess possible impact and vulnerability to climate change. The results will be incorporated into an Assessment and Decision Support System (ADSS) for the arctic, which shall raise awareness among stakeholders in the BALANCE study area about possible climate change impacts on their environment and way of live and support decisions concerning possible adaptation and mitigation strategies. The paper elucidates the concept of geoprocessing and service chaining with the help of a use case in the field of reindeer herding.

Accuracy and relevance are properties that are often mentioned in context with usability and quality of data. The use of GIS and mobile GI services in spatial planning and management requires and data. For usage of spatial data in time critical situations, e. g. disaster management, this is especially true. Decisions must be made in short terms, within hours or minutes. Such situations do not leave any time to evaluate whether the data used for decision making are “accurate” or “relevant”. This evaluation has to be done before. The use of inaccurate or irrelevant data, however, can lead to decisions that are inadequate or in the worst case harmful.

Increasingly, spatial data are provided via the Internet, and through wireless connections. This is potentially a great chance for GI services in general, but especially for time critical, spatial decision making. An important question that arises is how accuracy and relevance of data can be evaluated, a difficult task taking into account that data come from various sources, are distributed over several servers, and provided by different data producers. Such an evaluation is important because users must have confidence in the data. They must be sure that they use information that is reliable and adequate in a given situation. This requires not only a syntactical control, but especially an understanding of the semantic content of information.

The paper pursues these thoughts by investigating methods and techniques to search for specific spatial data sets and simultaneously evaluating the of data. “Fitness for use” can be examined - to a certain extent - using metadata. But metadata have shortcomings as they describe a fixed number of properties not taking into account adequately data semantics. Ontologies - an approach propagated currently within the framework of the semantic web (Fensel 2004) - are aimed at the identification of data sets.

This paper will examine where uncertainty may lie in the knowledge discovery process through the use of case studies in disaster management, in turn leading to a discussion of what future action is required to address the uncertainty that may lie within knowledge obtained through these techniques. We describe our approach to address three types of issues: accuracy, efficiency, and usability. Typically, data mining techniques have higher false positive rates than traditional data exploratory methods, making them unusable in real-time systems. Also, these techniques tend to be inefficient (that is, computationally expensive) during the steps of a knowledge discovery process, particularly during training and evaluation. This prevents them from being able to process data and detect anomalies, hot-spots, or patterns in real-time applications. Finally, disaster management applications require large amount of training data and are significantly more complex than traditional GIS applications. These problems are inherent in developing and deploying any real-time data mining based system, and although there are trade-offs between these three groups of issues, each can generally be handled separately. The paper concludes by presenting the key design elements for supporting a real-time knowledge discovery process and group them into which general issues they address.

In Russia, a Unified System of Emergency Response (RSER) one of the parts of which is Russia’s FAAE Branch System of Emergency Response has been created.

To provide scientific and technical support for the tasks on evaluation of consequences for the environment and the population in case of crisis situation at NPP, in 1996 IBRAE RAS has established the Technical Support Centre (TSC) for the Crisis Centre (CC) of REA Concern. IBRAE RAS TSC is part of the system to render assistance to nuclear power plants in radiation hazardous situations.

Since 1999 a united Technical Crisis Center (TCC) has been operating at IBRAE RAS on a twenty-four hour basis to support the Situation Crisis Center of FAEA (former Minatom) and crisis centers of the Russian Ministry for Emergency Situations (EMERCOM) and Rosenergoatom Concern.

In recent years TCC experts have worked through a number of requests related to real incidents at nuclear-hazardous facilities throughout the world. Within the scope of everyday activity, the check-up of TSC workability is under way through exercises and trainings.

The report represents the examples of practical approbation for the activity of FAAE and CC of REA Concern in their interaction with TSC during the exercises and actual incidents.

The Centre for Remote Sensing and Geographic Information Services (CERSGIS), a self-supporting, no-profit organization within the University of Ghana is an implementing agency an European Union supported project for “Establishing Mapping and Monitoring System for Development Activities in Ghana (EMMSDAG)”. The current EDF-funding cycle of three years covers the “Rural Phase” of the EMMSDAG”. The goal of the Rural Phase of the EMMSDAG is an on-line multi-functional/multi-application ArcGIS-driven government-owned data sets on social infrastructure facilities for rural community development projects and impact monitoring and evaluation. Five frontline government institutions are the primary stakeholders of the EMMSDAG — Ministries of Finance and Economic Planning, Local Government and Rural Development, Project Monitoring and Evaluation Unit of the Office of the President, National Development Planning Commission and House of Parliament. These are to be connected on-line through wide area network for direct use of the data resources of the EMMSDAG Laboratory at the CERSGIS.

The project should facilitate widespread use (by government and nongovernmental organizations, donors, investors and the private sector in general) of the GIS-driven socio-economic databases to be sustainably maintained by CERSGIS.

It will be the primary task of CERSGIS to assist potential and real users in defining profitable/practical uses of the data for problem-searching/defining and problem-solving. The National Disaster Management Organization (NADMO) will be one of the expected users of the EMMSDAG’s data resources.

In the Sahel region the unfavourable climatic conditions and the natural resources degradation are recognised to be the main constraints affecting the agricultural productivity; nevertheless more complex and less evident causes of food insecurity of population groups need to be investigated.

The analysis of biophysics and socio-economic dynamics acting in disadvantage areas is considered a basic element to put in action effective programs for the prevention and management of food crisis. The availability of effective and timely information during the development of agricultural season is strategic for the evaluation of famine warning level: it constitutes a support to the planning activity of decision makers.

In this context a schedule for food crisis prediction has been designed as an operational framework enabling to integrate different tools to produce appropriate information to define and monitoring the risk zones during the whole agro-pastoral season. The spatial extension and the importance of the population touched can be timely estimated in order to define the action plans to minimize the crisis damage.

An analysis module, a plugin of the free Geographic Information System VisualCarte, has been developed aiming at increasing the Sahelian National Services’ responsibility in emergencies management and at encouraging their involvement in the long-term resources planning for crisis prevention. The plugin PRVS (Structural Vulnerability Mapping Procedures) enables the elaboration of thematic maps through indicators of vulnerability and the creation of scenarios about the current vulnerability.

A structural frame of reference for the identification of vulnerable zones and populations at regional and national level is made available: homogeneous areas and administrative units are classified according to theirs productive capacity related to the economic structure of productive system. As consequence the monitoring of negative events can include their evaluation according to the incidence in the context of structural vulnerability.

The plugin PRVS can process different data type. The use of real or estimated data allows to assess the impact of unfavourable conditions on the primary productivity during the rain fed season. The estimation of the capability of the primary sector to sustain a given population leads to the identification of self-sufficient, surplus or deficit areas.

By means of integrated analyses tools and seasonal forecast, the early assessment of current vulnerability may occur. The continuous availability of reliable information for the mapping of vulnerability satisfy the needs of the various institutions involved in crisis prevention and natural resource management.

The work described in this paper has been conducted by Ibimet - CNR in collaboration with Agrhymet Regional Center within the SVS (Suivi de la Vulnérabilité au Sahel) project funded by Italian Cooperation.

The program’s objective is the strengthening of regional and national capacities of CILSS countries (Permanent Interstate Committee for Drought Control in the Sahel) in the crisis prevention and in the poverty reduction. Methodologies for vulnerability analyses have been pointed out to support different users with appropriate information. Furthermore environmental monitoring for the assessment of zoning associated with climate change is also included in the objective of the program.

Afghanistan is located in a zone of high-seismic activity. Given the rugged and mountainous nature of the country and the location of villages, towns and cities, there is propensity for widespread death and destruction due to landslides whenever an earthquake occurs.

Use of satellite imagery by humanitarian agencies in Afghanistan in preparation for and response to natural and man-made disasters has been very limited, mostly to International organizations such as the United Nations. Earth Observation Satellites (EOS) due to their vantage position have demonstrated their ability to rapidly provide vital information and services in a disaster situation. EOS has been used in emergency situations where the ground resources are often lacking.

The perception amongst humanitarian agencies and civil protection authorities in most developing countries is that the cost of satellite imagery is not cheap. With limited budgets available for purchasing satellite data, they tend to opt for less expensive solutions such as interagency survey teams to assess damages. The rugged and mountainous nature of Afghanistan and the lack of roads in most parts of the country, survey teams are most often hampered, leading to delays in delivery of information from the field to the decision makers.

Recent earthquake in the Hindu Kush of the country in April 2004 witnessed the triggering of the International Charter for free delivery of satellite imagery.

Image analysis and interpretation of both pre and crisis data did not show observable features of damages. The damage assessment maps were used by the humanitarian community for decision-making.

Availability and access to space technology in addressing natural disasters have been the main obstacles facing developing countries particularly those poor countries without their own space programs. This problem has been solved through the introduction of The International Charter for major disasters. However, knowledge about the Charter is not common knowledge in most developing countries; Disaster Management Authorities, the Academic Institutions, humanitarian agencies and the affected communities have very little idea about the availability and access to free satellite imagery. There is need for a massive awareness campaign to educate decision makers about the International Charter and the potentials of using space technology in addressing problems relating to disaster management and the environment. The skills to process satellite imagery and integrate it with other GIS layers are lacking in most developing countries; there is need to embark on a massive capacity building exercise to ensure optimization of the benefits of the technology. The Charter needs to find innovative ways of quickly sending value added information products to disaster management authorities instead of relying on in-country skills in image processing.

This paper elaborates on the experiences gained working with images received from the International Charter, and the immense pressures from the humanitarian community for rapid delivery of information.