Spatial Management of Risks E-Book

• Herausgeber: John Wiley & Sons
• Kategorie: Wissenschaft und neue Technologien
• Sprache: Englisch

Seitenzahl: 381

Veröffentlichungsjahr: 2013

Beschreibung

Spatial analysis is an increasingly important tool for detecting and preventing numerous risk and crisis phenomena such as floods in a geographical area. This book concentrates on examples of prevention but also gives crisis control advice and practical case studies. Some chapters address urban applications in which vulnerabilities are concentrated in area; others address more rural areas with more scattered phenomena.

Leseprobe

Table of Contents

Introduction

Chapter 1. From Prevention to Risk Management: Use of GIS

1.1. Introduction

1.2. GIS and public security

1.3. Examples of applications for public security

1.4. Prospects for development

1.5. Conclusion

1.6. Bibliography

Chapter 2. Coupled Use of Spatial Analysis and Fuzzy Arithmetic: Assessing the Vulnerability of a Watershed to Phytosanitary Products

2.1. Introduction

2.2. Construction of the index

2.3. Implementation of fuzzy calculations

2.4. Application to the watershed of Vannetin: vulnerability to atrazine

2.5. Conclusion

2.6. Bibliography

Chapter 3. Agricultural Non-Point Source Pollution

3.1. Introduction

3.2. Mapping non-point source pollution phenomenon

3.3. Territorial database building rules

3.4. The data sources used

3.5. Pollution risk zoning

3.6. Risk zoning applications

3.7. Conclusion

3.8. Bibliography

Chapter 4. Cartographic Index and History of Road Sites that Face Natural Hazards in the Province of Turin

4.1. Introduction

4.2. Principal risks

4.3. Research area

4.4. Working method

4.5. Computer-based synthetic analysis and transcription of historical data and information collected on the research area

4.6. First results

4.7. Structure of computer thematic mapping

4.8. Application and use of the method

4.9. Bibliography

Chapter 5. Forest and Mountain Natural Risks: From Hazard Representation to Risk Zoning – The Example of Avalanches

5.1. Introduction

5.2. Identification of protective forest zones

5.3. Perspectives

5.4. The creation of green zones in risk prevention plans

5.5. Conclusion: general recommendations

5.6. Bibliography

Chapter 6. GIS and Modeling in Forest Fire Prevention

6.1. Understanding forest fire risks

6.2. Forest fire management: risk mapping and the use of spatial analysis

6.3. Using GIS to map forest fire risks

6.4. Conclusion

6.5. Bibliography

Chapter 7. Spatial Decision Support and Multi-Agent Systems: Application to Forest Fire Prevention and Control

7.1. Introduction

7.2. Natural risk prevention support and the need for cooperation between the software programs

7.3. Towards an intelligent software agent model to satisfy the cooperation between the decision-support systems dedicated to natural risk prevention

7.4. Experiment in the field of forest fire prevention and control

7.5. Conclusions and perspectives

7.6. Bibliography

Chapter 8. Flood Monitoring Systems

8.1. Introduction.

8.2. Flood monitoring and warning

8.3. Situation diversity

8.4. Technical answers

8.5. Conclusion

8.6. Bibliography

Chapter 9. Geography Applied to Mapping Flood-Sensitive Areas: A Methodological Approach

9.1. Introduction.

9.2. A geographic analysis of flooding

9.3. A concrete example

9.4. Bibliography

Chapter 10. Information Systems and Diked Areas: Examples at the National, Regional and Local Levels

10.1. Context.

10.2. Analysis of the current situation for the management of diked areas

10.3. Spatial dimension and integrated management of diked areas

10.4. Examples of information systems dedicated to diked areas

10.5. Recent progress and perspectives

10.6. Bibliography

Chapter 11. Geomatics and Urban Risk Management: Expected Advances

11.1. Towns, risks and geomatics

11.2. Prevention stakeholders: their responsibilities, their current resources and expectations

11.3. Today’s methods and tools: strengths and weaknesses

11.4. New potentialities using geomatic methods and tools

11.5. Some ongoing initiatives since the beginning of 2001.

11.6. Assessment and outlook: fundamental elements of future systems

11.7. Bibliography

List of Authors

Index

First published in France in 2001 by Hermes Science/Lavoisier entitled “Gestion spatiale des risques” First published in Great Britain and the United States in 2008 by ISTE Ltd and John Wiley & Sons, Inc.

Apart from any fair dealing for the purposes of research or private study, or criticism or review, as permitted under the Copyright, Designs and Patents Act 1988, this publication may only be reproduced, stored or transmitted, in any form or by any means, with the prior permission in writing of the publishers, or in the case of reprographic reproduction in accordance with the terms and licenses issued by the CLA. Enquiries concerning reproduction outside these terms should be sent to the publishers at the undermentioned address:

ISTE Ltd27-37 St George’s RoadLondon SW19 4EUUKJohn Wiley & Sons, Inc.111 River StreetHoboken, NJ 07030USAwww.iste.co.ukwww.wiley.com

© ISTE Ltd, 2008

© LAVOISIER, 2001

The rights of Gérard Brugnot to be identified as the author of this work have been asserted by him in accordance with the Copyright, Designs and Patents Act 1988.

Library of Congress Cataloging-in-Publication Data

[Gestion spatiale des risques English] Spatial management of risks / Edited by Gérard Brugnot.

p. cm.

Includes bibliographical references and index.

ISBN 978-1-84821-046-2

1. Human geography--Mathematical models. 2. Environmental degradation--Mathematical models. 3. Environmental degradation--Statistical methods. 4. Geographic information systems. I. Brugnot, Gérard. II. Title.

GF23.M35G4713 2008

363.3401’1--dc22

2008027556

British Library Cataloguing-in-Publication Data

A CIP record for this book is available from the British Library

ISBN: 978-1-84821-046-2

Introduction

Geographic Information, Land Use Planning and Risks1

Risks, a growing issue

As clearly asserted by the titles of two books written by the German sociologist, Ulrich Beck, we have entered into a risk society (1992), and more recently we could even say that we have entered into the world risk society (2000). Risk is omnipresent in our daily life. Now, the question is naturally raised whether we are living in a more “risky” society than ever before. This statement can be analyzed in two ways. On the political level, which we will not enter, risk acts as the social cement of a “society without enemies”. On a more prosaic level, regarding our daily life, it is now commonly asserted that risk consists of the combination of a hazard (sometimes called danger, threat, etc.) and vulnerability. This analysis needs to be more comprehensive to paint a more accurate picture of reality, but it gives us something to work with.

This definition raises many difficulties, for it seems only to apply satisfactorily to the situations in which a phenomenon, totally independent from human activity, could assault people or damage their goods. In fact, this is true but only in borderline cases, such as, for instance, natural hazards related to crustal motions. Generally, we are both agent and victim, which means that not only do we not protect ourselves sufficiently from phenomena posing risks, but we create them. If this contrast appears artificial, yet we can more satisfactorily attest that risks can often be explained, whatever the real cause may be, such as poor use of land planning. This poor planning does not, in this case, stem from ill will, but from a lack of knowledge of spatial phenomena and issues. The territory, and the society that exists within it, is bounded by risk and every risk is written in the land.

As a result, the application of spatial analysis to any type of risk remains limited. The choice to give very concrete examples of spatial analyses led us to consider only certain types of risks with strong spatial logic. Therefore, we have focused on natural hazards, while some other risks, though important on the socio-political agenda, play less of a part. For example, food safety and health risks do not lend themselves to spatial analysis, although we do believe that the relatively small number of such analyses carried out on these phenomena is due to some other reasons.

The contribution of spatial analysis to risk analysis and prevention

According to our previous definition, threatening phenomena and human stakes are both clearly spatialized. For this reason, it is easy to see why spatial analysis is an indispensable tool for those in charge of risk management.

Risk management in large communities makes spatial analysis particularly relevant, since a high level of vulnerability is to be found there, and most large European cities have the necessary geomatic tools. Nevertheless, one of the major problems in large urban concentrations is that, although vulnerable concentrated elements are well known, hazards may originate from outside the urban territory – for example, water-related risks, whether they have to do with the quality (pollution) or the quantity (floods).

The chapters in this book have been chosen to illustrate various situations. Phenomena generating risks are quite diverse. Even though natural hazards make up the largest proportion of such applications, we have tried to compare other factors. This is the reason why some chapters focus on applications and others on theories.

Moreover, the examples given not only refer to prevention, but also to crisis management and feedbacks. Some chapters present urban applications with very highly spatially concentrated vulnerability, while some others present rural applications with more diffuse vulnerability and possibly more diffuse phenomena. It is certainly in the latter kind of case, which involves slow-dynamic phenomena, that spatial applications, which increasingly turn to temporal factors, are hugely beneficial to society, since they can detect both dangerous and irreversible slow changes on large territories. In this case, we can assert that spatial analysis is a tool serving sustainable development.

All the contributions in this book share a common point: they are all presented from a risk representation perspective, and not only from a potentially dangerous phenomena perspective. In all cases, human stakes are weighed against these phenomena and, even if, in most examples, we do not (yet) have an integrated risk management system, we do have an information and decision support tool. There is no doubt that the future, thanks to the expected continuing advances in software and equipment, will see the development of more and more sophisticated spatio-temporal interoperable systems. The field of risk management will probably welcome these systems more than any other field, since it requires the manipulation of numerous spatio-temporal objects, so as to support more and more complex decisions.

Presentation of chapters

In Chapter 1, the author gives a comprehensive summary of GISs used in crisis management. The spectacular evolution of problem management environments over the last 15 years is illustrated with the example of forest fire management performed by civil protection. The example of forest fire is particularly relevant to emphasize the obvious importance of spatial tools supporting risk management. Indeed, this natural hazard is very sensitive both in a temporal (the effect of a bucket of water after a minute of combustion is equivalent to the considerable means deployed an hour later) and spatial sense (not only for the management of preventive measures on vegetation and access, but also for pre-positioning of fire fighting and the conduct of fire-related operations). Two other examples are presented: one deals with the transportation of dangerous substances, while the other is about crisis management. In the first example, we discover a very concrete application, which takes special care to describe the notion of vulnerability. The second example introduces a very generic application that requires efficient telecommunications management. It enables the real-time acquisition of data on incidents and the issuing of the instructions necessary for implementing corrective actions.

Chapter 2 is dedicated to even more anthropogenic hazards, that is to say pollution risks generated by plant protection products. This type of pollution is widespread and related to agricultural practices that the so-called reasoned agriculture is willing to minimize. Yet, without further advances to improve water quality, it is necessary to implement and manage health information. To do so, the authors suggest the use of overall quality indexes to identify pollution levels in the logic of spatial representation. This index combines the determinants of pollutants leaching to ground water aquifers and waterways; these factors characterize the contaminants, the types of soil and rain. An original element of this contribution is the use of fuzzy numbers to list the results and reveal the inaccuracies related to spatial representation in general, especially when the purpose of the indicators is more to reflect the variation of phenomena in a space, rather than to represent them with precise physical parameters at each point. An example is given to illustrate the method and to test management actions aimed at controlling water pollution from atrazine.

In Chapter 3 we remain in the field of risks related to farming practices, for which the implementation of a space observatory is proposed, so as to monitor water pollution, in all its forms (pesticides, fertilizes, solid objects), as well as soil erosion. The authors’ approach rests on what they call process mapping, which corresponds to conceptual modeling. Their ambitious project led them to build a very comprehensive spatial database, consisting of elements related to topography, vegetation cover, structures (ditches, hedges, etc.) and to ground conditions. A risk/vulnerability analysis emphasizes the most exposed areas and proposes, as in the previous chapter, complementary management actions to improve the situation.

Chapter 4 was written in Italy, more precisely in the Piedmont region, and we would like to thank the authors who made the effort to write in French, for this book was first published in French. This chapter is an introduction to natural hazards and, in particular, to extremely severe events of nature. North-west Italy was hit very hard in October 2000, to such an extent that it led to the development of a spatial information and representation system. It lists a certain number of natural events characteristic of mountain zones near the Mediterranean Sea, and which are poorly defined by the French classifications. These phenomena correspond to flooding, landslides and torrential runoffs (formation of lavas). They are caused by heavy and long-lasting rain in geologically unstable areas, which generate several runoffs that sometimes stay away from thalwegs and carry huge amounts of solid objects, which can entail deposits exceeding several meters in thickness. Chapter 4 shows how these phenomena are inventoried through a specific survey, and then processed in a GIS, which in turn provides numerous information layers, among which the most prominent is related to the road network, assessing how vulnerable it is to these hydrological and geological phenomena through a list of accounted damage.

Chapter 5 also deals with mountain areas, albeit more peaceful mountain areas, with colder but less excessive climate conditions: the Northern Alps of France. In this area, the forest is a real protection structure that can be considered as ecological, because it is not natural, and results rather from an intensive gardening of the slopes, sometimes very steep, and dating back to very ancient times. The authors describe a very sophisticated multilayer spatial analysis system that makes it possible to emphasize the interactions between the forest and the various events disturbing it, and against which it provides protection: avalanches, rockfalls and landslides. This Geographic Information System highlights the weak areas in the forest ecosystem, where the slightest mistake, the slightest delay in terms of intervention could make whole areas at the foot of slopes unsuitable for building purposes. This type of concern explains the reason why this chapter was written by a researcher and a practitioner, who developed a method that can be used and is operational to draw up risk prevention plans (plans de prévention des risques, PPR).

As in Chapter 1, Chapter 6 presents an application for forest fire management. It is also similar to Chapter 5, in the sense that it focuses on prevention via natural habitat management. Naturally-caused forest fires are often contested, because the majority of fires are caused by human activities, whether intentional or unintentional. The authors analyze the constraints related to this type of situation in terms of risk definition: the forest, but also humans are both risk creators and victims. Natural habitats are strongly affected by this phenomenon, which is not, ecologically speaking, completely negative. Moreover, forest fire being a physical phenomenon, its propagation suffers from greater uncertainties than rockfalls or avalanches influenced by slope inclination or even rivers running down their beds. All these circumstances make forest fire risk zone mapping very delicate. This explains why there are very few “forest fire” PPR. The authors propose to develop an interesting hazard mapping support system for the Massif des Maures, based on physical characteristics such as wind, slope or vegetation, to assess fire risks and fighting conditions. The application is presented in a very educational way, and comments and illustrations are provided for all the development phases of the spatial information system.

Chapter 7 also deals with forest fires and confirms the fact that this phenomenon is particularly relevant, due to its complexity and numerous feedbacks, to test the most sophisticated spatial analysis systems. The author thus proposes a very ambitious and very generic approach to spatial and temporal multi-agent risk management that integrates some decision support aspects in situations of uncertainty. He gives concrete examples of wind intensity changes, and especially of wind direction that can greatly endanger the resources deployed in the field. This type of management, which is highly decentralized in a multi-agent context, gives the author the opportunity to present distinctive theoretical results from a multi agent system. ISA are neither firemen nor a new kind of forest firefighters, but intelligent software agents exchanging information and coordinating their actions. The author gives a concrete example of crisis management to illustrate how such tools could foster theoretical developments that are not discussed in this volume, which is dedicated to the presentation of applications.

Chapters 8 and 9 describe applications used in the case of a specific phenomenon that no region of our country is immune from, even if it takes different forms according to geographic location (climate): floods. Brittany, Aude, Somme, Meuse and Var are among the most recently disaster-struck and/or susceptible regions, which does not mean that the next flood will necessarily occur in one of these specific locations of which, among others, the Loire and the Seine are not included.

As is clearly explained in Chapter 8, flood hazard management, and especially flood hazard warning largely depends on the size and slope of watersheds. Entering the geographic information field with great care, and staying away from debate among hydrologists, and even farther from political considerations regarding land planning, such analyses should enable us to define flood control measures that could be implemented to the entire French territory, and especially, to stay in the realm of GI, lead to the development of spatio-temporal information systems adaptable to local climatic and geomorphic conditions. The system presented by the authors in this chapter is used to manage the watersheds that drain into the Garonne, for though they are large, they are vulnerable to heavy rainfall. The authors also describe the meteo-hydrological forecasting chain, as well as the spatial tools supporting crisis managers. As in the previous chapter, we focus on short-range forecasting (nowcasting). Unfortunately, a disciplinary and administrative barrier between hydrologists and meteorologists has limited the advances necessary to reach the level of the application dedicated to forest fires presented in the previous chapter.

Chapter 9 is less ambitious, in the sense that it only targets the representation of historical floods. Yet, this inventory is very topical since we are in a field where spatial analysis uses both proven tools and large surfaces of buildable or already built zones. It concretely illustrates the risk issue, the assessment of which is based on a study of the phenomena that must be extremely accurate due to the economic stakes involved, as well as relevant when delivering results. To illustrate this, the author not only provides an inventory of the questions raised and the methods used in flood mapping, which is very valuable, but also an example relating to the Garonne river.

Chapter 10 is also dedicated to flooding, but its approach is very different from those used in the previous chapters. It describes a comprehensive project with ambitious plans to inventory and diagnose river dikes over the whole national territory. Above all this, this chapter is particularly fundamental in this volume because it provides an example of a major spatial system that integrates all the characteristics of a comprehensive public decision-support system. The average time for such projects is 10 years, and the proportion of resources necessary to carry it out is similar. The genesis of the application (the Camargue flooding and the concerns with the Loire embankments) is interesting, because it is based on the Government’s willingness to find a long-term solution to this problem, and because it conducted a thorough analysis to identify the needs of a multi-scale spatial information system in nature, according to the variety of the objects involved. In the end, this system integrates the notions of hazard and vulnerability, from the most concrete and accurate geotechnical aspects related to dikes (e.g. rabbit burrows) to the most realistic scenarios of vulnerability, such as what if (e.g. what would happen if such a dike, which had received a diagnosis of weakness, finally breaks)

Chapter 11 concludes a volume essentially devoted to natural risks, or at least risks related to vast territories of low-density occupation, with an overview of spatial information systems dedicated to urban risks. This chapter is presented in a course format, which completes the volume by addressing spatial risk issues in a conceptually clear manner, by discussing alternatively application questions and examples, which will enable readers to shed new light on some developments already presented in the previous chapters. The author provides many different examples, including space risk management systems developed by the Urban Community of Lyon, which are, with those developed in Marseille, the most ambitious of their kind. He sets all the tools used in a public political context, which concludes the volume with an emphasis on the social and political nature of risk, as expressed at the beginning of our introduction.

Conclusion

Risk analysis involves a fundamental spatial component; there is no need to demonstrate this point again. The chapters of this volume illustrate the possible uses of spatial analysis tools. Without some of these tools, many delicate issues relating to land planning would be impossible to manage at the political level.

Some may be surprised from the above statements that our conclusion is actively pessimistic. Viewed more broadly, spatial risk analysis appears to be poorly developed in France. It is scarce in numerous fields, and a little more developed with respect to country-related risks, due to the agro-rural tradition of our society that some bodies, sometimes academic bodies, have acquired.

Nevertheless, we are still unable, for instance, to overlay natural hazard-related information layers, such as floods, with other information layers illustrating land use in urban and peri-urban environments. Moreover, information on flood damage is managed independently and its spatialization is not on the agenda, at least for now. Therefore, we are still unable to integrate the drainage system to a digital elevation model.

Many examples could be given to demonstrate how important it is for major managers of spatial databases, without whom applications would only remain academic monographs or systems of local interest, to provide quality and economic research products, such as topographic, land use, physical or economic databases. Some areas of study are still wide open, such as the creation of areal postal codes as in the UK, and the georeferencing of vulnerable components.

These issues can only be addressed with political support. They are a fundamental ingredient to the development of interoperated land use management systems, without which no risk integrated management is possible; only partial management, often implemented in catastrophic events, which can lead to disappointing results, let alone negative results.

1 Written by Gérard BRUGNOT.

Chapter 1

From Prevention to Risk Management: Use of GIS1

1.1. Introduction

Territory mapping has always been of paramount importance for society [IGN 90].

Since ancient times, maps have had a functional role:

– “route” maps, in ancient Rome and the Middle Ages, such as the Tabula Peutingeriana;

– commercial maps during the 15th century and the long voyages around the world;

– military maps, of which the most significant development occurred at the instigation of Napoleon.

It was at the beginning of the 19th century that Napoleon formalized the fact that knowing the terrain was a necessary condition for victory. He created the 1:80,000-scale ordnance survey maps produced by the military services. They were high-precision maps providing detailed information on relief, remote communities, bridges, vegetation, etc. Moreover, such maps enhanced the necessity for regular updates.

During the two World Wars, maps progressively became an obvious decision-making support tool for crisis management:

– road maps appeared with the transport revolution, but their use was adapted to the needs of World War I, that is, to follow the evolution of the Front with nearly real-time updates;

– the French National Geographic Institute (IGN) was created in 1940, and replaced the Army Geographic Service that had been dismantled by the Germans;

– Michelin provided the French, English and American armies with maps to drive their troops.

Some of the working conditions of firemen are similar to the context of conflict, and this is why they have always paid great attention to prior knowledge of the terrain. Maps have always been critical for any type of response (emergency relief to people, flooding, accidents on transportation linkages, etc.). However, they are mainly used to locate an event, to dispatch the resources, to know about the crisis area and emergency plans (prevention, aid). When responding to a disaster or an accident, this knowledge is determinant in order to take the right and most appropriate decisions given situational factors. The time to plan a response is limited to the few tens of seconds between the moment the call is received and the movement of the emergency team.

In the case of toxic gas dispersion, for instance, it is essential to know the environment in order to take action, such as the confinement or evacuation of people.

In March 2000, in Saint-Galmier (Loire), a train hauling highly toxic substances derailed, thus releasing a gas cloud. The operational analysis carried out just after the event revealed that, among the elements that had supported the decision-making process for the rescue of people, accurate knowledge of land use had been fundamental [GRI 00]. In such contexts, the most comprehensive and synthetic tool to picture land use is the map.

The use of “conventional” topographic maps, which was dominant for a long time, progressively turned to “profession” maps targeting specific issues. The need for “profession” maps produced for a particular theme increased more and more:

– maps dedicated to urban public security and defense management [CHE 00];

– maps to prevent and fight forest fires [JAP 00];

– maps for the management of dangerous goods transportation-related accidents [GLA 97].

Nowadays, the most effective tool to answer these needs is a Geographic Information System (GIS). The evolution of its use over time will be discussed using examples of existing applications.

The features related to the complex issue of updating data will not be dealt with in this chapter.

1.2. GIS and public security

Within their respective sphere of competence, the French Fire and Rescue Department Services cover the following missions: public security risk prevention and assessment, planning safeguards and implementing emergency measures, life, property and environment protection, emergency assistance to people who have suffered an accident, damage or a disaster as well as their evacuation [SNO 00]. These missions are grouped into three themes:

– prevention: gathering the measures implemented to prevent a disaster occurring again or becoming worse;

– forecasting: to know and forecast the initial conditions and evolution of a disaster;

– operations: the implementation of disaster control measures.

Three main reasons account for the increasing importance of GISs in the execution of public security plans:

– GISs are involved in each of the missions mentioned above;

– the professional profile of those using the GIS tool;

– the role of GISs in decision-making processes in crises.

In the field of forest fires, in which mapping is a fundamental tool, the missions of the French Fire and Rescue Department Services are characterized by [DSC 94]:

– forest fire prevention or protection (DFCI), which includes, among others, forest massif management, monitoring (patrols and fire towers) and public outreach;

– forecasting, aiming at assessing local risks of forest fire outbreaks and spread, based on meteorological and vegetation condition parameters;

– fighting, which consists of coordinating land and air resources to stop the fire from spreading and to extinguish it.

During these missions, firemen make considerable use of mapping. Indeed, it is quite impossible to manage disaster control measures without information on the surface topography, road transportation systems, populated areas, etc. Maps are a privileged tool at the heart of decision-making processes.

At the beginning of the 1970s, a reform of the forest fire control mechanisms was launched in the South-West of France. This reform entailed several consequences such as the creation of new agencies aiming at implementing actions to protect forests from fire, the creation of a statistical database on forest fires and forest fire control structures, such as tracks, water points, forest towers, etc. [KER 99].

In 1987, DFCI maps appeared, that is, maps specifically produced for forest fires control and prevention practitioners. Provided by the IGN, these maps were based on 1:25,000 and 1:100,000-scale topographic maps [RON 87]. They display, in superimposition, the specific coordinate system (bikilometric DFCI grid) and all of the DFCI structures.

Figure 1.1.Initial organizational chart of DFCI map production

Figure 1.1 presents the organization in 1987. There were, however, some disadvantages to these maps [SAU 97]:

– the cost: the IGN spent hundreds of thousands of Euros to produce 1:25,000 and 1:100,000-scale DFCI maps just for medium-sized departments;

– information update: a year after the production, the maps were no longer operationally usable.

The development of GISs was then mainly limited to the research sphere [DID 90], and their market reflected “their youth by its instability and lack of maturity” [POR 92].

In 1992, a first attempt to implement a geographic information system dedicated to public security was made in the French Mediterranean zone. The purpose was to deploy an operational coordination information system for public security integrating messaging capabilities, databases, mapping and decision-support [MAR 93]. Yet, the lack of geographic digital databases (both in terms of costs and of geographic coverage of the Mediterranean zone) led to the suspension of the mapping dimension of this tool, and consequently of the use of a GIS.

The interest in using a mapping information tool was renewed in 1995. At the national level, this date also seems to be an important step in the use of GISs by firemen [SDI 00].

In the French Mediterranean zone, this was illustrated by the SIGASC application project (GIS applied to public security). The objectives of this application emphasized GIS functionalities so as to achieve several goals. Indeed, SIGASC must [SAU 97]:

– produce up-to-date paper maps;

– provide a constant knowledge of the DFCI structures across a specific area;

– manage and plan forest massifs to help control forest fires.

A transfer of expertise regarding paper maps can be observed, from the “conventional” producers to the users (see Figure 1.2).

End users (firemen) seek to develop the necessary skills to manage their own mapping production. Moreover, the routine use of GISs introduces users to more complex functions, which creates new needs: data processing, spatial analysis, quantitative analysis, geographic database management, etc.

Figure 1.2.Introduction of GISs aimed at firemen

A major step was taken with the introduction of the automated processing of geographic data: geomatics.

GISs were then actually used to gather, store and manipulate heterogenous data that, once they were made coherent, could be restored in various forms: reference maps, thematic maps, reviews and tables (see Figure 1.3).

Figure 1.3.Use of GISs

The developments and use of GISs continued in two major areas [SDI 00]:

– functional developments: the use of spatial analysis capabilities, the production of new information, the use of technologies producing geographic information (remote sensing, GPS);

– developments in the issues addressed: natural and technological risks, radiological risks, common risks.

GISs also became support tools for the retrieval of simulated processing. These functionalities are especially related to the following areas:

– forest fires [SAU 98],

– technological risks [DUS 97],

– radiological risks [PRE 00],

– floods [COR 99].

Their use remained especially focused on prevention and prediction for, even though the functions gained in complexity, the core purpose still remained the production of maps.

Figure 1.4.Today’s telegeomatics in fire stations

It is only recently that this purpose has evolved [SDI 00]. Today, GIS is at the core of an increasingly complex organization (see Figure 1.4).

This organization makes it possible not only to produce thematic maps from a variety of sources, but also factual maps [FOR 98], which will be progressively introduced in operational areas responsible for risk management.

Today, GISs have become tools processing information to achieve an immediate objective, such as maps on demand, mobile tracking, etc. They are more and more involved into decision-making processes in emergency situations, for they provide the required information with almost real-time refresh rate [GAL 96, SAU 00].

Today, there are a growing number of GIS applications integrating the principles of telegeomatics [OLI 99]: the communications between operational areas and command posts in situ are essentially cartographic in nature: the resources implemented in the field include geographic information survey tools (GIS, GPS, cameras, etc.) allowing the edition of maps “on demand” to track an event [BOU 01]. These maps are then transmitted to the command post in the field to plan or modify fighting tactics, or to the operational area to anticipate the actual requirements in terms of resources.

Yet, despite their increasing use in public security, GISs are still basically used by in situ commanders to acquire and manage as much information as possible to make the best decision they can.

GISs, and more particularly the maps they produce, are information-sharing tools fundamental to decision-making.

The strength of GISs is related to the fact that the volume of information, its level of synthesis and the typology of the information on the map have greatly increased.

1.3. Examples of applications for public security

The role of GISs within departments responsible for public security is illustrated in the three following examples of existing applications.

1.3.1. SIGASC application

Formalized in 1995, the objectives of the SIGASC application were determined by the necessity of supporting forest fire management and prevention [SAU 98].

The main purpose was to provide the 15 Departmental Fire and Rescue Services of the south defense zone (Languedoc-Roussillon, Provence-Alpes-Côte d’Azur, Corsica, Drôme, Ardèche) and the public security and defense top managers of the south zone with a GIS-based tool and methodologies applied to public security practitioners of the French Mediterranean zone, and mainly to forest fires.

Defined in collaboration with the users, this application always provides an updated vision of the field. The large geographic coverage (the 15 departments of the Mediterranean front) requires homogenous information across the research area in terms of content and cartographic representation.

Such a work required establishing a certain number of working groups, consisting of users and GIS specialists. Indeed, the Departmental Fire and Rescue Services are financially independent, and consequently, they could not all afford the purchase of the necessary data or the services of a specialist to carry out this work [SAU 00].

Figure 1.5.General structure of the SIGASC application

The work was organized through thematic groups (application architecture, financial negotiations with the IGN, acquisition of specific data, cartographic production, etc.), and thus, all the Departmental Fire and Rescue Services could benefit from the results.

In order to meet the intermediate objectives presented in Figure 1.5, several steps were achieved:

– to define precisely the computer support (hardware and software) according to the needs;

– to identify a common cartographic reference system that would address both the information and visualization expectations;

– to identify as accurately as possible the availability of specific mapping, and to assess discrepancies;

– to standardize the definition and the representation system of specific geographic information;

– to provide the technological and methodological resources to have constantly updated maps and to be able to edit an annual departmental atlas.

To fit the departmental and zonal use, the common cartographic reference system was chosen on a small or mid-scale.

The following choices were made:

– BD CARTO® of the IGN (digital vector products that include all the information present on 1:100,000-scale maps, providing decametric precision), because of its availability across the research zone and its adaptability to the research field;

– Scan25® and Scan100® products from the IGN (raster digital products resulting from the scanning of paper maps at different scales, from 1:25,000 to 1:250,000, to be used exclusively as base maps), so as to keep what we already had, as well as the comfort while reading maps at both scales.

With the appearance of DFCI maps in 1987, the use of specific maps to serve public security was assigned differently within each department, and soon, discrepancies came to light. An overview of these discrepancies is presented in Figure 1.6.

These results led to the writing of a standards guide [DPF 97] to lay down the fact that: “every piece of field equipment used by the DFCI corresponds to a specific standards category that enables its symbolization and production on maps”.

The geographic information specific to public security services being precisely defined, the technology and methodology dedicated to the acquisition of these data were identified in their turn.

The problem relating to data acquisition is a major issue. There is a lot at stake, the reason for producing maps for public security being twofold [SAU 98]:

– to provide a comprehensive knowledge of the field, so as to optimize decision-making at the different levels of operational command;

– to ensure, with a minimum amount of risk, the veracity and relevancy of the elements represented.

With respect to specific data acquisition, a brief comparative study was carried out between the traditional methodology for surveying geographic information (compass and decameter) and the methodology using GPS [SAU 00]. The results emphasize a factor of 10 between these two methodologies regarding the time for survey and mapping transfer.

Figure 1.6.Examples of symbols used before the standardization

The SIGASC application was implemented in 1997 in some departmental fire and rescue services.

In the Department of Gard, a protocol was established between the National Forest Office, the Departmental Fire and Rescue Service, the Agriculture and Forestry Departmental Directorate and the General Council. It aims at creating a common pool of DFCI data, for which the processes of initial GPS acquisition, of management and processing, of update and mapping are jointly carried out by the four signatory organizations. The target here is the homogenity of DFCI data across the department, the constant updating of DFCI 1:25,000-scale mapping and a cost-effective production.

A structure (DFCI GIS cell) and a specific vehicle with a GIS and a GPS on board (DFCI four-wheel drive liaison vehicle) were implemented. The DFCI GIS cell consists of a joint-team of some staff from the National Forest Office and firemen. Their objective is to perform a GPS-based survey of all the DFCI structures to map them and characterize them.

The GPS-surveyed data are then sent to the administrator, the DDAF (Agriculture and Forestry Departmental Directorate), who structures them and integrates them to the departmental database. The resulting DFCI database thus conforms in all respects to the requirements of the standards guide. This database is then transferred to the signatory organizations of the convention.

In summer, the systematic GPS-based survey of DFCI structures stops. The DFCI four-wheel drive liaison vehicle is then mobilized for forest fires to map in real-time the starting point and successive contours. The maps produced can be printed in situ. This also contributes to the development of a cartographic database for forest fire annual reports.

The SIGASC project gave birth to many others: working groups are carrying out more research into the use of GIS for public security, with additional technologies (aeronautical application of GISs) and other risks [MIS 00].

1.3.2. Application

The SIGRISK project (GISs related to the risk of transporting dangerous substances applied to public security) is part of the implementation of an operational tool to support decision-making in times of crisis. This tool meets the needs related to public security preparedness in the face of accidents resulting from the transportation of dangerous substances.

Dangerous substances transportation risk is characterized by random occurrence both in space and time. This specific risk presents two major categories of uncertainty related to risk assessment and quantification, and to environmental variability [DUS 97, GRI 99].

Among the specificities of typical accidents resulting from the transportation of dangerous substances [LAG 95], we find:

– kinetics, which varies a lot according to the type of transportation, the type of goods and the type of accident;

– the necessity to understand very quickly the environment (human, physical, natural) where the accident occurs.

Within the very first minutes following an accident resulting from the transportation of dangerous goods, it is of an absolute necessity:

– to know, even generally, about the kinetics of the accident, and its possible spreading, for instance how a toxic gas cloud might disperse according to weather conditions [FUL 96];

– to have as much information as possible on the population, the potential presence of public assembly buildings, of industrial sites at risk, of drinking water installations, nature of the surface, of the subsurface, of the road network, etc. [GRI 00], so as to assess as soon as possible the direct risks as well as the potential indirect impacts.

Consequently, the objective is to design a GIS-based computer tool enabling us to:

– quickly assess the consequences of a dangerous goods transportation accident via effect distance calculations;

– identify and quantify the vulnerability of the area impacted from a human, material and environmental point of view; vulnerability levels previously determined and translated into a map.

These objectives include some intermediary steps described in Figure 1.7.

Assessing the consequences of a dangerous goods transportation accident is achieved using OSIRIS. This software was developed in collaboration with the Ecole des Mines d’Alès and firemen from the Department of Gard. OSIRIS is “capable of providing information relating to the safety of individuals during dangerous goods transportation accidents” [OSI 00]. It is a crisis management training software program that informs, with baseline figures (on meteorology, type and quantity of spilled material, etc.), on the consequences (rate of flow, of evaporation) and effect distances for various types of accidents: explosions, toxic gas dispersion, and hydrocarbon fires [DUS 97].

OSIRIS is coupled with a GIS software to transfer onto a map the effect distances previously calculated. The spatial analysis capabilities of the GIS are then used to emphasize the information relevant to crisis management.

With the GISRISK application, the effect distances of sample accidents are cross-referenced with a vulnerability map [GRI 01].

The term map refers to the need for digital geographic information. The term vulnerability refers to land use knowledge.

The production of a vulnerability map consists of several steps.

The digital mapping of land use is the first to be edited from [GRI 01]:

– SPOT satellite images providing a land use overview adapted to the selected issue. This map is based on a nomenclature derived from CORINE land cover (www.ifen.fr) to emphasize the important elements to the assessment of vulnerability;

– other data sources (urban databases, National Institute for Statistics and Economic Studies, etc.) add more accuracy to specific points previously edited, for instance, the localization of public assembly buildings, schools, etc.

Figure 1.7.GISRISK application architecture

At that point, the land use map contains all the elements necessary to the assessment of human, material and environmental vulnerabilities, with respect to dangerous goods transportation risks.

The next step, which consists of the production of the vulnerability map, requires the development of a method to identify the potential targets, their sensitivity to a certain type of accident and their degree of exposure to potential hazards.

Among the information edited on the land use map, human targets (dense or dispersed residential developments, healthcare facilities, etc.), environmental targets (intensive agriculture, wetland, etc.) and material targets (water installations, commercial stocks, etc.) are identified.

Subsequently, these targets are characterized according to:

– the potential effects of accidents to which they are sensitive (thermic, toxic effects for instance);

– the foreseeable consequences (difficulty to evacuate people, financial depreciation, etc.), possible indirect consequences (psychomedia impacts, daily life impacts, etc.).

Then, all these parameters are structured in a hierarchical format using a multicriteria decision-making method [SAA 80]. This step aims at drawing attention to the targets showing some kind of sensitivity, for all endpoints, rating high in terms security of people, property and environment.

The results are finally edited on a map using five categories (which are the five legend keys). This final vulnerability map will support preparedness and response.

Tackling extreme situations, this tool provides the public security services with the capacity for quicker access to specifically processed information.

1.3.3. SIG CODIS application

The third application presented here is an illustration of the current use of GISs for crisis management.

Indeed, it is the increased knowledge of GIS capabilities that led to the implementation of this application in operational areas [SDI 00].

This tool was designed to address a certain number of constraints entailed by the operational management of alerts, whatever the type of accident (forest fire, emergency relief to people, road accident, etc.):

– localization of the event,

– to send the rescue resources to the scene of the accident,

– knowledge of the environment of the site,

– to anticipate the possible increase of the emergency requirements, by knowing the evolution of the event.

The purpose is to provide a tool that improves response time. The users are those in charge of processing alerts and of the operational management of the event.

The application is not dedicated to one type of risk, but makes it possible to manage any operational contexts.

Developed on the basis of the consultation version of a GIS software program, the first version of the SIG CODIS application was very simple. This was to prevent any corruption of the databases due to misuse.

Initially, the application provided users with a reference database consisting of BD CARTO®, SCAN 25® and SCAN 100® from the IGN, as well as with rapid locating functions, using information such as the name of the district, the road number, etc.

Today, this tool has evolved, and its current design is more complex (see Figure 1.8).