Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Coastal flood risk: improving operational response, a case study on the municipality of Leucate, Languedoc, France

Coastal flood risk: improving operational response, a case study on the municipality of Leucate,... Background: Climate change and rising sea level will certainly lead to significant changes in the management of low-lying coastal areas in the coming decades. While the most recent studies in the field of coastal storms-related flooding are increasingly integrated, simultaneously addressing hazards and vulnerability, as well as population risk perception, there is still little work to consider the preparedness of stakeholders to manage crises whose frequency and intensity are likely to increase in the next years. Methods: The aim of this paper is to expose the major results of the CRISSIS research program, which proposed a multi-disciplinary approach to the management of coastal flood risk in a town particularly exposed on the French Mediterranean coast. The originality of the project was to offer both an integrated approach to risk by analysing its 3 dimensions (hazard, impact and vulnerability, and representations and perceptions held by populations and stakeholders, or « risk culture ») and a very operational section focused on the evaluation of crisis management measures led by local stakeholders. To achieve this objective, two crisis exercises were organized, the first one to test the assimilation by the municipality staff of existing crisis management procedures and the second to allow local actors to integrate in their crisis management procedures the new risk knowledge data issued from research conducted under the program. Results: The project has had three main features; (i) it accurately mapped the submersible areas that present a critical vulnerability, both material and human; (ii) it revelated the poor social representation of marine submersion risk, as well as the obvious lack of awareness of crisis management systems and tools and the behavior to adopt in the event of flooding; (iii) Finally, it highlighted the need, through the crisis exercises, for a better assimilation by the municipality staff of the crisis management procedures defined in the Municipal Rescue Plan. Conclusion: The CRISSIS project has demonstrated the usefulness of an integrated and operational approach of coastal flood risk, not only in terms of studying hazards, stakes and vulnerability, but also in terms of crisis management, in particular through the organisation of crisis simulation exercises. Keywords: Coastal flood risk, Numerical modeling, Vulnerability, Risk perception, Crisis exercise, Leucate * Correspondence: brice.anselme@univ-paris1.fr Université Paris 1 Panthéon-Sorbonne, 191 rue Saint Jacques, F-75005 Paris, France Pôle de Recherche pour l’Organisation et la Diffusion de l’Information Géographique, UMR 8586 2 rue Valette, F-75005 Paris, France Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. Durand et al. Geoenvironmental Disasters (2018) 5:19 Page 2 of 17 Background of the program, with the objective of testing the PCS again With global warming causing a rise of sea levels, the but this time by integrating new data produced as part of vulnerability of coastal towns to flooding risk during the program’s findings. Overall, the objective of CRISSIS storms has increased. This issue is bound to become a has been to both generate new knowledge to improve risk key element in the management of low coastlines in de- prevention and help optimize crisis management by local cades to come, and multiple studies have been pub- stakeholders by facilitating the prioritization and classifi- lished on this subject in recent years (Cariolet et al. cation of actions in the event of a flood. After a brief intro- 2012. Vinet et al., 2012; Hurlimann and al., 2014; Duan duction of its study area, this paper will describe the et al. 2014 and Duan et al. 2015;King and al., 2014; methodology that guided its operational approach to Rulleau et al., 2015). While earlier studies were often coastal flood risk, before presenting its key findings. relatively segmented, focusing either on the hazard itself or on impact and vulnerability, more recent research pro- Study area grams on the topic have been much more integrated, sim- Theareachosenfor this studyisthe municipality of ultaneously addressing hazards and vulnerability, as well Leucate, a town located on the French coastline of the as the population’sperceptionofrisk – see for instance Languedoc region. The municipality comprises of four the French programs MISEEVA (2008–2011), JOHANNA seaside resorts located on both sides of the Leucate (2010–2013), BARCASUB (2010–2013), COCORISCO cape. These include from South to North: the resort of (2011–2014), and European studies on the management Port Leucate, built in the seventies as part of the Racine of marine flooding and submersion risk (Samuels et al. development scheme, on the coastal strip that cuts 2008;Sorensen et al. 2017). across the Salses-Leucate lake; the nudist village located The two-year CRISSIS program (Characterizing on the narrowest part of the strip; Leucate Plage, at the Submersion Risks in Sensitive Sites), launched in 2015 foot of the cape’s southernmost cliff; and La Franqui, at in a coastal town of the French region of Languedoc the foot of the cape’s northern cliff. The area is affected (municipality of Leucate), falls under this last category: by large-scale urban and touristic issues, as the town is through a multi-disciplinary approach associating ge- densely populated during the summer season, growing ographers, modelers, GIS specialists and crisis and risk from 5000 year-round residents to almost 100,000 dur- management experts, the project offers an integrated ing the high season. The urban areas of Port Leucate perspective on coastal flooding by exploring this risk’s and Leucate Plage are also very exposed to coastal flood three dimensions in coastal areas (hazard; impact and vul- hazards, and in particular Port Leucate which was built nerability; and representations and perceptions held by on a strip of low altitude coastal land (for the most part populations and stakeholders, or « risk culture »). This under 3 m NGF). Both areas are periodically flooded study does however stand out from other literature, in that when strong south-eastern storms occur (Ullmann it also includes a very operational section focusing on the 2009), during which the waves can cause the sea level evaluation of crisis management measures led by local to rise to up to 3 m NGF (Anselme et al. 2011). The stakeholders – an aspect that has rarely considered by floods can also be aggravated by a combination of research programs so far. wind and heavy rainfall, causing the lake’slevel to rise The preparedness of stakeholders to manage crises and provoking major runoff down the cliffs of the whose frequency and intensity is likely to increase is a key cape (Fig. 1). element in the management of vulnerable coastal spaces. Paradoxically, in spite of its high vulnerability to risk Several studies have shown that the damages caused by re- (the state of natural disaster has been declared 14 times in cent disasters in coastal areas have been aggravated by the the area since 1982, including 13 times for “flood/submer- local authorities’ lack of preparedness (Daniels et al. 2011; sion” events, GASPAR database of the French Ministry of Genovese et al. 2011; Genovese and Przyluski 2013). It is Ecological Transition, http://www.georisques.gouv.fr/ not enough to know and prevent risk: we also need to be acces-aux-donnees-gaspar), the town of Leucate has been prepared to manage the resulting crises (Lagadec 1993, relatively slow in applying risk regulation measures in 2001, 2004, 2012; Egli 2013). In this perspective, the comparison with other municipalities located on the CRISSIS project aims to help local stakeholders use the Languedoc-Roussillon coastline. According to a re- new findings produced through the analysis of the three cent publication of the Department of Planning and dimensions of risk. Two crisis management exercises were Housing of the Rhône-Mediterranean basin (DREAL thus organized from the outset: one took place halfway Rhône-Méditerranée 2016), the first PPRI (“Plan de through the project to test the uptake of crisis manage- Prévention des Risques Inondation”,or “Flood Risk ment procedures (« Plan Communal de Sauvegarde » Prevention Plan”) was only adopted in November (PCS), or « Municipal Rescue Plan ») by local authority 2012, while all the other municipality in the region staff; a second more ambitious test took place at the end subject to marine submersion risk had already had Durand et al. Geoenvironmental Disasters (2018) 5:19 Page 3 of 17 Fig. 1 Location of study area one for several years, and to this day this document – The hazard and its predictable evolution in the has not made it past assessment stage. A local Muni- context of climate change: digital simulations of sea cipal Safeguarding Plan (PCS) was adopted in De- level variations caused by storms or other factors, cember 2014 after the town was hit by two violent and of runoff propagation into urban areas, and storms in March 2013 and November 2014. high-resolution mapping of findings; In this context combining strong hazard and major – Vulnerability: structural vulnerability of buildings and vulnerability, the local authorities only appear to have human vulnerability of inhabitants (based on age and become aware of risk very recently. The town of Leucate mobility criteria), mapping of potential damages to therefore provides an interesting site to experiment with constructions and associated human issues; the integrated and operational marine submersion risk – Perceptions and representations held by the population management approach developed by the CRISSIS (or “risk culture”), captured through a geo-sociological program. survey aimed at quantifying risk culture levels; produc- tion of a map of findings to help improve information Methodology: An integrated and operational management and awareness-raising activities. approach to marine submersion risk The CRISSIS approach integrates the three dimensions This project aims to improve both risk prevention and of submersion risk: crisis management by local authorities. This aspect forms Durand et al. Geoenvironmental Disasters (2018) 5:19 Page 4 of 17 the fourth phase of the program: it consisted in developing The CRISSIS project therefore chose to work from a exercises to test crisis management mechanisms and pro- numerical modeling system to characterize flood haz- cedures based on existing operational documents (PCS), ard. The modeling is based on a chain of models – complemented by new data produced through our ana- MARS (Lazure and Dumas 2008), SWAN (Booij et al. lysis of the three dimensions of risk. 1999), SWASH (Zijlema et al. 2011) – in order to model the variation in sea levels caused by storm con- ditions (atmospheric pressure, wind set-up) and the Analyzing hazard using a numerical modeling system contribution of wave set-up to the rise of average sea In France, coastal flood hazards are still often approached levels, and to assess the volume of water overtopping through basic mapping methods obtained by cross-refer- the seafront and the propagation of the water on land encing the topography of exposed areas with extreme (Nicolae et al. 2015; Nicolae et al. 2018). The modeling water levels, estimated through a statistical analysis of process comprised of two stages. marine weather forcing conditions. This methodology was for instance used to elaborate official risk exposure docu- Stage 1: Testing and validating the modeling system ments (Coastal Flood Risk Prevention Plan or PPRSM) for The test consisted in simulate two recent storms the Mediterranean coastline. These basic methods are not (March 2013 and November 2014) that presented di- however suited to optimal crisis management. They only verse weather conditions and consequences in terms of produce schematic maps that do not include the event’s flood on the area of study. The all description of the kinetics, the detail of the process that caused the flood (in- validation process is detailed in Nicolae et al. 2018 and undation, overtopping of a barrier, breaching of flood only main results are presented here. Data, which was defense structures…), or the specificities of water propaga- used to recreate similar weather conditions on the site tion in urban area. Furthermore, this type of mapping as- of Leucate, is listed in Table 1. sociates the hazard with extreme statistical characteristics, Information has been collected from multiple sources which is not always compatible with the need for precise (press, photos, interviews, testimonies, report from tech- information in cases where the crisis might have been nical services, etc.) to characterize flood events (intensity, caused by a lower-intensity event. sea level, chronology). We were thus able to retrace the af- Digital modeling is now increasingly used to map out fected areas and the chronology of events. These observa- hazards, using chains of models to recreate the process tions, although mostly of a qualitative nature, enabled us that caused the flood: inundation, overtopping, breaching to estimate the level reached locally by the flood based on of flood defenses, etc. (Gallien et al. 2014; Guimarães et al. human landmarks (sidewalks, walls, quays). These points 2015; Le Roy et al. 2015; Nicolae et al. 2018). Recent im- of observation were cross-referenced with LIDAR data provements, both in terms of technical developments and and/or DGPS measurements collected through a field of pre- and post-data processing, have made it possible to campaign, in order to quantify in term of water height integrate both the phenomena that contribute to floods each qualitative observation. The reach of the flood waters (rise of average sea level, waves, river water) and their extension limit in the most heavily affected areas was also chronology (duration of floods, breaching of defense mapped out, consulting local authority staff who had structures, synchronicity of the maximum values of sea worked on the ground during the storms. In the absence and river levels). This has made it possible not only to of local records of water levels in the port, the information identify risk areas, but also to evaluate timescales and po- collected and converted in water height (in reference to tential response times between the start of an event and the French terrestrial datum, IGN69) provided a valuable the point where communication lines or strategic build- information to assess the quality and precision of the ings are damaged. Some important efforts are currently modeling of water levels and submersion in various sites being deployed using meta-modeling or machine learning Table 1 Observed and simulated data to achieve a real-time forecast of floods (Goulby et al., Data Past events 2014; Jia et al., 2015; Rohmer et al., 2016). However, this work remains too experimental to be used in an oper- March 13 Nov 14 ational setting. The most effective solution for mapping Sea Observed SHOM tide gauge in Sète SHOM tide gauge in levels Port La Nouvelle out such hazards remains combining a statistical analysis of the varying intensity of marine weather conditions with Waves Observed Candhis buoy, Leucate Candhis buoy, Leucate a field diagnosis of local vulnerabilities (coastal narrow Simulated Model IFREMER Model IFREMER dune, fragile seafront structures, water drainage system…): MEDNORD MEDNORD we can thus elaborate multiple scenarios in consultation Winds Observed France Leucate weather France Leucate with local stakeholders, and produce a tool to support station weather station decision-making in crisis situations. Durand et al. Geoenvironmental Disasters (2018) 5:19 Page 5 of 17 across the municipality. The water levels generated As part of this research project in Leucate, the diagnosis through the simulations were compared with those de- of material and human vulnerability was evaluated on duced from the analysis of topographic landmarks (quays, multiple scales. We first assessed the vulnerability of ma- roads) that appeared on photographs to be submersed or terial goods exposed to hazards, considering that vulner- not by the floods. By looking at various landmarks across ability also affects the rest of the area (D’Ercole and the port area, we established that the average water level Metzger 2009). In addition, we carried out a comprehen- in the port was 0.85 m IGN +/− 5 cm IGN in 2013, and sive diagnosis of material and human vulnerability, consid- 1.05 m IGN +/− 5 cm in 2014. The water levels generated ering the dangerousness of each building located in a by simulations in the area were within a similar range to high-risk area in the municipality. Within this sample, we those estimated from photographic observations (a dis- carried out a diagnosis of first homes, second homes and crepancy of under 5 cm was observed for the 2013 simula- tourist accommodation (about 4000 buildings) as well as tion, and an underestimation of about 10 cm for 2014), commercial premises (about 100). However, to further which indicated that the performance of the modeling sys- elaborate our vulnerability criteria – and more specifically tem was relatively reliable. those relevant to human vulnerability – a diagnosis was drawn based on a perception survey (see below). Add- itional evaluation criteria were used for this sample, but Stage 2: Elaboration of submersion scenarios they only covered 400 buildings (tourists who do not own After validating the modeling system, we developed various their accommodation were excluded from the sample). submersion scenarios, based on a multivariate statistical The criteria were set to capture the potential damages analysis of extreme values (joint probability of waves/water caused by coastal floods, assessing both material vulner- levels). This analysis enabled us to define low-probability ability (type and condition of building) and human vulner- marine conditions (with a return period of 100 years), ability with indicators assessing people’s access to safety. whichwerepropagatedtothe shoreand then to theland The advantage of using two different samples was that (Nicolae et al. 2018). These simulations enabled us to they provided an operational diagnosis across the area af- characterize the reach and intensity of floods for hundred- fected by risk. This also helped us respond to the expecta- year events, and to anticipate the consequences of a gradual tions of local authorities, by showing that a more in-depth rise of the average sea level under the effect of global warm- diagnosis could be carried out to improve crisis manage- ing. This data was fed into a GIS in order to create relevant ment by identifying and locating vulnerable individuals. maps for crisis management exercises. The surveys included questions based on vulnerability cri- teria and indicators (extract: see Table 2). The methodology Assessing the vulnerability of buildings and populations used to assess vulnerability was adapted to the local context Over the past few years, the geographic understanding of of the municipality of Leucate, using evaluation criteria that natural hazards shifted from a hazard-centered approach had been designed for other locations and hazards and had to a vulnerability-centered approach, integrating the social been proven to be relevant to this site (Leone 2007; dimension of hazard. Such a diagnosis and evaluation of Meur-Ferec et al. 2011; Lagahé and Vinet 2014). vulnerability presents advantages in terms of anticipating Some criteria capture both material and human risk and mitigating its impacts. Although this notion has vulnerability, while others only capture one type of evolved, vulnerability can be understood both in its pri- vulnerability. mary meaning as the capacity to withstand damage (D’Er- A vulnerability index was developed from these criteria. cole et al. 1994) but also as the capacity to cope with When taken individually, a criterion only captures a par- damage (Thouret and D’Ercole 1996), thus echoing the tial assessment of one specific type of vulnerability. How- concept of resilience. Resilience is perceived as the posi- ever these criteria and indicators can be combined to tive counterpart to vulnerability, which usually carries provide a global assessment of vulnerability, where some negative connotations due to its association with fragility, criteria are compensated by others. although the two terms are not mutually exclusive The methodology followed to create this index con- (Reghezza Zitt and Rufat 2015). sisted in setting a hierarchy between indicators and cri- This study considers the geographic dimension of risk on teria, by using an intra- and inter-criteria weighing an infra-municipal scale, with a vulnerability diagnosis fo- system. The hierarchy was based on the bibliography cusing on individual issues (on the scale of a building). The and on past studies on this topic, as well as on a global notion of vulnerability implies an assessment-centered ap- reflection on theimportanceofagivencriteriaincom- proach (Leone et al. 1996;Becerra 2012), which must begin parison with another, always from the perspective of a with the identification of criteria and indicators. Many re- scenario of potential damage. We were thus able to cent studies on the subject show the relevance of this meth- elaborate maps that were fed into the development of odology (Barroca et al. 2005;Leone 2007). crisis exercise scenarios. Durand et al. Geoenvironmental Disasters (2018) 5:19 Page 6 of 17 Table 2 Extract from list of vulnerability criteria and indicators applied to Leucate (the sample of individuals surveyed) Inter-criteria Criteria Indicators Intra-criterion weighting weighting 0to6 Type of property Multi-storey accommodation; single-storey with roof exit; single-storey 0 to 8 Material depending on without roof exit. vulnerability criteria Structure and condition Age of accommodation; floor level; types of materials; waterproofing 0to 4 of property of openings; height of electrical connections. Type and structure Multi-storey accommodation; single-storey with roof exit; single-storey 0 to 8 Human of property without roof exit; ground floor level; number of bedrooms and inhabitants vulnerability Accessibility Type of access road; emergency exits 0 to 4 Characteristics of Number of persons; under 10 years old; over 60 years old; over 0to 3 population 80 years old; persons with reduced mobility and/or dependent persons Perception analysis and representations of risk In order to measure risk culture levels in the munici- An understanding of the perceptions, behaviors and re- pality of Leucate, the methodology focused on three actions of individuals and groups of individuals in the main angles: questionnaire-based surveys of inhabi- face of any kind of risk should form a prerequisite to tants; semi-structured interviews with key stakeholders; the development of prevention and management pol- and a discourse analysis of the local news bulletin, Cap icies (Ruin 2010). This process can help researchers Leucate. In total, we screened 46 issues ranging from better grasp the discrepancy between the knowledge of January 2010 to June 2015 (the bulletin is monthly in experts and that of laypersons (Baggio and Rouquette theory), which underwent an assisted textual analysis 2006). The perception of risk varies between individ- using the Tropes software (Pont 2015). The term uals: studies on this topic attempt to characterize “submersion” (“submersion”) appears in the newsletter groups of individuals that share a relatively similar per- for the first time in the March 2015 issue, to refer to ception of a given risk (Hellequin et al. 2013). An indi- the municipal rescue plan established by the local au- vidual’s perception of risk is influenced by four factors: thorities, and in an article that announces the launch of their experience of risk (through direct or indirect ex- the CRISSIS research project. Before the creation of of- perience), their knowledge on this risk (lay or aca- ficial documents (PCS and PPRL), municipal authorities demic), their social and economic interests (attachment would use the word “inondation” (“inundation”)when to a place, value of property, socio-economic standing referring to coastal floods. The word “inondation” was of household, etc.), and their values (moral, political, written sixteen times, including three times during the etc.). Over the past twenty years, the analysis of percep- 2013 flood that hit Leucate Plage, and three times in tions and social representations has played an increasingly 2015 in relation to the announcement of this research prominent part in research programs on French coastal program. towns’ vulnerability to marine hazards (Goeldner-Gianella The questionnaire survey was communicated to 493 and Chionne 2014; 2002–2004: PNEC, Nord-Pas-de-Calais; people including first home owners, holiday home owners 2007–2011: ANR MISEEVA, Languedoc - Roussillon; and tourists living in Port Leucate, the naturist village and 2011–2014: ANR COCORISCO, Bretagne, etc.). These Leucate Plage (Table 3). The survey was carried out from studies have highlighted the weakness of social repre- april to june 2015. At this time of year, many holiday sentations of coastal risks (erosion and marine submer- homes are unoccupied, making it easier for us to survey sion) and even the “blatant absence of a risk culture” residents and secondary home owners. The surveys were (Chauveau et al. 2011). Most surveys show that the conducted face-to-face at the respondents’ homes and populations affected feel little concern for coastal risks. lasted 20 to 40 min. The people present were rather suspi- They do not reflect on their situation, and nor do they cious, we had several refusals. Overall, the questionnaires actively seek out information or design any action plans were well completed, with respondents answering 93% of in the event of a disaster (Hellequin et al. 2013; the questions. It would have been useful to base our popu- Flanquart 2014;Krien 2014). When questioned, the lation and location sampling on the modeling of submers- people surveyed tend to make up a spontaneous dis- ible areas and of the vulnerability of buildings, which were course on the question by drawing from their memories, described above. However, as the duration of the pro- sensations and incomplete reflections on the question. gram was limited to two years, we had to carry out all Krien (2014) explains that people’s representation of research activities simultaneously, and were not able to coastal risks is the product of a social construct developed use this sampling methodology. Individuals were sur- from their overall representation of risk, the sea, storms veyed on: (1) the appeal of the Leucate municipality and the place where they live. and its leisure activities; (2) their perception of the Durand et al. Geoenvironmental Disasters (2018) 5:19 Page 7 of 17 Table 3 Breakdown of surveys in Leucate based on area of residence and residential profile of individual surveyed Resident profile / First home owners Second home owners Tourists Total surveyed area Leucate Plage 2% (n = 12) 4% (n = 18) 9% (n = 42) 15% (n = 72) Naturist village 5% (n = 23) 4% (n = 20) 1% (n =3) 9% (n = 46) Port Leucate 32% (n = 158) 36% (n = 174) 9% (n = 43) 76% (n = 375) Total 39% (n = 193) 43% (n = 212) 18% (n = 88) 100% (n = 493) area’s and of their accommodation’s exposure to coastal work effectively and efficiently together to prepare for, flood risk; (3) their awareness of this risk in Leucate; respond to, and recover from domestic incidents.. To (4) their presumed reaction in the event of a submer- provide for interoperability and compatibility among sion; and (5) their expectations in terms of communica- Federal, State, and local capabilities, the NIMS will in- tion about this risk. The final objective was to map out clude a core set of concepts, principles, terminology, representations of risk, in order to help shape the local and technologies covering multi-agency coordination authorities’ information and awareness-raising activities systems; unified command; training; identification and on coastal flood risk. management of resources ; qualifications and certification; and the collection, tracking, and reporting of incident Testing crisis management procedures through crisis information and incident resources” (Anneli 2006). To exercises regularly train the stakeholders, the FEMA (Federal Crisis exercises: A brief state of the art Emergency Management Agency) has a National Exercise The last stage in the CRISSIS program’s integrated ap- Program, https://www.fema.gov/ned) which regularly runs proach brought added operational value to the project. large-scale drills simulating the occurrence of cyclones, Recent research on risk assessment in coastal areas – for earthquakes and tsunamis. In Europe, this task is covered instance in the Languedoc Roussillon region, the projects by the European Civil Protection Mechanism (ECPM, RNACC (“Risques Naturels, Assurances et Changement http://ec.europa.eu/echo/what/civil-protection/mechanis- Climatique” [“Natural Risk, Insurance and Climate m_en) which organizes and funds a series of drills in the Change”], Yates-Michelin et al. 2011)and ANR MISEEVA field of civil protection in different member countries (2008–2011, Vinchon et al. 2011; Meur-Férec et al., 2011) every year. The French example provides a good gauge for – did not go as far as integrating crisis management. And a national policy for crisis drills. With a concern for opti- yet the preparedness of local authorities is essential. Any mizing the response from stakeholders and the public in a failings on their part can significantly amplify the impact crisis, the 2004 law on the modernization of civil security of extreme phenomena in coastal areas, turning a crisis (law #2004–811 dated 13 August 2004) makes regular ex- into a full-blown disaster (Daniels et al. 2011;Genovese et ercises compulsory, with a requirement for large-scale al. 2011). This is particularly true in regions where the drills several times a year at national and regional levels stakes are high and diverse, and where a crisis can involve (Richter-type earthquake drills are a good example) and a multiple public and private stakeholders. Local players requirement for all towns and villages where a major nat- should therefore be trained in crisis management, includ- ural or technological risk has been identified to conduct at ing by organizing crisis exercises (Lagadec 1993, 2001, least one crisis drill per year (DGSCGC or Direction 2012;Stern 2014). Générale de la Sécurité Civile et de la Gestion des Crises, To meet this challenge, public authorities in many 2008). According to the guidelines set by the General countries have developed emergency management sys- Directorate for Civil Security and Crisis Management tems for using by stakeholders and organize regularly (DGSCGC), a body whose equivalent can be found in large-scale crisis exercises to prepare institutions and most countries, there are two main types of exercises populations to cope with major crises. Thus, in the (Institut National des Hautes Etudes de la Sécurité et de la USA, following huge forest fires in California and Justice (INHESJ) 2015): “table-top exercises” (“exercices Arizona in the 1970s, which had highlighted a lack of cadres”)and “field conditions exercises” (“exercices ter- coordination of emergency response between different rain”). The former are desk-based and take place at the actors, several states have developed the Incident crisis headquarters. They do not involve the deployment Command System (ICS), which, after the attacks of of any resources on the ground. Participants receive infor- September 11, 2001, was extended to all US states and mation by radio, phone, fax, television, SMS and the social integrated into the National Incident Management networks. They are required to analyze, and synthetize System (NIMS). NIMS is “a consistent nationwide ap- this information, react, report to others, make proposi- proach for Federal State, and local governments to tions, set priorities and make choices to manage the crisis Durand et al. Geoenvironmental Disasters (2018) 5:19 Page 8 of 17 for the better. The second type involves the deployment of analysis, vulnerability and population perceptions/repre- men and equipment on the ground. One of its objectives sentations. The training followed the format of a frame- is to test the transportation and deployment of equipment work exercise, which was organized in Paris by students in real-life and real-time conditions. Good examples of from the Master’sdegree “Gestion Globale des Risques et such activities include the exercises organized periodically des Crises” (“Global Risk and Crisis Management”,or in the US by the FEMA or those organized by area prefec- GGRC), working from a low-intensity crisis scenario. In tures in France. For instance, the European Sequana 2016 practice, the exercise ran over three hours and the partici- exercise, held in March 2016 and which simulated the oc- pants broke out into two units: (i) a “facilitation” unit currence of a 100-year flood of the Seine in Paris, involv- where the participants played the parts of Préfecture staff, ing public stakeholders at various levels (national, zonal, firefighters, security forces, the media and the population, departmental and municipal) and an array of private oper- who were responsible for sending “inputs” based on the ators (transport firms, telecommunications companies, chosen scenario to (ii) the crisis unit that simulated the banks, supermarkets, hospitals, and so on). On the one work of the municipal command team (PCC). hand, the goal of this exercise was to test Paris area re- Following this first training session, a second, more am- gional stakeholders’ ability to coordinate and to cope with bitious exercise took place locally a year later, in March response and crisis management. On the other hand, that 2016 (duration: 3 h). This time, the exercise involved all test was also an opportunity to assess the relevance of op- the municipal staff including elected members of the erators’ crisis management plans as well as the consistency Council, as well as senior representatives of the local au- of operational procedures. Eventually, public authorities thorities: the Préfecture of the Aude département (where were also provided with a feedback and the scope of infor- Leucate is located), département managers of the fire bri- mation for population was measured for ensuring their gade and the DDTM (“Direction de l’Aménagement et du awareness can be improved in the event of a disaster. The Territoire”,or “Planning and Territorial Directorate”, a de- exercise uncovered two points to improve in the emer- volved State body in charge of risk prevention). This time, gency response: the coordination of public and private ac- the session included both framework and field activities. tors, very heterogeneous according to the sectors of Some elements were performed in a real-life setting activity, and the information of the population, which (on-the-ground deployment of technical staff and munici- sometimes lacked efficiency (some instructions have not pal police forces) to test the coordination of the municipal been well assimilated) (Creton-Cazenave and November crisis unit (municipal command team, or PCC), with re- 2017). However, in France, in spite of these efforts, crisis ex- sources deployed on the ground. However, because of fi- ercises remain all too rare, especially at local level. In its nancial and public security constraints, it was not possible 2012 activity report, the French Committee for Civil to perform the entire exercise in real-life conditions: for Defense (Haut Comité Français pour la Défense Civile instance, civil safety (firefighters) and security forces (gen- (HCFDC) 2012) pointed out that in France “crisis man- darmerie) were not involved in real life but fictionally. Fa- agement procedures are too often untested” for cilitators played the part of the media (television, press), high-intensity events. In addition, such exercises, whether the social networks, the population affected by the crisis they are “table-top” or “in real conditions” exercises, take and the private operators involved. place at the initiative of major public institutions (govern- This time, the scenario included new data (on hazard ment agencies, prefectures, civil protection departments) and vulnerability) acquired during the project’sprevious but are rarely if ever initiated by the research sector. stages. The main objective was to test the “marine sub- In this context, the CRISSIS program aimed to test the mersion” section of the PCS to which detailed specific municipality’s current crisis management procedures (set in and intermediary objectives had been added, as shown the December 2013 PCS) and help optimize them through in Table 4. two crisis exercises involving local stakeholders, organized The scenario was entirely designed by the students of and designed according to a principle of progressivity. the GGRC Master’s degree from Université Paris 1, under the supervision of crisis exercise specialists, and Methodology of program exercises with the assistance of two senior members of municipal The first exercise took place at the start of the program in staff (the Head of environment and a Municipal Police March 2015. It consisted in a simple training session manager) as well as a manager from the Aude préfecture aimed at informing the municipality’ssenior administra- (deputy risk manager) who brought their field experi- tive staff (Head of municipal police, Head of technical de- ence on board and helped develop credible “inputs”. The partment, Director General of administration) of the main danger when creating such exercises is that participants procedures and mechanisms listed in the Municipal might in retrospect challenge their credibility – for in- Rescue Plan, which had recently been adopted. This ses- stance because they consider that their impact was too sion did not include the program’sfindings onhazard low or too high, or too remote from real-life situations. Durand et al. Geoenvironmental Disasters (2018) 5:19 Page 9 of 17 Table 4 Extract from the objectives of the March 2016 exercise Defense and Civil Protection (Service Interdépartemen- tal de Défense et de Protection Civile, SIDPC) of the Global objective Intermediary objectives Specific objectives Aude Préfecture in Carcassone (sent out emergency noti- Testing the marine Testing the SAIP Testing the logbook submersion section (SIRENS) / SMS Alerts fications from the Préfecture, responded to the munici- of the PCS pality’s requests for additional resources); (ii) at Understanding the PCS Testing the relevance and its role in the crisis and proficiency of the municipal level, from a site in Leucate, by the students room crisis room’s tools from the GGRC Master’s and by the town’s two contact Confirming the crisis Testing communication persons (sent out all other “inputs”: reports of local management sites damage caused by the storm, requests from the popula- Testing the municipality’s Testing the power tion, media queries, pressure on the social networks, organization in the event generator etc.). The crisis was simulated on the evening of Satur- of a crisis (staff, communications, etc.) day 9 April 2016 – that is, over a weekend during mid-tourist season (the week-end before a major sport- Testing the municipal Testing connections staff’s knowledge and (phone, radio) ing event organized in Leucate, the “Mondial du Vent”). understanding of During this period the town usually hosts 8000 to the PCS 10,000 people (for a permanent resident population of 5000) staying in secondary homes, campsites and If this had been the case, the exercise would not have camper van parks. Figure 2 shows an overview of the achieved its learning objective of consolidating the mu- roll-out of the three-hour exercise, which has also been nicipal staff’s knowledge of crisis management tools and documented in a film that is available on the program procedures, and where possible helping improve them. It website (http://crissis2015.free.fr/). was therefore essential for this scenario to be developed in collaboration with individuals who had an excellent Results and discussion knowledge of local realities and institutions. Production of high-resolution marine submersion maps Stimuli were produced on two levels: (i) at Préfecture The validation of the digital modeling system and the level, by staff based at the Interdepartmental Service of multivariate statistical analysis of extreme values on record Fig. 2 Flowchart of the roll-out of the crisis exercise, March 2016. Source: ©Master GGRC, 2016 (PCC: municipal post of command; PREDICT: consultancy, service provider for the local authority; DEBEX: start of exercise; FINEX: end of exercise) Durand et al. Geoenvironmental Disasters (2018) 5:19 Page 10 of 17 (see above) enabled us to elaborate several submersion sce- Mapping out critical vulnerability levels narios. Simulations were run to characterize the spread and The analysis of vulnerability using the method described intensity of submersions for a hundred-year event (Nicolae above enabled us to establish a vulnerability index (see et al. 2018), and to forecast the consequences of a gradual Table II), which was mapped out to highlight the most rise of the average sea level under the effect of climate vulnerable buildings and populations. For Leucate, three change. For this purpose, we estimated the impact of a categories were defined and prioritized, ranging from low minor rise in sea level working from a variation of + 0.2 m to medium and high vulnerability (Figs. 4 and 5). (the average global rise forecasted according to the median Spatial mapping makes it possible to draw compari- scenario for the 2046–2065periodincomparisonwith sons within one single entity, but also between entities 1986–2005 – source: IPCC WG1 Ch13, Church et al. 2013) themselves, highlighting in this case a variation in and + 0.6 m (the average sea level rise forecasted in the levels of material vulnerability between areas. After Mediterranean by 2100, Slangen et al. 2014). assessing vulnerability, this ongoing study aims to rec- For the March 2016 crisis exercise, our consultation ommend strategies to adapt buildings depending on with the two members of municipal staff who contrib- their exposure to risk. This diagnosis will be expanded uted to the script drove us to simulate a coastal flood to the totality of buildings in risk areas. By globalizing caused by a twenty-year storm, which combined the vulnerability, we will thus be able to assess not only characteristics of the last two major storms (March material vulnerability but also territorial vulnerability 2013 and November 2014) in terms of wave height and (D’Ercole and Metzger 2009). sea levels – the former had caused the breach of a sea- These maps of vulnerability were integrated to the front wall in Leucate Plage, and the latter had provoked PCS as appendixes, and made available to participants a flood by overtopping on the seafront and overflowing for the second exercise in May 2016. Their purpose in Port Leucate (Fig. 3). The objective was therefore to was, where possible, to inform the first protection engage the municipal staff in a role play using a sce- measures taken by municipal staff during the crisis nario that combined events they might already have en- (for instance, setting up coffer dams), as well as the countered, but separately. This helped ensure that the organization of evacuation. scenario had enough impact to place the players under pressure, while not being too “overblown”, which could A poor risk culture have been the case with a simulation of a hundred-year Our analysis of perceptions and representations showed storm that factored in the rise of the average sea level. a poor awareness of coastal flood risk amongst surveyed Fig. 3 Submersion scenario, a Leucate Plage, b naturist village, c Port Leucate, created from the digital model and used for the March 2016 field exercise Durand et al. Geoenvironmental Disasters (2018) 5:19 Page 11 of 17 Fig. 4 Material vulnerability of people surveyed in Leucate (source: field survey) populations. For instance, of all residents and tourists existence of this system – and this despite the fact that surveyed in areas that were shown by our model to be of all second home residents surveyed, only 29% de- potentially submersible, only 50% of people surveyed de- clared they were never present in Leucate between Octo- clared that they were living in a risk area (Fig. 6). ber and March (the high-risk period for coastal floods). 43% of first and second homeowners declared that the Consequently, it appears necessary to implement a com- municipality of Leucate was not threatened by coastal munication strategy to inform the population, and in par- floods. This finding is combined with an underestima- ticular second-home owners, of existing emergency alerts. tion of risk. In comments recorded as annexes to the Although the findings of this study of perceptions/repre- questions, some people who had been informed of past sentations were not used in the March 2016 exercise, they marine submersion events explained that they did not informed the recommendations drafted in our exercise consider the few intrusions of sea water into the streets debrief notes to improve the municipality’s information, and houses as floods. According to these people, the awareness raising and communication strategy. hazard does not exist until there are human casualties, and material damage is considered unimportant and re- mediable. We also sought to find out whether residents Lessons learned from the exercise debrief: A poor were aware of the systems set up by the municipality to integration of crisis management systems and procedures alert them in the event of a risk, including the volunteer-run SMS alert system outlined in the PCS. On An exercise debrief includes a methodical analysis of this point, we found a significant difference between first the exercise, to highlight its strong points and home and second home owners: 34% of the former had improvement points in order to perfect the communicated their contact details to the local author- organization’s crisis management processes. In this ities, while 74% of the latter were unaware of the case, the scope of the lessons learned went beyond a Durand et al. Geoenvironmental Disasters (2018) 5:19 Page 12 of 17 Fig. 5 Location of vulnerable populations based on age and health status (source: field surveys) sole technical fixing of failing tools or processes. The crisis management. Following these observations, the aim was to question individual and organizational municipality of Leucate introduced geo-referenced responses to extreme event that might pose a challenge flood markers into its PCS and committed to fully par- to the system (Lagadec 1993). ticipating in the second exercise to improve municipal staff’s command of the PCS. Our debrief from the first exercise (March 2015) re- The debrief for the second exercise (March 2016) vealed a poor knowledge of the Municipal Rescue Plan was captured in a report that was presented in early (PCS) on the part of municipal staff, and helped us May 2016 to all municipal staff. It highlighted the mu- present a few observations on its contents. For in- nicipal teams’ excellent field knowledge, as well as a stance, the PCS had no objective quantifiable mile- good coordination between the municipal post of stones (in meters) regarding the water levels attained in command (PCC) and the teams deployed on the submersed areas: the decision to activate the successive ground. However, it also revealed the municipal staff’s phases of the PCS (closing roadways, evacuating certain obvious lack of knowledge of the PCS: most staff had neighborhoods) were only based on field information no clear understanding of their respective roles and communicated by technical staff who have been work- responsibilities within the crisis unit, and therefore ing for the municipality for years and therefore have an faced difficulties in organizing and coordinating the in-depth knowledge of local sites. However this lack of PCC. Furthermore, these challenges were aggravated quantifiable milestones can be a disadvantage for inex- by the crisis headquarters’ lack of practicality: the perienced staff (for instance people who are new in functions of secretariat, command, logistics and co- post, or covering for more experienced permanent staff ordination were all gathered in a single room, which members who might be off work when a crisis occurs), caused a great deal of confusion. In this context, the leading to damaging delays in a context of emergency fear of an error was a major source of stress, causing Durand et al. Geoenvironmental Disasters (2018) 5:19 Page 13 of 17 Fig. 6 Perceptions of residents on their property’s exposure to marine submersion risk communication challenges both internally within the (iii)Writing a checklist to be inserted at the start of the crisis unit (absence of situation updates) and exter- PCS, outlining the key principles of crisis nally (communication with Préfecture services and the management: sharing the logbook; updating the media). Consequently, the debrief report included four schedule and map in real time; regular situation main recommendations: updates, and communication of human casualties to the higher echelons (Préfecture), etc. (i) Reorganizing the PCC according to the traditional (iv) In addition, the municipality should improve its organizational structure of a crisis unit, to facilitate communication with the population ahead of the communication and decision-making in the event of crisis (for instance via the local news bulletin) on a crisis. The space should be split into five open current crisis alert systems and in particular the SMS units (Fig. 7): decision-making and coordination alert (this recommendation directly derives from the unit; situation unit (secretariat/log); logistical unit; analysis of perceptions/representations of risk). communication unit; forecasting unit) (ii) Drafting concise post descriptions to be included Discussion: What are the obstacles faced by an into the PCS, so that from the moment of their operational approach to coastal flood risk? arrival into the PCC each member of the crisis The resolutely operational approach adopted by the unit knows what their role is and what they are CRISSIS program faced three challenges. First of all, it has expected to do. Examples of post descriptions: been difficult to convince local representatives to organize crisis unit Director (DOS); unit Coordinator a crisis exercise locally. While the first exercise, which (needs to be very mobile and move between took place in Paris in March 2015, did not pose any major units); Secretary (keeping a log, handling calls); issues as its stakes were lower (low-intensity exercise, re- communication unit / forecasting unit / logistical mote location, framework exercise only, involving only unit Managers. two technical staff members and the Director General of Durand et al. Geoenvironmental Disasters (2018) 5:19 Page 14 of 17 Fig. 7 Example of proposed reorganization of the PCC suggested in the exercise debrief report municipal services as observers), organizing the second had received from hydro-meteorological experts Predict proved markedly more complex. As this exercise needed (the municipality’s current service provider) were used to take place locally and involve large numbers of Council properly. However, this was not the case with the maps members and municipal staff, we were initially faced with showing people and buildings’ critical vulnerability levels, a clear reluctance on their part. Eventually, only two which could have informed the first protection measures Council members (two Deputy Mayors) did take part in (for instance, setting up coffer dams) taken by personnel the exercise, along with about twenty administrative and on the ground. This challenge shows the necessity of pro- technical staff members. The 2004 Law for the viding the PCC staff with user-friendly documents that modernization of civil security (Act 2004–811 of 13 can be read quickly. Similar observations had been made August 2004) states that any municipality where at least during the crisis exercise organized for a group of munici- one major natural or technological risk has been identified palities in La Réunion by students from the École des must organize at least one crisis exercise per year Mines in Alès as part of the only other French program (DGSCGC, 2008). However, as highlighted by the French on submersion risk (Wassner et al. 2016)thatalsoin- Committee for Civil Defense in its 2012 report (Haut cluded an operational dimension and a crisis exercise. Comité Français pour la Défense Civile (HCFDC) 2012), Finally, the actual adoption of the recommendations very few municipalities actually comply with this obliga- presented in the exercise debrief report is a very tion. This reluctance may be due to the lack of municipal long-term process. Although these recommendations resources, or perhaps to the fear of being evaluated and were approved by the municipality when the report judged. In this respect, a lot of educational work is needed was first presented to them, over a year on only the to convince potential participants that the objective of first has led to concrete action being taken (the such exercises is not to evaluate individuals but systems reorganization of the PCC). The PCS has still not been and procedures, in order to improve them and make them modified. Besides, this plan has never been tested in easier to memorize. Only when the exercise was com- real-life conditions (as the last storm occurred in No- pleted did participants finally drop their guard and admit vember 2014, before its adoption): it still remains un- to its usefulness. sure whether the municipality does have the capacity In addition, municipal staff made a variable use of the to tackle a high-intensity storm. The most “extreme” operational documents (maps) presenting the findings of scenarios tested as part of the project’shazard analysis the analysis of hazards and vulnerability. The submersion show for instance that the spread and volume of the forecast maps that were provided to the PCC coordinator flood would increase by respectively 160% and 188% (Head of the municipal police) at various points during in the event of a hundred-year storm with a 0.6 rise of the crisis, in replacement for the maps he would normally sea level (Nicolae et al. 2018, Anselme et al. 2017). Just Durand et al. Geoenvironmental Disasters (2018) 5:19 Page 15 of 17 as it is necessary to implement a robust prevention Funding This research work was supported by the French Conseil Supérieur de la policy (the Submersion Risk Prevention Plan for Formation et de la Recherche Stratégiques (CSFRS), within the framework of Leucate was approved in January 2017), improving cri- the call for non-thematic projects 2012. The funding body has played no role sis management procedures should also be a priority in the design of the study, the collection, analysis and interpretation of data, or the writing of the manuscript. for the municipality in the years to come. Authors’ contributions BA CRISSIS program manager, manuscript’s redaction: intro, conclusion, study area, figures design and layout. PD preparation of crisis simulation exercises, Conclusion manuscript’s redaction: general structure of the paper, intro, conclusion, The initial objective of the CRISSIS program was to method, results and discussion (crisis exercises). ALN and SE numerical improve both the anticipation of risk and the imple- modeling, manuscript’s redaction: method, results and discussion (hazard). SD and MG vulnerability survey, vulnerability index design, manuscript’sredaction: mentation of adapted responses to better tackle its un- method, results and discussion (vulnerability mapping). LG-G and EL survey’s predictability and help with decision-making, in the conception, survey in the field, statistical treatment, survey’s analysis method, context of a highly urbanized municipality that is peri- results and discussion (risk perceptions and representations). All authors read and approved the final manuscript. odically impacted by sea water floods (inundation, bar- rier overtopping, breaching of barriers). To achieve Ethics approval and consent to participate this, our research focused simultaneously on the nat- Not applicable ural (hazard), material and social (vulnerability, per- Consent for publication ception/representations) dimensions of risk. However, Not applicable the objective was also to engage with local authorities Competing interests by inviting them to work on the operational aspect of The authors declare that they have no competing interests. crisis management, which is usually neglected by re- search programs on flood risk. In this perspective, we Publisher’sNote worked to improve the various stakeholders’ oper- Springer Nature remains neutral with regard to jurisdictional claims in published ational response by using crisis exercises to assess the maps and institutional affiliations. degree to which they had adopted the tools and sys- Author details tems placed at their disposal (PCC, PCS), but also by Université Paris 1 Panthéon-Sorbonne, 191 rue Saint Jacques, F-75005 Paris, helping improve these tools and systems using the France. Laboratoire de Géographie Physique, UMR 85911 place Aristide Briand cedex, F-92195 Meudon, France. Pôle de Recherche pour new findings produced by the program’sfirst three l’Organisation et la Diffusion de l’Information Géographique, UMR 8586 2 rue sections. Valette, F-75005 Paris, France. Université Paul Valéry, 2 rue du Professeur Overall, this project helped improve our understand- Henri Serre, 34080 Montpellier, France. UMR GRED – IRD (Gouvernance, Risque, Environnement, Développement), route de Mende, 34199 ing of submersible areas and create maps of critical vul- Montpellier, France. BRGM, 3 Avenue Claude Guillemin, 45100 Orléans, nerability, both material and human. It also revealed France. Centre universitaire de formation et de recherche de Mayotte, the poor social representation of marine submersion Dembéni, Mayotte. UMR Espace Dev, 500 rue JF Breton cedex 5, 34093 Montpellier, France. risk, as well as a clear lack of awareness of crisis man- agement systems and tools, and of behaviors that Received: 4 May 2018 Accepted: 3 October 2018 should be adopted in the event of a flood. These find- ings highlight the need for setting up a communication References strategy, to raise awareness of risk and inform the Anneli, J.F. 2006. 2006, the National Incident Management System: A multi- population of current alert mechanisms. The crisis ex- agency approach to emergency response in the United States of America. Rev. sci. tech. Off. int. Epiz. 25 (1): 223–231. ercise conducted in March 2016 with relevant munici- Anselme, B., P. Durand, Y.F. Thomas, and A. Nicolae-Lerma. 2011. Storm extreme pal departments showed the importance of creating levels and coastal flood hazards. A parametric approach on the french coast of such exercises working from a realistic scenario, to en- languedoc (district of Leucate). Comptes Rendus Geosciences 343 (10): 677–690. Anselme, B., P. Durand, Y.F. Thomas, and A. Nicolae-Lerma. 2017. Coastal flood sure that all staff in positions of responsibility are aware Hazard. Risk and crisis management : Questions raised by sea level rise ? 10–14. of procedures, but also to detect any potential gaps in New York: International WRCP/IOC Conference, (poster) Columbia University. the contents of the PCS. In a context where hazard and Baggio, S., and M.-L. Rouquette. 2006. La représentation sociale de l’inondation : influence croisée de la proximité au risque et de l’importance de l’enjeu. vulnerabilityare boundtobecomemoreextreme,this Bulletin de Psychologie 59 (1): 103–117. should be a key action in the necessary optimization of Barroca B., Pottier N., Lefort E., 2005, Analyse et évaluation de la vulnérabilité aux crisis management systems and procedures. inondations du bassin de l’Orge aval. Actes des septièmes rencontres de TheoQuant, Atelier 3 « Risques, vulnérabilité », Besançon, 12p. Becerra, S. 2012. Vulnérabilité, risques et environnement : l’itinéraire chaotique d’un paradigme sociologique contemporain. VertigO - la revue électronique en Acknowledgements sciences de l’environnement [En ligne] 12 (1). URL : http://vertigo.revues.org/ We would like to thank the Conseil Supérieur de la Formation et de la 11988). https://doi.org/10.4000/vertigo.11988. Recherche Stratégiques (CSFRS) for funding this research program. Our Booij, N., R.C. Ris, and L.H. Holthuijsen. 1999. A third-generation wave model for thanks also go to the municipality of Leucate who volunteered to take part coastal regions, part I: Model description and validation. Journal of in this program and greatly facilitated our work on the ground. Geophysical Research 104 (C4): 7649–7666. Durand et al. Geoenvironmental Disasters (2018) 5:19 Page 16 of 17 Cariolet, J.-M., S. Suanez, C. Meur-Férec, and A. Postec. 2012, 2012. URL : https:// Institut National des Hautes Etudes de la Sécurité et de la Justice (INHESJ), 2015. journals.openedition.org/cybergeo/25077. Cartographie de l’aléa de La gestion de crise en Europe. In Vers une coexistence des organisations submersion marine et PPR : éléments de réflexion à partir de l’analyse de la actuelles basée sur la culture des états membres ou vers une convergence commune de Guissény (Finistère, France). Cybergeo : European Journal of structurelle dans un système européen? Rapport de la 26ème session « Sécurité Geography [En ligne], Espace, Société, Territoire, document 586, mis en ligne et Justice » 2014–2015, Groupe de Sécurité stratégique, Paris, 147 p. le 02 février. https://doi.org/10.4000/cybergeo.25077. Krien N., 2014, Place des risques côtiers dans la représentation du cadre de vie Chauveau, E., C. Chadenas, B. Comentale, P. Pottier, A. Blanlœil, T. Feuillet, D. d’individus possédant des enjeux sur des communes « à risque », thèse de Mercier, L. Pourinet, N. Rollo, I. Tillier, and B. Trouillet. 2011. Xynthia : leçons l’Université de Bretagne occidentale, 237p. d’une catastrophe. Cybergeo : European Journal of Geography [En ligne], Lagadec P., 1993, Apprendre à gérer les crises - société vulnérable, acteurs Environnement, Nature, Paysage, document 538, mis en ligne le 09 juin responsables, Editions d’Organisation, 120p. 2011. URL : https://journals.openedition.org/cybergeo/23763. https://doi.org/ Lagadec P., 2001, Les exercices de crise : pour des ruptures créatrices, La lettre 10.4000/cybergeo.23763. des cindyniques, n°34. Church, J.A., P.U. Clark, A. Cazenave, J.M. Gregory, S. Jevrejeva, A. Levermann, M.A. Lagadec, P. 2004. Understanding the French 2003 heat wave experience: Beyond Merrifield, G.A. Milne, R.S. Nerem, P.D. Nunn, A.J. Payne, W.T. Pfeffer, D. the heat, a multi-layered challenge. Journal of Contengencies and Crisis Stammer, and A.S. Unnikrishnan. 2013. Sea level change. In Climate change Management 12 (4): 160–169. 2013: The physical science basis. Contribution of Working Group I to the Fifth Lagadec, P. 2012. Du risque majeur aux mégachocs. Dossier spécial : la gestion de Assessment Report of the Intergovernmental Panel on Climate Change, ed. T. crise, méthodologies et retours d’expérience, Sécurité et stratégie 10: 50–52. F. Stocker, D. Qin, G.-K. Plattner, M. Tignor, S.K. Allen, J. Boschung, A. Nauels, Lagahé, E., and F. Vinet. 2014. Evaluation de la vulnérabilité des logements face à la Y. Xia, V. Bex, and P.M. Midgley, 1137–1216. Cambridge: Cambridge submersion sur l’île d’Oléron, Rapport de projet Risks, 103. Île d’Oléron: UMR University Press. https://doi.org/10.1017/ CBO9781107415324.026. Liens Université de la Rochelle, PAPI. Creton-Cazenave, L., and V. November. 2017. La gestion de crise à l’épreuve de Lazure, P., and F. Dumas. 2008. An external–internal mode coupling for a 3D l’exercice. EU SEQUANA, La documentation française, 240:237 p. hydrodynamical model for applications at regional scale (MARS). Advances in D’Ercole, R., and P. Metzger. 2009. La vulnérabilité territoriale : une nouvelle Water Resources 31: 233–250. approche des risques en milieu urbain, Cybergeo. European Journal of Le Roy, S., R. Pedreros, C. André, F. Paris, S. Lecacheux, F. Marche, and C. Geography [En ligne], Dossiers, document 447, mis en ligne le 31 mars 2009, Vinchon. 2015. Coastal flooding of urban areas by overtopping: Dynamic consulté le 14 octobre 2016. URL : https://journals.openedition.org/cybergeo/ modelling application to the Johanna storm (2008) in Gâvres (France). 22022. https://doi.org/10.4000/cybergeo.22022. Natural Hazards and Earth System Sciences 15: 2497–2510. https://doi.org/ D’Ercole, R., J.-C. Thouret, O. Dolfus, and J.-P. Asté. 1994. Les vulnérabilités des 10.5194/nhess-15-2497-2015. sociétés et des espaces urbanisés : concepts, typologie, modes d’analyse. Leone F., 2007, Caractérisation des vulnérabilités aux catastrophes « naturelles » : Revue de géographie alpine, tome 82, n°4: 87–96. contribution à une évaluation géographique multirisque (mouvements de Daniels, R.J., D.F. Kettl, and H. Kunreuther. 2011. On risk an disaster: Lessons from terrain, séismes, tsunamis, éruptions volcaniques, cyclones) – HDR de hurricane Katrina, Philadelphia: University of Pensylvania Press, 304 p. l’Université Paul Valéry Montpellier 3, 244 p. + annexes. DREAL de bassin Rhône-Méditerranée, 2016, Plan de gestion des risques Leone F., Asté J-P., Leroi E., 1996, L’évaluation de la vulnérabilité aux inondations, bassin Rhône-Méditerranée, 2016–2021, vol 2, 295 p. mouvements de terrains : pour une meilleure quantification du risque, Revue Duan, W., B. He, K. Takara, P. Luo, M. Hu, N.E. Alias, and D. Nover. 2015. Changes de géographie alpine, tome 84, n°1, pp. 35–46. https://doi.org/10.3406/rga. of precipitation amounts and extremes over Japan between 1901 and 2012 1996.3846. http://www.ano-omiv.cnrs.fr/images/Publications/PDFs/ and their connection to climate indices. Clim Dynam 45: 2273–2292. MasAvignonet/JournalManuscripts/1996-Leone_evaluation.pdf. Duan, W., B. He, K. Takara, P. Luo, M. Hu, N.E. Alias, et al. 2014. Climate change Meur-Ferec C., Flanquart H., Hellequin A.P., Rulleau B., 2011, Risk perception, a key impacts on wave characteristics along the coast of Japan from 1986 to 2012. component of systemic vulnerability of coastal zones to erosion-submersion. Journal of Coastal Research 68: 97–104. Case study on the French Mediterranean coast. Littoral 2010 - Adapting to Egli, D.S. 2013. Beyond the storms. Strengthning preparedness, Response & Global Change at the Coast: Leadership, Innovation and Investment 2011, Resilience in the 21 st century. Journal of Strategic Security 6 (2): 32–45. Royaume-Uni (2010), 8p. Flanquart, H. 2014. Perception versus représentation du risque de submersion et Nicolae, Lerma A., S. Elineau, T. Bulteau, F. Paris, P. Durand, B. Anselme, and R. autres risques : ce que révèle une querelle sémantique, In Connaissance et Pedreros. 2018. High resolution flood modeling coupling overflowing and compréhension des risques côtiers : aléas, enjeux, représentations, gestion, actes de overtopping process, framing the hazard based on historical and statistic colloque, 3-4 juillet 2014, Université de Bretagne occidentale, Brest, 353–364. approach. Natural Hazards and Earth System Sciences 18: 207–229. https://doi. Gallien, T.W., B.F. Sanders, and R.E. Flick. 2014. Urban coastal flood prediction: org/10.5194/nhess-18-2017-2018. Integrating wave overtopping, flood defenses and drainage. Coastal Nicolae, Lerma A., S. Elineau, F. Paris, Y. Balouin, S. Lecacheux, P. Durand, B. Anselme, Engineering 91: 18–28. https://doi.org/10.1016/j.coastaleng.2014.04.007. E. Longepe, S. Defossez, and Gianella L. Goeldner. 2015. Modélisation numérique Genovese, E., S. Hallegatte, and P. Dumas. 2011. Damage assessment from storm de l’aléa de submersion appliquée à l’élaboration des plans d’évacuation et à la surge to coastal cities: Lessons from the Miami area. In Advancing gestion de crise, exemple de la commune de Leucate. Ferrare: Paralia, In: Actes 3rd Geoinformation Science for a Changing World. Lecture Notes in Geoinformation Coastal And Maritime Mediterranean Conference. and Cartography, (eds). Geertman S., Reinhardt W., Toppen F. vol 1. 21–43. Pont C., 2015, Aborder et caractériser le risque de submersion marine sur les Berlin, Heidelberg: Springer. littoraux sableux méditerranéens : l’exemple de la commune de Leucate, Genovese, E., and V. Przyluski. 2013. Storm surge disaster risk management : The mémoire de master, université Paris 1 Panthéon Sorbonne, 150 p. Xynthia case study in France. Journal of Risk Research 16(7):825–841. Reghezza Zitt, M., and S. Rufat. 2015. Résiliences. Sociétés et territoires face à Goeldner-Gianella L., Chionne D., 2014, Des enquêtes sociales sur les savoirs, les l’incertitude, aux risques et aux catastrophes, 226. London: ISTE Editions. pratiques et les comportements face aux risques littoraux, Colloque Ruin I., 2010, Conduite à contre-courant et crues rapides, le conflit du quotidien international Réduire les risques littoraux et s’adapter au changement et de l’exceptionnel, Annales de la Géographie, v4, n°674, p. 419–432. climatique, La Rochelle, 2-4 Avril 2014. https://lienss.univ-larochelle.fr/IMG/ Samuels, P., Huntington S., Allsop W., Harrop W., 2008, Flood risk management : pdf/actes_colloque_la_rochelle_2-4_avril_2014.pdf. Research and Practice. Proceedings of the European conference on flood risk Guimarães, P.V., L. Farina, E. Toldo Jr., G. Diaz-Hernandez, and E. Akhmatskaya. management research into practice (Floodrisk 2008), Oxford, extended 2015. Numerical simulation of extreme wave runup during storm events in abstracts volume, 332 p. Tramandaí Beach, Rio Grande do Sul, Brazil. Coastal Engineering 95: 171–180. Slangen, A.B.A., M. Carson, C.A. Katsman, R.S.W. van de Wal, A. Köhl, L.L.A. https://doi.org/10.1016/j.coastaleng.2014.10.008. Vermeersen, and D. Stammer. 2014. Projecting twenty-first century regional Haut Comité Français pour la Défense Civile (HCFDC), 2012, Risques et menaces sea-level changes. Climatic Change 124 (1): 317–332. exceptionnels. Quelle préparation ?, Rapport d’activité 2011, 139p. Sorensen, C., M. Jebens, and T. Piontkowitz. 2017. Danish risk management plans Hellequin, A.-P., C. Meur-Férec, H. Flanquart, and B. Rulleau. 2013. Perceptions du of the EU flood directive. La Houille Blanche, n° 4: 31–39. risque de submersion marine par la population locale du littoral Stern, E. 2014. Designing Crisis Management Training and Exercises for Strategic languedocien : contribution à l’analyse de la vulnérabilité côtière. Natures Leaders. A Swedish and United States Collaborative Project., 112. Stockholm: Sciences Sociétés 21 (4): 385–399. Swedish National Defense College, Elanders Sverige AB. Durand et al. Geoenvironmental Disasters (2018) 5:19 Page 17 of 17 Thouret, J.-C., and R. D’Ercole. 1996. Vulnérabilité aux risques naturels en milieu urbain : effets, facteurs et réponses sociales. Cahiers des sciences humaines, ORSTOM 32 (2): 407–422. Ullmann, A. 2009. Changement climatique et évolution des tempêtes dans le Golfe du Lion : approche par intégration d’échelles spatio-temporelles. Cybergeo : European Journal of Geography [En ligne], Environnement, Nature, Paysage, document 441, mis en ligne le 18 mars 2009, consulté le 25 janvier 2017. URL : https://journals.openedition.org/cybergeo/22013. https://doi.org/10.4000/ cybergeo.22013. Vinchon C., Baron-Yelles N, Berthelier E., Hérivaux C., Lecacheux S., Meur-Ferec C., Pedreros R., Rey-Valette H., Rulleau B., 2011, MISEEVA: Set up of a transdisciplinary approach to assess vulnerability of the coastal zone to marine inundation at regional and local scale, within a global change context, Littoral 2010 – Adapting to Global Change at the Coast: Leadership, Innovation, and Investment, London 21-23 September 2010, 10p. Wassner T.S., Sauvagnargues, P.A. Ayral, F. Tena-Chollet, N. Frealle, 2016. Retour sur l’exercice de simulation de crise au sein du simulateur de l’école des mines d’Alès, rapport WP4 - D4 - 3a, projet ANR SPICy "Système de prévision des inondations côtières et fluviales en contexte cyclonique", 142 p. Yates-Michelin M., Le Cozannet G., Balouin Y., 2011, Etat des connaissances sur les effets potentiels du changement climatique sur les aléas côtiers en région Languedoc-Roussillon, rapport final BRGM/RP 58 872- FR, 83p., 26 III., 2 annexes. Zijlema M., Stelling G., Smit P., 2011, “SWASH: An operational public domain code for simulating wave fields and rapidly varied flows in coastal waters”. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Geoenvironmental Disasters Springer Journals

Coastal flood risk: improving operational response, a case study on the municipality of Leucate, Languedoc, France

Download PDF
17 pages

Loading next page...
 
/lp/springer-journals/coastal-flood-risk-improving-operational-response-a-case-study-on-the-KLWXRrWruh

References (61)

Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s).
Subject
Environment; Environment, general; Earth Sciences, general; Geography, general; Geoecology/Natural Processes; Natural Hazards; Environmental Science and Engineering
eISSN
2197-8670
DOI
10.1186/s40677-018-0109-1
Publisher site
See Article on Publisher Site

Abstract

Background: Climate change and rising sea level will certainly lead to significant changes in the management of low-lying coastal areas in the coming decades. While the most recent studies in the field of coastal storms-related flooding are increasingly integrated, simultaneously addressing hazards and vulnerability, as well as population risk perception, there is still little work to consider the preparedness of stakeholders to manage crises whose frequency and intensity are likely to increase in the next years. Methods: The aim of this paper is to expose the major results of the CRISSIS research program, which proposed a multi-disciplinary approach to the management of coastal flood risk in a town particularly exposed on the French Mediterranean coast. The originality of the project was to offer both an integrated approach to risk by analysing its 3 dimensions (hazard, impact and vulnerability, and representations and perceptions held by populations and stakeholders, or « risk culture ») and a very operational section focused on the evaluation of crisis management measures led by local stakeholders. To achieve this objective, two crisis exercises were organized, the first one to test the assimilation by the municipality staff of existing crisis management procedures and the second to allow local actors to integrate in their crisis management procedures the new risk knowledge data issued from research conducted under the program. Results: The project has had three main features; (i) it accurately mapped the submersible areas that present a critical vulnerability, both material and human; (ii) it revelated the poor social representation of marine submersion risk, as well as the obvious lack of awareness of crisis management systems and tools and the behavior to adopt in the event of flooding; (iii) Finally, it highlighted the need, through the crisis exercises, for a better assimilation by the municipality staff of the crisis management procedures defined in the Municipal Rescue Plan. Conclusion: The CRISSIS project has demonstrated the usefulness of an integrated and operational approach of coastal flood risk, not only in terms of studying hazards, stakes and vulnerability, but also in terms of crisis management, in particular through the organisation of crisis simulation exercises. Keywords: Coastal flood risk, Numerical modeling, Vulnerability, Risk perception, Crisis exercise, Leucate * Correspondence: brice.anselme@univ-paris1.fr Université Paris 1 Panthéon-Sorbonne, 191 rue Saint Jacques, F-75005 Paris, France Pôle de Recherche pour l’Organisation et la Diffusion de l’Information Géographique, UMR 8586 2 rue Valette, F-75005 Paris, France Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. Durand et al. Geoenvironmental Disasters (2018) 5:19 Page 2 of 17 Background of the program, with the objective of testing the PCS again With global warming causing a rise of sea levels, the but this time by integrating new data produced as part of vulnerability of coastal towns to flooding risk during the program’s findings. Overall, the objective of CRISSIS storms has increased. This issue is bound to become a has been to both generate new knowledge to improve risk key element in the management of low coastlines in de- prevention and help optimize crisis management by local cades to come, and multiple studies have been pub- stakeholders by facilitating the prioritization and classifi- lished on this subject in recent years (Cariolet et al. cation of actions in the event of a flood. After a brief intro- 2012. Vinet et al., 2012; Hurlimann and al., 2014; Duan duction of its study area, this paper will describe the et al. 2014 and Duan et al. 2015;King and al., 2014; methodology that guided its operational approach to Rulleau et al., 2015). While earlier studies were often coastal flood risk, before presenting its key findings. relatively segmented, focusing either on the hazard itself or on impact and vulnerability, more recent research pro- Study area grams on the topic have been much more integrated, sim- Theareachosenfor this studyisthe municipality of ultaneously addressing hazards and vulnerability, as well Leucate, a town located on the French coastline of the as the population’sperceptionofrisk – see for instance Languedoc region. The municipality comprises of four the French programs MISEEVA (2008–2011), JOHANNA seaside resorts located on both sides of the Leucate (2010–2013), BARCASUB (2010–2013), COCORISCO cape. These include from South to North: the resort of (2011–2014), and European studies on the management Port Leucate, built in the seventies as part of the Racine of marine flooding and submersion risk (Samuels et al. development scheme, on the coastal strip that cuts 2008;Sorensen et al. 2017). across the Salses-Leucate lake; the nudist village located The two-year CRISSIS program (Characterizing on the narrowest part of the strip; Leucate Plage, at the Submersion Risks in Sensitive Sites), launched in 2015 foot of the cape’s southernmost cliff; and La Franqui, at in a coastal town of the French region of Languedoc the foot of the cape’s northern cliff. The area is affected (municipality of Leucate), falls under this last category: by large-scale urban and touristic issues, as the town is through a multi-disciplinary approach associating ge- densely populated during the summer season, growing ographers, modelers, GIS specialists and crisis and risk from 5000 year-round residents to almost 100,000 dur- management experts, the project offers an integrated ing the high season. The urban areas of Port Leucate perspective on coastal flooding by exploring this risk’s and Leucate Plage are also very exposed to coastal flood three dimensions in coastal areas (hazard; impact and vul- hazards, and in particular Port Leucate which was built nerability; and representations and perceptions held by on a strip of low altitude coastal land (for the most part populations and stakeholders, or « risk culture »). This under 3 m NGF). Both areas are periodically flooded study does however stand out from other literature, in that when strong south-eastern storms occur (Ullmann it also includes a very operational section focusing on the 2009), during which the waves can cause the sea level evaluation of crisis management measures led by local to rise to up to 3 m NGF (Anselme et al. 2011). The stakeholders – an aspect that has rarely considered by floods can also be aggravated by a combination of research programs so far. wind and heavy rainfall, causing the lake’slevel to rise The preparedness of stakeholders to manage crises and provoking major runoff down the cliffs of the whose frequency and intensity is likely to increase is a key cape (Fig. 1). element in the management of vulnerable coastal spaces. Paradoxically, in spite of its high vulnerability to risk Several studies have shown that the damages caused by re- (the state of natural disaster has been declared 14 times in cent disasters in coastal areas have been aggravated by the the area since 1982, including 13 times for “flood/submer- local authorities’ lack of preparedness (Daniels et al. 2011; sion” events, GASPAR database of the French Ministry of Genovese et al. 2011; Genovese and Przyluski 2013). It is Ecological Transition, http://www.georisques.gouv.fr/ not enough to know and prevent risk: we also need to be acces-aux-donnees-gaspar), the town of Leucate has been prepared to manage the resulting crises (Lagadec 1993, relatively slow in applying risk regulation measures in 2001, 2004, 2012; Egli 2013). In this perspective, the comparison with other municipalities located on the CRISSIS project aims to help local stakeholders use the Languedoc-Roussillon coastline. According to a re- new findings produced through the analysis of the three cent publication of the Department of Planning and dimensions of risk. Two crisis management exercises were Housing of the Rhône-Mediterranean basin (DREAL thus organized from the outset: one took place halfway Rhône-Méditerranée 2016), the first PPRI (“Plan de through the project to test the uptake of crisis manage- Prévention des Risques Inondation”,or “Flood Risk ment procedures (« Plan Communal de Sauvegarde » Prevention Plan”) was only adopted in November (PCS), or « Municipal Rescue Plan ») by local authority 2012, while all the other municipality in the region staff; a second more ambitious test took place at the end subject to marine submersion risk had already had Durand et al. Geoenvironmental Disasters (2018) 5:19 Page 3 of 17 Fig. 1 Location of study area one for several years, and to this day this document – The hazard and its predictable evolution in the has not made it past assessment stage. A local Muni- context of climate change: digital simulations of sea cipal Safeguarding Plan (PCS) was adopted in De- level variations caused by storms or other factors, cember 2014 after the town was hit by two violent and of runoff propagation into urban areas, and storms in March 2013 and November 2014. high-resolution mapping of findings; In this context combining strong hazard and major – Vulnerability: structural vulnerability of buildings and vulnerability, the local authorities only appear to have human vulnerability of inhabitants (based on age and become aware of risk very recently. The town of Leucate mobility criteria), mapping of potential damages to therefore provides an interesting site to experiment with constructions and associated human issues; the integrated and operational marine submersion risk – Perceptions and representations held by the population management approach developed by the CRISSIS (or “risk culture”), captured through a geo-sociological program. survey aimed at quantifying risk culture levels; produc- tion of a map of findings to help improve information Methodology: An integrated and operational management and awareness-raising activities. approach to marine submersion risk The CRISSIS approach integrates the three dimensions This project aims to improve both risk prevention and of submersion risk: crisis management by local authorities. This aspect forms Durand et al. Geoenvironmental Disasters (2018) 5:19 Page 4 of 17 the fourth phase of the program: it consisted in developing The CRISSIS project therefore chose to work from a exercises to test crisis management mechanisms and pro- numerical modeling system to characterize flood haz- cedures based on existing operational documents (PCS), ard. The modeling is based on a chain of models – complemented by new data produced through our ana- MARS (Lazure and Dumas 2008), SWAN (Booij et al. lysis of the three dimensions of risk. 1999), SWASH (Zijlema et al. 2011) – in order to model the variation in sea levels caused by storm con- ditions (atmospheric pressure, wind set-up) and the Analyzing hazard using a numerical modeling system contribution of wave set-up to the rise of average sea In France, coastal flood hazards are still often approached levels, and to assess the volume of water overtopping through basic mapping methods obtained by cross-refer- the seafront and the propagation of the water on land encing the topography of exposed areas with extreme (Nicolae et al. 2015; Nicolae et al. 2018). The modeling water levels, estimated through a statistical analysis of process comprised of two stages. marine weather forcing conditions. This methodology was for instance used to elaborate official risk exposure docu- Stage 1: Testing and validating the modeling system ments (Coastal Flood Risk Prevention Plan or PPRSM) for The test consisted in simulate two recent storms the Mediterranean coastline. These basic methods are not (March 2013 and November 2014) that presented di- however suited to optimal crisis management. They only verse weather conditions and consequences in terms of produce schematic maps that do not include the event’s flood on the area of study. The all description of the kinetics, the detail of the process that caused the flood (in- validation process is detailed in Nicolae et al. 2018 and undation, overtopping of a barrier, breaching of flood only main results are presented here. Data, which was defense structures…), or the specificities of water propaga- used to recreate similar weather conditions on the site tion in urban area. Furthermore, this type of mapping as- of Leucate, is listed in Table 1. sociates the hazard with extreme statistical characteristics, Information has been collected from multiple sources which is not always compatible with the need for precise (press, photos, interviews, testimonies, report from tech- information in cases where the crisis might have been nical services, etc.) to characterize flood events (intensity, caused by a lower-intensity event. sea level, chronology). We were thus able to retrace the af- Digital modeling is now increasingly used to map out fected areas and the chronology of events. These observa- hazards, using chains of models to recreate the process tions, although mostly of a qualitative nature, enabled us that caused the flood: inundation, overtopping, breaching to estimate the level reached locally by the flood based on of flood defenses, etc. (Gallien et al. 2014; Guimarães et al. human landmarks (sidewalks, walls, quays). These points 2015; Le Roy et al. 2015; Nicolae et al. 2018). Recent im- of observation were cross-referenced with LIDAR data provements, both in terms of technical developments and and/or DGPS measurements collected through a field of pre- and post-data processing, have made it possible to campaign, in order to quantify in term of water height integrate both the phenomena that contribute to floods each qualitative observation. The reach of the flood waters (rise of average sea level, waves, river water) and their extension limit in the most heavily affected areas was also chronology (duration of floods, breaching of defense mapped out, consulting local authority staff who had structures, synchronicity of the maximum values of sea worked on the ground during the storms. In the absence and river levels). This has made it possible not only to of local records of water levels in the port, the information identify risk areas, but also to evaluate timescales and po- collected and converted in water height (in reference to tential response times between the start of an event and the French terrestrial datum, IGN69) provided a valuable the point where communication lines or strategic build- information to assess the quality and precision of the ings are damaged. Some important efforts are currently modeling of water levels and submersion in various sites being deployed using meta-modeling or machine learning Table 1 Observed and simulated data to achieve a real-time forecast of floods (Goulby et al., Data Past events 2014; Jia et al., 2015; Rohmer et al., 2016). However, this work remains too experimental to be used in an oper- March 13 Nov 14 ational setting. The most effective solution for mapping Sea Observed SHOM tide gauge in Sète SHOM tide gauge in levels Port La Nouvelle out such hazards remains combining a statistical analysis of the varying intensity of marine weather conditions with Waves Observed Candhis buoy, Leucate Candhis buoy, Leucate a field diagnosis of local vulnerabilities (coastal narrow Simulated Model IFREMER Model IFREMER dune, fragile seafront structures, water drainage system…): MEDNORD MEDNORD we can thus elaborate multiple scenarios in consultation Winds Observed France Leucate weather France Leucate with local stakeholders, and produce a tool to support station weather station decision-making in crisis situations. Durand et al. Geoenvironmental Disasters (2018) 5:19 Page 5 of 17 across the municipality. The water levels generated As part of this research project in Leucate, the diagnosis through the simulations were compared with those de- of material and human vulnerability was evaluated on duced from the analysis of topographic landmarks (quays, multiple scales. We first assessed the vulnerability of ma- roads) that appeared on photographs to be submersed or terial goods exposed to hazards, considering that vulner- not by the floods. By looking at various landmarks across ability also affects the rest of the area (D’Ercole and the port area, we established that the average water level Metzger 2009). In addition, we carried out a comprehen- in the port was 0.85 m IGN +/− 5 cm IGN in 2013, and sive diagnosis of material and human vulnerability, consid- 1.05 m IGN +/− 5 cm in 2014. The water levels generated ering the dangerousness of each building located in a by simulations in the area were within a similar range to high-risk area in the municipality. Within this sample, we those estimated from photographic observations (a dis- carried out a diagnosis of first homes, second homes and crepancy of under 5 cm was observed for the 2013 simula- tourist accommodation (about 4000 buildings) as well as tion, and an underestimation of about 10 cm for 2014), commercial premises (about 100). However, to further which indicated that the performance of the modeling sys- elaborate our vulnerability criteria – and more specifically tem was relatively reliable. those relevant to human vulnerability – a diagnosis was drawn based on a perception survey (see below). Add- itional evaluation criteria were used for this sample, but Stage 2: Elaboration of submersion scenarios they only covered 400 buildings (tourists who do not own After validating the modeling system, we developed various their accommodation were excluded from the sample). submersion scenarios, based on a multivariate statistical The criteria were set to capture the potential damages analysis of extreme values (joint probability of waves/water caused by coastal floods, assessing both material vulner- levels). This analysis enabled us to define low-probability ability (type and condition of building) and human vulner- marine conditions (with a return period of 100 years), ability with indicators assessing people’s access to safety. whichwerepropagatedtothe shoreand then to theland The advantage of using two different samples was that (Nicolae et al. 2018). These simulations enabled us to they provided an operational diagnosis across the area af- characterize the reach and intensity of floods for hundred- fected by risk. This also helped us respond to the expecta- year events, and to anticipate the consequences of a gradual tions of local authorities, by showing that a more in-depth rise of the average sea level under the effect of global warm- diagnosis could be carried out to improve crisis manage- ing. This data was fed into a GIS in order to create relevant ment by identifying and locating vulnerable individuals. maps for crisis management exercises. The surveys included questions based on vulnerability cri- teria and indicators (extract: see Table 2). The methodology Assessing the vulnerability of buildings and populations used to assess vulnerability was adapted to the local context Over the past few years, the geographic understanding of of the municipality of Leucate, using evaluation criteria that natural hazards shifted from a hazard-centered approach had been designed for other locations and hazards and had to a vulnerability-centered approach, integrating the social been proven to be relevant to this site (Leone 2007; dimension of hazard. Such a diagnosis and evaluation of Meur-Ferec et al. 2011; Lagahé and Vinet 2014). vulnerability presents advantages in terms of anticipating Some criteria capture both material and human risk and mitigating its impacts. Although this notion has vulnerability, while others only capture one type of evolved, vulnerability can be understood both in its pri- vulnerability. mary meaning as the capacity to withstand damage (D’Er- A vulnerability index was developed from these criteria. cole et al. 1994) but also as the capacity to cope with When taken individually, a criterion only captures a par- damage (Thouret and D’Ercole 1996), thus echoing the tial assessment of one specific type of vulnerability. How- concept of resilience. Resilience is perceived as the posi- ever these criteria and indicators can be combined to tive counterpart to vulnerability, which usually carries provide a global assessment of vulnerability, where some negative connotations due to its association with fragility, criteria are compensated by others. although the two terms are not mutually exclusive The methodology followed to create this index con- (Reghezza Zitt and Rufat 2015). sisted in setting a hierarchy between indicators and cri- This study considers the geographic dimension of risk on teria, by using an intra- and inter-criteria weighing an infra-municipal scale, with a vulnerability diagnosis fo- system. The hierarchy was based on the bibliography cusing on individual issues (on the scale of a building). The and on past studies on this topic, as well as on a global notion of vulnerability implies an assessment-centered ap- reflection on theimportanceofagivencriteriaincom- proach (Leone et al. 1996;Becerra 2012), which must begin parison with another, always from the perspective of a with the identification of criteria and indicators. Many re- scenario of potential damage. We were thus able to cent studies on the subject show the relevance of this meth- elaborate maps that were fed into the development of odology (Barroca et al. 2005;Leone 2007). crisis exercise scenarios. Durand et al. Geoenvironmental Disasters (2018) 5:19 Page 6 of 17 Table 2 Extract from list of vulnerability criteria and indicators applied to Leucate (the sample of individuals surveyed) Inter-criteria Criteria Indicators Intra-criterion weighting weighting 0to6 Type of property Multi-storey accommodation; single-storey with roof exit; single-storey 0 to 8 Material depending on without roof exit. vulnerability criteria Structure and condition Age of accommodation; floor level; types of materials; waterproofing 0to 4 of property of openings; height of electrical connections. Type and structure Multi-storey accommodation; single-storey with roof exit; single-storey 0 to 8 Human of property without roof exit; ground floor level; number of bedrooms and inhabitants vulnerability Accessibility Type of access road; emergency exits 0 to 4 Characteristics of Number of persons; under 10 years old; over 60 years old; over 0to 3 population 80 years old; persons with reduced mobility and/or dependent persons Perception analysis and representations of risk In order to measure risk culture levels in the munici- An understanding of the perceptions, behaviors and re- pality of Leucate, the methodology focused on three actions of individuals and groups of individuals in the main angles: questionnaire-based surveys of inhabi- face of any kind of risk should form a prerequisite to tants; semi-structured interviews with key stakeholders; the development of prevention and management pol- and a discourse analysis of the local news bulletin, Cap icies (Ruin 2010). This process can help researchers Leucate. In total, we screened 46 issues ranging from better grasp the discrepancy between the knowledge of January 2010 to June 2015 (the bulletin is monthly in experts and that of laypersons (Baggio and Rouquette theory), which underwent an assisted textual analysis 2006). The perception of risk varies between individ- using the Tropes software (Pont 2015). The term uals: studies on this topic attempt to characterize “submersion” (“submersion”) appears in the newsletter groups of individuals that share a relatively similar per- for the first time in the March 2015 issue, to refer to ception of a given risk (Hellequin et al. 2013). An indi- the municipal rescue plan established by the local au- vidual’s perception of risk is influenced by four factors: thorities, and in an article that announces the launch of their experience of risk (through direct or indirect ex- the CRISSIS research project. Before the creation of of- perience), their knowledge on this risk (lay or aca- ficial documents (PCS and PPRL), municipal authorities demic), their social and economic interests (attachment would use the word “inondation” (“inundation”)when to a place, value of property, socio-economic standing referring to coastal floods. The word “inondation” was of household, etc.), and their values (moral, political, written sixteen times, including three times during the etc.). Over the past twenty years, the analysis of percep- 2013 flood that hit Leucate Plage, and three times in tions and social representations has played an increasingly 2015 in relation to the announcement of this research prominent part in research programs on French coastal program. towns’ vulnerability to marine hazards (Goeldner-Gianella The questionnaire survey was communicated to 493 and Chionne 2014; 2002–2004: PNEC, Nord-Pas-de-Calais; people including first home owners, holiday home owners 2007–2011: ANR MISEEVA, Languedoc - Roussillon; and tourists living in Port Leucate, the naturist village and 2011–2014: ANR COCORISCO, Bretagne, etc.). These Leucate Plage (Table 3). The survey was carried out from studies have highlighted the weakness of social repre- april to june 2015. At this time of year, many holiday sentations of coastal risks (erosion and marine submer- homes are unoccupied, making it easier for us to survey sion) and even the “blatant absence of a risk culture” residents and secondary home owners. The surveys were (Chauveau et al. 2011). Most surveys show that the conducted face-to-face at the respondents’ homes and populations affected feel little concern for coastal risks. lasted 20 to 40 min. The people present were rather suspi- They do not reflect on their situation, and nor do they cious, we had several refusals. Overall, the questionnaires actively seek out information or design any action plans were well completed, with respondents answering 93% of in the event of a disaster (Hellequin et al. 2013; the questions. It would have been useful to base our popu- Flanquart 2014;Krien 2014). When questioned, the lation and location sampling on the modeling of submers- people surveyed tend to make up a spontaneous dis- ible areas and of the vulnerability of buildings, which were course on the question by drawing from their memories, described above. However, as the duration of the pro- sensations and incomplete reflections on the question. gram was limited to two years, we had to carry out all Krien (2014) explains that people’s representation of research activities simultaneously, and were not able to coastal risks is the product of a social construct developed use this sampling methodology. Individuals were sur- from their overall representation of risk, the sea, storms veyed on: (1) the appeal of the Leucate municipality and the place where they live. and its leisure activities; (2) their perception of the Durand et al. Geoenvironmental Disasters (2018) 5:19 Page 7 of 17 Table 3 Breakdown of surveys in Leucate based on area of residence and residential profile of individual surveyed Resident profile / First home owners Second home owners Tourists Total surveyed area Leucate Plage 2% (n = 12) 4% (n = 18) 9% (n = 42) 15% (n = 72) Naturist village 5% (n = 23) 4% (n = 20) 1% (n =3) 9% (n = 46) Port Leucate 32% (n = 158) 36% (n = 174) 9% (n = 43) 76% (n = 375) Total 39% (n = 193) 43% (n = 212) 18% (n = 88) 100% (n = 493) area’s and of their accommodation’s exposure to coastal work effectively and efficiently together to prepare for, flood risk; (3) their awareness of this risk in Leucate; respond to, and recover from domestic incidents.. To (4) their presumed reaction in the event of a submer- provide for interoperability and compatibility among sion; and (5) their expectations in terms of communica- Federal, State, and local capabilities, the NIMS will in- tion about this risk. The final objective was to map out clude a core set of concepts, principles, terminology, representations of risk, in order to help shape the local and technologies covering multi-agency coordination authorities’ information and awareness-raising activities systems; unified command; training; identification and on coastal flood risk. management of resources ; qualifications and certification; and the collection, tracking, and reporting of incident Testing crisis management procedures through crisis information and incident resources” (Anneli 2006). To exercises regularly train the stakeholders, the FEMA (Federal Crisis exercises: A brief state of the art Emergency Management Agency) has a National Exercise The last stage in the CRISSIS program’s integrated ap- Program, https://www.fema.gov/ned) which regularly runs proach brought added operational value to the project. large-scale drills simulating the occurrence of cyclones, Recent research on risk assessment in coastal areas – for earthquakes and tsunamis. In Europe, this task is covered instance in the Languedoc Roussillon region, the projects by the European Civil Protection Mechanism (ECPM, RNACC (“Risques Naturels, Assurances et Changement http://ec.europa.eu/echo/what/civil-protection/mechanis- Climatique” [“Natural Risk, Insurance and Climate m_en) which organizes and funds a series of drills in the Change”], Yates-Michelin et al. 2011)and ANR MISEEVA field of civil protection in different member countries (2008–2011, Vinchon et al. 2011; Meur-Férec et al., 2011) every year. The French example provides a good gauge for – did not go as far as integrating crisis management. And a national policy for crisis drills. With a concern for opti- yet the preparedness of local authorities is essential. Any mizing the response from stakeholders and the public in a failings on their part can significantly amplify the impact crisis, the 2004 law on the modernization of civil security of extreme phenomena in coastal areas, turning a crisis (law #2004–811 dated 13 August 2004) makes regular ex- into a full-blown disaster (Daniels et al. 2011;Genovese et ercises compulsory, with a requirement for large-scale al. 2011). This is particularly true in regions where the drills several times a year at national and regional levels stakes are high and diverse, and where a crisis can involve (Richter-type earthquake drills are a good example) and a multiple public and private stakeholders. Local players requirement for all towns and villages where a major nat- should therefore be trained in crisis management, includ- ural or technological risk has been identified to conduct at ing by organizing crisis exercises (Lagadec 1993, 2001, least one crisis drill per year (DGSCGC or Direction 2012;Stern 2014). Générale de la Sécurité Civile et de la Gestion des Crises, To meet this challenge, public authorities in many 2008). According to the guidelines set by the General countries have developed emergency management sys- Directorate for Civil Security and Crisis Management tems for using by stakeholders and organize regularly (DGSCGC), a body whose equivalent can be found in large-scale crisis exercises to prepare institutions and most countries, there are two main types of exercises populations to cope with major crises. Thus, in the (Institut National des Hautes Etudes de la Sécurité et de la USA, following huge forest fires in California and Justice (INHESJ) 2015): “table-top exercises” (“exercices Arizona in the 1970s, which had highlighted a lack of cadres”)and “field conditions exercises” (“exercices ter- coordination of emergency response between different rain”). The former are desk-based and take place at the actors, several states have developed the Incident crisis headquarters. They do not involve the deployment Command System (ICS), which, after the attacks of of any resources on the ground. Participants receive infor- September 11, 2001, was extended to all US states and mation by radio, phone, fax, television, SMS and the social integrated into the National Incident Management networks. They are required to analyze, and synthetize System (NIMS). NIMS is “a consistent nationwide ap- this information, react, report to others, make proposi- proach for Federal State, and local governments to tions, set priorities and make choices to manage the crisis Durand et al. Geoenvironmental Disasters (2018) 5:19 Page 8 of 17 for the better. The second type involves the deployment of analysis, vulnerability and population perceptions/repre- men and equipment on the ground. One of its objectives sentations. The training followed the format of a frame- is to test the transportation and deployment of equipment work exercise, which was organized in Paris by students in real-life and real-time conditions. Good examples of from the Master’sdegree “Gestion Globale des Risques et such activities include the exercises organized periodically des Crises” (“Global Risk and Crisis Management”,or in the US by the FEMA or those organized by area prefec- GGRC), working from a low-intensity crisis scenario. In tures in France. For instance, the European Sequana 2016 practice, the exercise ran over three hours and the partici- exercise, held in March 2016 and which simulated the oc- pants broke out into two units: (i) a “facilitation” unit currence of a 100-year flood of the Seine in Paris, involv- where the participants played the parts of Préfecture staff, ing public stakeholders at various levels (national, zonal, firefighters, security forces, the media and the population, departmental and municipal) and an array of private oper- who were responsible for sending “inputs” based on the ators (transport firms, telecommunications companies, chosen scenario to (ii) the crisis unit that simulated the banks, supermarkets, hospitals, and so on). On the one work of the municipal command team (PCC). hand, the goal of this exercise was to test Paris area re- Following this first training session, a second, more am- gional stakeholders’ ability to coordinate and to cope with bitious exercise took place locally a year later, in March response and crisis management. On the other hand, that 2016 (duration: 3 h). This time, the exercise involved all test was also an opportunity to assess the relevance of op- the municipal staff including elected members of the erators’ crisis management plans as well as the consistency Council, as well as senior representatives of the local au- of operational procedures. Eventually, public authorities thorities: the Préfecture of the Aude département (where were also provided with a feedback and the scope of infor- Leucate is located), département managers of the fire bri- mation for population was measured for ensuring their gade and the DDTM (“Direction de l’Aménagement et du awareness can be improved in the event of a disaster. The Territoire”,or “Planning and Territorial Directorate”, a de- exercise uncovered two points to improve in the emer- volved State body in charge of risk prevention). This time, gency response: the coordination of public and private ac- the session included both framework and field activities. tors, very heterogeneous according to the sectors of Some elements were performed in a real-life setting activity, and the information of the population, which (on-the-ground deployment of technical staff and munici- sometimes lacked efficiency (some instructions have not pal police forces) to test the coordination of the municipal been well assimilated) (Creton-Cazenave and November crisis unit (municipal command team, or PCC), with re- 2017). However, in France, in spite of these efforts, crisis ex- sources deployed on the ground. However, because of fi- ercises remain all too rare, especially at local level. In its nancial and public security constraints, it was not possible 2012 activity report, the French Committee for Civil to perform the entire exercise in real-life conditions: for Defense (Haut Comité Français pour la Défense Civile instance, civil safety (firefighters) and security forces (gen- (HCFDC) 2012) pointed out that in France “crisis man- darmerie) were not involved in real life but fictionally. Fa- agement procedures are too often untested” for cilitators played the part of the media (television, press), high-intensity events. In addition, such exercises, whether the social networks, the population affected by the crisis they are “table-top” or “in real conditions” exercises, take and the private operators involved. place at the initiative of major public institutions (govern- This time, the scenario included new data (on hazard ment agencies, prefectures, civil protection departments) and vulnerability) acquired during the project’sprevious but are rarely if ever initiated by the research sector. stages. The main objective was to test the “marine sub- In this context, the CRISSIS program aimed to test the mersion” section of the PCS to which detailed specific municipality’s current crisis management procedures (set in and intermediary objectives had been added, as shown the December 2013 PCS) and help optimize them through in Table 4. two crisis exercises involving local stakeholders, organized The scenario was entirely designed by the students of and designed according to a principle of progressivity. the GGRC Master’s degree from Université Paris 1, under the supervision of crisis exercise specialists, and Methodology of program exercises with the assistance of two senior members of municipal The first exercise took place at the start of the program in staff (the Head of environment and a Municipal Police March 2015. It consisted in a simple training session manager) as well as a manager from the Aude préfecture aimed at informing the municipality’ssenior administra- (deputy risk manager) who brought their field experi- tive staff (Head of municipal police, Head of technical de- ence on board and helped develop credible “inputs”. The partment, Director General of administration) of the main danger when creating such exercises is that participants procedures and mechanisms listed in the Municipal might in retrospect challenge their credibility – for in- Rescue Plan, which had recently been adopted. This ses- stance because they consider that their impact was too sion did not include the program’sfindings onhazard low or too high, or too remote from real-life situations. Durand et al. Geoenvironmental Disasters (2018) 5:19 Page 9 of 17 Table 4 Extract from the objectives of the March 2016 exercise Defense and Civil Protection (Service Interdépartemen- tal de Défense et de Protection Civile, SIDPC) of the Global objective Intermediary objectives Specific objectives Aude Préfecture in Carcassone (sent out emergency noti- Testing the marine Testing the SAIP Testing the logbook submersion section (SIRENS) / SMS Alerts fications from the Préfecture, responded to the munici- of the PCS pality’s requests for additional resources); (ii) at Understanding the PCS Testing the relevance and its role in the crisis and proficiency of the municipal level, from a site in Leucate, by the students room crisis room’s tools from the GGRC Master’s and by the town’s two contact Confirming the crisis Testing communication persons (sent out all other “inputs”: reports of local management sites damage caused by the storm, requests from the popula- Testing the municipality’s Testing the power tion, media queries, pressure on the social networks, organization in the event generator etc.). The crisis was simulated on the evening of Satur- of a crisis (staff, communications, etc.) day 9 April 2016 – that is, over a weekend during mid-tourist season (the week-end before a major sport- Testing the municipal Testing connections staff’s knowledge and (phone, radio) ing event organized in Leucate, the “Mondial du Vent”). understanding of During this period the town usually hosts 8000 to the PCS 10,000 people (for a permanent resident population of 5000) staying in secondary homes, campsites and If this had been the case, the exercise would not have camper van parks. Figure 2 shows an overview of the achieved its learning objective of consolidating the mu- roll-out of the three-hour exercise, which has also been nicipal staff’s knowledge of crisis management tools and documented in a film that is available on the program procedures, and where possible helping improve them. It website (http://crissis2015.free.fr/). was therefore essential for this scenario to be developed in collaboration with individuals who had an excellent Results and discussion knowledge of local realities and institutions. Production of high-resolution marine submersion maps Stimuli were produced on two levels: (i) at Préfecture The validation of the digital modeling system and the level, by staff based at the Interdepartmental Service of multivariate statistical analysis of extreme values on record Fig. 2 Flowchart of the roll-out of the crisis exercise, March 2016. Source: ©Master GGRC, 2016 (PCC: municipal post of command; PREDICT: consultancy, service provider for the local authority; DEBEX: start of exercise; FINEX: end of exercise) Durand et al. Geoenvironmental Disasters (2018) 5:19 Page 10 of 17 (see above) enabled us to elaborate several submersion sce- Mapping out critical vulnerability levels narios. Simulations were run to characterize the spread and The analysis of vulnerability using the method described intensity of submersions for a hundred-year event (Nicolae above enabled us to establish a vulnerability index (see et al. 2018), and to forecast the consequences of a gradual Table II), which was mapped out to highlight the most rise of the average sea level under the effect of climate vulnerable buildings and populations. For Leucate, three change. For this purpose, we estimated the impact of a categories were defined and prioritized, ranging from low minor rise in sea level working from a variation of + 0.2 m to medium and high vulnerability (Figs. 4 and 5). (the average global rise forecasted according to the median Spatial mapping makes it possible to draw compari- scenario for the 2046–2065periodincomparisonwith sons within one single entity, but also between entities 1986–2005 – source: IPCC WG1 Ch13, Church et al. 2013) themselves, highlighting in this case a variation in and + 0.6 m (the average sea level rise forecasted in the levels of material vulnerability between areas. After Mediterranean by 2100, Slangen et al. 2014). assessing vulnerability, this ongoing study aims to rec- For the March 2016 crisis exercise, our consultation ommend strategies to adapt buildings depending on with the two members of municipal staff who contrib- their exposure to risk. This diagnosis will be expanded uted to the script drove us to simulate a coastal flood to the totality of buildings in risk areas. By globalizing caused by a twenty-year storm, which combined the vulnerability, we will thus be able to assess not only characteristics of the last two major storms (March material vulnerability but also territorial vulnerability 2013 and November 2014) in terms of wave height and (D’Ercole and Metzger 2009). sea levels – the former had caused the breach of a sea- These maps of vulnerability were integrated to the front wall in Leucate Plage, and the latter had provoked PCS as appendixes, and made available to participants a flood by overtopping on the seafront and overflowing for the second exercise in May 2016. Their purpose in Port Leucate (Fig. 3). The objective was therefore to was, where possible, to inform the first protection engage the municipal staff in a role play using a sce- measures taken by municipal staff during the crisis nario that combined events they might already have en- (for instance, setting up coffer dams), as well as the countered, but separately. This helped ensure that the organization of evacuation. scenario had enough impact to place the players under pressure, while not being too “overblown”, which could A poor risk culture have been the case with a simulation of a hundred-year Our analysis of perceptions and representations showed storm that factored in the rise of the average sea level. a poor awareness of coastal flood risk amongst surveyed Fig. 3 Submersion scenario, a Leucate Plage, b naturist village, c Port Leucate, created from the digital model and used for the March 2016 field exercise Durand et al. Geoenvironmental Disasters (2018) 5:19 Page 11 of 17 Fig. 4 Material vulnerability of people surveyed in Leucate (source: field survey) populations. For instance, of all residents and tourists existence of this system – and this despite the fact that surveyed in areas that were shown by our model to be of all second home residents surveyed, only 29% de- potentially submersible, only 50% of people surveyed de- clared they were never present in Leucate between Octo- clared that they were living in a risk area (Fig. 6). ber and March (the high-risk period for coastal floods). 43% of first and second homeowners declared that the Consequently, it appears necessary to implement a com- municipality of Leucate was not threatened by coastal munication strategy to inform the population, and in par- floods. This finding is combined with an underestima- ticular second-home owners, of existing emergency alerts. tion of risk. In comments recorded as annexes to the Although the findings of this study of perceptions/repre- questions, some people who had been informed of past sentations were not used in the March 2016 exercise, they marine submersion events explained that they did not informed the recommendations drafted in our exercise consider the few intrusions of sea water into the streets debrief notes to improve the municipality’s information, and houses as floods. According to these people, the awareness raising and communication strategy. hazard does not exist until there are human casualties, and material damage is considered unimportant and re- mediable. We also sought to find out whether residents Lessons learned from the exercise debrief: A poor were aware of the systems set up by the municipality to integration of crisis management systems and procedures alert them in the event of a risk, including the volunteer-run SMS alert system outlined in the PCS. On An exercise debrief includes a methodical analysis of this point, we found a significant difference between first the exercise, to highlight its strong points and home and second home owners: 34% of the former had improvement points in order to perfect the communicated their contact details to the local author- organization’s crisis management processes. In this ities, while 74% of the latter were unaware of the case, the scope of the lessons learned went beyond a Durand et al. Geoenvironmental Disasters (2018) 5:19 Page 12 of 17 Fig. 5 Location of vulnerable populations based on age and health status (source: field surveys) sole technical fixing of failing tools or processes. The crisis management. Following these observations, the aim was to question individual and organizational municipality of Leucate introduced geo-referenced responses to extreme event that might pose a challenge flood markers into its PCS and committed to fully par- to the system (Lagadec 1993). ticipating in the second exercise to improve municipal staff’s command of the PCS. Our debrief from the first exercise (March 2015) re- The debrief for the second exercise (March 2016) vealed a poor knowledge of the Municipal Rescue Plan was captured in a report that was presented in early (PCS) on the part of municipal staff, and helped us May 2016 to all municipal staff. It highlighted the mu- present a few observations on its contents. For in- nicipal teams’ excellent field knowledge, as well as a stance, the PCS had no objective quantifiable mile- good coordination between the municipal post of stones (in meters) regarding the water levels attained in command (PCC) and the teams deployed on the submersed areas: the decision to activate the successive ground. However, it also revealed the municipal staff’s phases of the PCS (closing roadways, evacuating certain obvious lack of knowledge of the PCS: most staff had neighborhoods) were only based on field information no clear understanding of their respective roles and communicated by technical staff who have been work- responsibilities within the crisis unit, and therefore ing for the municipality for years and therefore have an faced difficulties in organizing and coordinating the in-depth knowledge of local sites. However this lack of PCC. Furthermore, these challenges were aggravated quantifiable milestones can be a disadvantage for inex- by the crisis headquarters’ lack of practicality: the perienced staff (for instance people who are new in functions of secretariat, command, logistics and co- post, or covering for more experienced permanent staff ordination were all gathered in a single room, which members who might be off work when a crisis occurs), caused a great deal of confusion. In this context, the leading to damaging delays in a context of emergency fear of an error was a major source of stress, causing Durand et al. Geoenvironmental Disasters (2018) 5:19 Page 13 of 17 Fig. 6 Perceptions of residents on their property’s exposure to marine submersion risk communication challenges both internally within the (iii)Writing a checklist to be inserted at the start of the crisis unit (absence of situation updates) and exter- PCS, outlining the key principles of crisis nally (communication with Préfecture services and the management: sharing the logbook; updating the media). Consequently, the debrief report included four schedule and map in real time; regular situation main recommendations: updates, and communication of human casualties to the higher echelons (Préfecture), etc. (i) Reorganizing the PCC according to the traditional (iv) In addition, the municipality should improve its organizational structure of a crisis unit, to facilitate communication with the population ahead of the communication and decision-making in the event of crisis (for instance via the local news bulletin) on a crisis. The space should be split into five open current crisis alert systems and in particular the SMS units (Fig. 7): decision-making and coordination alert (this recommendation directly derives from the unit; situation unit (secretariat/log); logistical unit; analysis of perceptions/representations of risk). communication unit; forecasting unit) (ii) Drafting concise post descriptions to be included Discussion: What are the obstacles faced by an into the PCS, so that from the moment of their operational approach to coastal flood risk? arrival into the PCC each member of the crisis The resolutely operational approach adopted by the unit knows what their role is and what they are CRISSIS program faced three challenges. First of all, it has expected to do. Examples of post descriptions: been difficult to convince local representatives to organize crisis unit Director (DOS); unit Coordinator a crisis exercise locally. While the first exercise, which (needs to be very mobile and move between took place in Paris in March 2015, did not pose any major units); Secretary (keeping a log, handling calls); issues as its stakes were lower (low-intensity exercise, re- communication unit / forecasting unit / logistical mote location, framework exercise only, involving only unit Managers. two technical staff members and the Director General of Durand et al. Geoenvironmental Disasters (2018) 5:19 Page 14 of 17 Fig. 7 Example of proposed reorganization of the PCC suggested in the exercise debrief report municipal services as observers), organizing the second had received from hydro-meteorological experts Predict proved markedly more complex. As this exercise needed (the municipality’s current service provider) were used to take place locally and involve large numbers of Council properly. However, this was not the case with the maps members and municipal staff, we were initially faced with showing people and buildings’ critical vulnerability levels, a clear reluctance on their part. Eventually, only two which could have informed the first protection measures Council members (two Deputy Mayors) did take part in (for instance, setting up coffer dams) taken by personnel the exercise, along with about twenty administrative and on the ground. This challenge shows the necessity of pro- technical staff members. The 2004 Law for the viding the PCC staff with user-friendly documents that modernization of civil security (Act 2004–811 of 13 can be read quickly. Similar observations had been made August 2004) states that any municipality where at least during the crisis exercise organized for a group of munici- one major natural or technological risk has been identified palities in La Réunion by students from the École des must organize at least one crisis exercise per year Mines in Alès as part of the only other French program (DGSCGC, 2008). However, as highlighted by the French on submersion risk (Wassner et al. 2016)thatalsoin- Committee for Civil Defense in its 2012 report (Haut cluded an operational dimension and a crisis exercise. Comité Français pour la Défense Civile (HCFDC) 2012), Finally, the actual adoption of the recommendations very few municipalities actually comply with this obliga- presented in the exercise debrief report is a very tion. This reluctance may be due to the lack of municipal long-term process. Although these recommendations resources, or perhaps to the fear of being evaluated and were approved by the municipality when the report judged. In this respect, a lot of educational work is needed was first presented to them, over a year on only the to convince potential participants that the objective of first has led to concrete action being taken (the such exercises is not to evaluate individuals but systems reorganization of the PCC). The PCS has still not been and procedures, in order to improve them and make them modified. Besides, this plan has never been tested in easier to memorize. Only when the exercise was com- real-life conditions (as the last storm occurred in No- pleted did participants finally drop their guard and admit vember 2014, before its adoption): it still remains un- to its usefulness. sure whether the municipality does have the capacity In addition, municipal staff made a variable use of the to tackle a high-intensity storm. The most “extreme” operational documents (maps) presenting the findings of scenarios tested as part of the project’shazard analysis the analysis of hazards and vulnerability. The submersion show for instance that the spread and volume of the forecast maps that were provided to the PCC coordinator flood would increase by respectively 160% and 188% (Head of the municipal police) at various points during in the event of a hundred-year storm with a 0.6 rise of the crisis, in replacement for the maps he would normally sea level (Nicolae et al. 2018, Anselme et al. 2017). Just Durand et al. Geoenvironmental Disasters (2018) 5:19 Page 15 of 17 as it is necessary to implement a robust prevention Funding This research work was supported by the French Conseil Supérieur de la policy (the Submersion Risk Prevention Plan for Formation et de la Recherche Stratégiques (CSFRS), within the framework of Leucate was approved in January 2017), improving cri- the call for non-thematic projects 2012. The funding body has played no role sis management procedures should also be a priority in the design of the study, the collection, analysis and interpretation of data, or the writing of the manuscript. for the municipality in the years to come. Authors’ contributions BA CRISSIS program manager, manuscript’s redaction: intro, conclusion, study area, figures design and layout. PD preparation of crisis simulation exercises, Conclusion manuscript’s redaction: general structure of the paper, intro, conclusion, The initial objective of the CRISSIS program was to method, results and discussion (crisis exercises). ALN and SE numerical improve both the anticipation of risk and the imple- modeling, manuscript’s redaction: method, results and discussion (hazard). SD and MG vulnerability survey, vulnerability index design, manuscript’sredaction: mentation of adapted responses to better tackle its un- method, results and discussion (vulnerability mapping). LG-G and EL survey’s predictability and help with decision-making, in the conception, survey in the field, statistical treatment, survey’s analysis method, context of a highly urbanized municipality that is peri- results and discussion (risk perceptions and representations). All authors read and approved the final manuscript. odically impacted by sea water floods (inundation, bar- rier overtopping, breaching of barriers). To achieve Ethics approval and consent to participate this, our research focused simultaneously on the nat- Not applicable ural (hazard), material and social (vulnerability, per- Consent for publication ception/representations) dimensions of risk. However, Not applicable the objective was also to engage with local authorities Competing interests by inviting them to work on the operational aspect of The authors declare that they have no competing interests. crisis management, which is usually neglected by re- search programs on flood risk. In this perspective, we Publisher’sNote worked to improve the various stakeholders’ oper- Springer Nature remains neutral with regard to jurisdictional claims in published ational response by using crisis exercises to assess the maps and institutional affiliations. degree to which they had adopted the tools and sys- Author details tems placed at their disposal (PCC, PCS), but also by Université Paris 1 Panthéon-Sorbonne, 191 rue Saint Jacques, F-75005 Paris, helping improve these tools and systems using the France. Laboratoire de Géographie Physique, UMR 85911 place Aristide Briand cedex, F-92195 Meudon, France. Pôle de Recherche pour new findings produced by the program’sfirst three l’Organisation et la Diffusion de l’Information Géographique, UMR 8586 2 rue sections. Valette, F-75005 Paris, France. Université Paul Valéry, 2 rue du Professeur Overall, this project helped improve our understand- Henri Serre, 34080 Montpellier, France. UMR GRED – IRD (Gouvernance, Risque, Environnement, Développement), route de Mende, 34199 ing of submersible areas and create maps of critical vul- Montpellier, France. BRGM, 3 Avenue Claude Guillemin, 45100 Orléans, nerability, both material and human. It also revealed France. Centre universitaire de formation et de recherche de Mayotte, the poor social representation of marine submersion Dembéni, Mayotte. UMR Espace Dev, 500 rue JF Breton cedex 5, 34093 Montpellier, France. risk, as well as a clear lack of awareness of crisis man- agement systems and tools, and of behaviors that Received: 4 May 2018 Accepted: 3 October 2018 should be adopted in the event of a flood. These find- ings highlight the need for setting up a communication References strategy, to raise awareness of risk and inform the Anneli, J.F. 2006. 2006, the National Incident Management System: A multi- population of current alert mechanisms. The crisis ex- agency approach to emergency response in the United States of America. Rev. sci. tech. Off. int. Epiz. 25 (1): 223–231. ercise conducted in March 2016 with relevant munici- Anselme, B., P. Durand, Y.F. Thomas, and A. Nicolae-Lerma. 2011. Storm extreme pal departments showed the importance of creating levels and coastal flood hazards. A parametric approach on the french coast of such exercises working from a realistic scenario, to en- languedoc (district of Leucate). Comptes Rendus Geosciences 343 (10): 677–690. Anselme, B., P. Durand, Y.F. Thomas, and A. Nicolae-Lerma. 2017. Coastal flood sure that all staff in positions of responsibility are aware Hazard. Risk and crisis management : Questions raised by sea level rise ? 10–14. of procedures, but also to detect any potential gaps in New York: International WRCP/IOC Conference, (poster) Columbia University. the contents of the PCS. In a context where hazard and Baggio, S., and M.-L. Rouquette. 2006. La représentation sociale de l’inondation : influence croisée de la proximité au risque et de l’importance de l’enjeu. vulnerabilityare boundtobecomemoreextreme,this Bulletin de Psychologie 59 (1): 103–117. should be a key action in the necessary optimization of Barroca B., Pottier N., Lefort E., 2005, Analyse et évaluation de la vulnérabilité aux crisis management systems and procedures. inondations du bassin de l’Orge aval. Actes des septièmes rencontres de TheoQuant, Atelier 3 « Risques, vulnérabilité », Besançon, 12p. Becerra, S. 2012. Vulnérabilité, risques et environnement : l’itinéraire chaotique d’un paradigme sociologique contemporain. VertigO - la revue électronique en Acknowledgements sciences de l’environnement [En ligne] 12 (1). URL : http://vertigo.revues.org/ We would like to thank the Conseil Supérieur de la Formation et de la 11988). https://doi.org/10.4000/vertigo.11988. Recherche Stratégiques (CSFRS) for funding this research program. Our Booij, N., R.C. Ris, and L.H. Holthuijsen. 1999. A third-generation wave model for thanks also go to the municipality of Leucate who volunteered to take part coastal regions, part I: Model description and validation. Journal of in this program and greatly facilitated our work on the ground. Geophysical Research 104 (C4): 7649–7666. Durand et al. Geoenvironmental Disasters (2018) 5:19 Page 16 of 17 Cariolet, J.-M., S. Suanez, C. Meur-Férec, and A. Postec. 2012, 2012. URL : https:// Institut National des Hautes Etudes de la Sécurité et de la Justice (INHESJ), 2015. journals.openedition.org/cybergeo/25077. Cartographie de l’aléa de La gestion de crise en Europe. In Vers une coexistence des organisations submersion marine et PPR : éléments de réflexion à partir de l’analyse de la actuelles basée sur la culture des états membres ou vers une convergence commune de Guissény (Finistère, France). Cybergeo : European Journal of structurelle dans un système européen? Rapport de la 26ème session « Sécurité Geography [En ligne], Espace, Société, Territoire, document 586, mis en ligne et Justice » 2014–2015, Groupe de Sécurité stratégique, Paris, 147 p. le 02 février. https://doi.org/10.4000/cybergeo.25077. Krien N., 2014, Place des risques côtiers dans la représentation du cadre de vie Chauveau, E., C. Chadenas, B. Comentale, P. Pottier, A. Blanlœil, T. Feuillet, D. d’individus possédant des enjeux sur des communes « à risque », thèse de Mercier, L. Pourinet, N. Rollo, I. Tillier, and B. Trouillet. 2011. Xynthia : leçons l’Université de Bretagne occidentale, 237p. d’une catastrophe. Cybergeo : European Journal of Geography [En ligne], Lagadec P., 1993, Apprendre à gérer les crises - société vulnérable, acteurs Environnement, Nature, Paysage, document 538, mis en ligne le 09 juin responsables, Editions d’Organisation, 120p. 2011. URL : https://journals.openedition.org/cybergeo/23763. https://doi.org/ Lagadec P., 2001, Les exercices de crise : pour des ruptures créatrices, La lettre 10.4000/cybergeo.23763. des cindyniques, n°34. Church, J.A., P.U. Clark, A. Cazenave, J.M. Gregory, S. Jevrejeva, A. Levermann, M.A. Lagadec, P. 2004. Understanding the French 2003 heat wave experience: Beyond Merrifield, G.A. Milne, R.S. Nerem, P.D. Nunn, A.J. Payne, W.T. Pfeffer, D. the heat, a multi-layered challenge. Journal of Contengencies and Crisis Stammer, and A.S. Unnikrishnan. 2013. Sea level change. In Climate change Management 12 (4): 160–169. 2013: The physical science basis. Contribution of Working Group I to the Fifth Lagadec, P. 2012. Du risque majeur aux mégachocs. Dossier spécial : la gestion de Assessment Report of the Intergovernmental Panel on Climate Change, ed. T. crise, méthodologies et retours d’expérience, Sécurité et stratégie 10: 50–52. F. Stocker, D. Qin, G.-K. Plattner, M. Tignor, S.K. Allen, J. Boschung, A. Nauels, Lagahé, E., and F. Vinet. 2014. Evaluation de la vulnérabilité des logements face à la Y. Xia, V. Bex, and P.M. Midgley, 1137–1216. Cambridge: Cambridge submersion sur l’île d’Oléron, Rapport de projet Risks, 103. Île d’Oléron: UMR University Press. https://doi.org/10.1017/ CBO9781107415324.026. Liens Université de la Rochelle, PAPI. Creton-Cazenave, L., and V. November. 2017. La gestion de crise à l’épreuve de Lazure, P., and F. Dumas. 2008. An external–internal mode coupling for a 3D l’exercice. EU SEQUANA, La documentation française, 240:237 p. hydrodynamical model for applications at regional scale (MARS). Advances in D’Ercole, R., and P. Metzger. 2009. La vulnérabilité territoriale : une nouvelle Water Resources 31: 233–250. approche des risques en milieu urbain, Cybergeo. European Journal of Le Roy, S., R. Pedreros, C. André, F. Paris, S. Lecacheux, F. Marche, and C. Geography [En ligne], Dossiers, document 447, mis en ligne le 31 mars 2009, Vinchon. 2015. Coastal flooding of urban areas by overtopping: Dynamic consulté le 14 octobre 2016. URL : https://journals.openedition.org/cybergeo/ modelling application to the Johanna storm (2008) in Gâvres (France). 22022. https://doi.org/10.4000/cybergeo.22022. Natural Hazards and Earth System Sciences 15: 2497–2510. https://doi.org/ D’Ercole, R., J.-C. Thouret, O. Dolfus, and J.-P. Asté. 1994. Les vulnérabilités des 10.5194/nhess-15-2497-2015. sociétés et des espaces urbanisés : concepts, typologie, modes d’analyse. Leone F., 2007, Caractérisation des vulnérabilités aux catastrophes « naturelles » : Revue de géographie alpine, tome 82, n°4: 87–96. contribution à une évaluation géographique multirisque (mouvements de Daniels, R.J., D.F. Kettl, and H. Kunreuther. 2011. On risk an disaster: Lessons from terrain, séismes, tsunamis, éruptions volcaniques, cyclones) – HDR de hurricane Katrina, Philadelphia: University of Pensylvania Press, 304 p. l’Université Paul Valéry Montpellier 3, 244 p. + annexes. DREAL de bassin Rhône-Méditerranée, 2016, Plan de gestion des risques Leone F., Asté J-P., Leroi E., 1996, L’évaluation de la vulnérabilité aux inondations, bassin Rhône-Méditerranée, 2016–2021, vol 2, 295 p. mouvements de terrains : pour une meilleure quantification du risque, Revue Duan, W., B. He, K. Takara, P. Luo, M. Hu, N.E. Alias, and D. Nover. 2015. Changes de géographie alpine, tome 84, n°1, pp. 35–46. https://doi.org/10.3406/rga. of precipitation amounts and extremes over Japan between 1901 and 2012 1996.3846. http://www.ano-omiv.cnrs.fr/images/Publications/PDFs/ and their connection to climate indices. Clim Dynam 45: 2273–2292. MasAvignonet/JournalManuscripts/1996-Leone_evaluation.pdf. Duan, W., B. He, K. Takara, P. Luo, M. Hu, N.E. Alias, et al. 2014. Climate change Meur-Ferec C., Flanquart H., Hellequin A.P., Rulleau B., 2011, Risk perception, a key impacts on wave characteristics along the coast of Japan from 1986 to 2012. component of systemic vulnerability of coastal zones to erosion-submersion. Journal of Coastal Research 68: 97–104. Case study on the French Mediterranean coast. Littoral 2010 - Adapting to Egli, D.S. 2013. Beyond the storms. Strengthning preparedness, Response & Global Change at the Coast: Leadership, Innovation and Investment 2011, Resilience in the 21 st century. Journal of Strategic Security 6 (2): 32–45. Royaume-Uni (2010), 8p. Flanquart, H. 2014. Perception versus représentation du risque de submersion et Nicolae, Lerma A., S. Elineau, T. Bulteau, F. Paris, P. Durand, B. Anselme, and R. autres risques : ce que révèle une querelle sémantique, In Connaissance et Pedreros. 2018. High resolution flood modeling coupling overflowing and compréhension des risques côtiers : aléas, enjeux, représentations, gestion, actes de overtopping process, framing the hazard based on historical and statistic colloque, 3-4 juillet 2014, Université de Bretagne occidentale, Brest, 353–364. approach. Natural Hazards and Earth System Sciences 18: 207–229. https://doi. Gallien, T.W., B.F. Sanders, and R.E. Flick. 2014. Urban coastal flood prediction: org/10.5194/nhess-18-2017-2018. Integrating wave overtopping, flood defenses and drainage. Coastal Nicolae, Lerma A., S. Elineau, F. Paris, Y. Balouin, S. Lecacheux, P. Durand, B. Anselme, Engineering 91: 18–28. https://doi.org/10.1016/j.coastaleng.2014.04.007. E. Longepe, S. Defossez, and Gianella L. Goeldner. 2015. Modélisation numérique Genovese, E., S. Hallegatte, and P. Dumas. 2011. Damage assessment from storm de l’aléa de submersion appliquée à l’élaboration des plans d’évacuation et à la surge to coastal cities: Lessons from the Miami area. In Advancing gestion de crise, exemple de la commune de Leucate. Ferrare: Paralia, In: Actes 3rd Geoinformation Science for a Changing World. Lecture Notes in Geoinformation Coastal And Maritime Mediterranean Conference. and Cartography, (eds). Geertman S., Reinhardt W., Toppen F. vol 1. 21–43. Pont C., 2015, Aborder et caractériser le risque de submersion marine sur les Berlin, Heidelberg: Springer. littoraux sableux méditerranéens : l’exemple de la commune de Leucate, Genovese, E., and V. Przyluski. 2013. Storm surge disaster risk management : The mémoire de master, université Paris 1 Panthéon Sorbonne, 150 p. Xynthia case study in France. Journal of Risk Research 16(7):825–841. Reghezza Zitt, M., and S. Rufat. 2015. Résiliences. Sociétés et territoires face à Goeldner-Gianella L., Chionne D., 2014, Des enquêtes sociales sur les savoirs, les l’incertitude, aux risques et aux catastrophes, 226. London: ISTE Editions. pratiques et les comportements face aux risques littoraux, Colloque Ruin I., 2010, Conduite à contre-courant et crues rapides, le conflit du quotidien international Réduire les risques littoraux et s’adapter au changement et de l’exceptionnel, Annales de la Géographie, v4, n°674, p. 419–432. climatique, La Rochelle, 2-4 Avril 2014. https://lienss.univ-larochelle.fr/IMG/ Samuels, P., Huntington S., Allsop W., Harrop W., 2008, Flood risk management : pdf/actes_colloque_la_rochelle_2-4_avril_2014.pdf. Research and Practice. Proceedings of the European conference on flood risk Guimarães, P.V., L. Farina, E. Toldo Jr., G. Diaz-Hernandez, and E. Akhmatskaya. management research into practice (Floodrisk 2008), Oxford, extended 2015. Numerical simulation of extreme wave runup during storm events in abstracts volume, 332 p. Tramandaí Beach, Rio Grande do Sul, Brazil. Coastal Engineering 95: 171–180. Slangen, A.B.A., M. Carson, C.A. Katsman, R.S.W. van de Wal, A. Köhl, L.L.A. https://doi.org/10.1016/j.coastaleng.2014.10.008. Vermeersen, and D. Stammer. 2014. Projecting twenty-first century regional Haut Comité Français pour la Défense Civile (HCFDC), 2012, Risques et menaces sea-level changes. Climatic Change 124 (1): 317–332. exceptionnels. Quelle préparation ?, Rapport d’activité 2011, 139p. Sorensen, C., M. Jebens, and T. Piontkowitz. 2017. Danish risk management plans Hellequin, A.-P., C. Meur-Férec, H. Flanquart, and B. Rulleau. 2013. Perceptions du of the EU flood directive. La Houille Blanche, n° 4: 31–39. risque de submersion marine par la population locale du littoral Stern, E. 2014. Designing Crisis Management Training and Exercises for Strategic languedocien : contribution à l’analyse de la vulnérabilité côtière. Natures Leaders. A Swedish and United States Collaborative Project., 112. Stockholm: Sciences Sociétés 21 (4): 385–399. Swedish National Defense College, Elanders Sverige AB. Durand et al. Geoenvironmental Disasters (2018) 5:19 Page 17 of 17 Thouret, J.-C., and R. D’Ercole. 1996. Vulnérabilité aux risques naturels en milieu urbain : effets, facteurs et réponses sociales. Cahiers des sciences humaines, ORSTOM 32 (2): 407–422. Ullmann, A. 2009. Changement climatique et évolution des tempêtes dans le Golfe du Lion : approche par intégration d’échelles spatio-temporelles. Cybergeo : European Journal of Geography [En ligne], Environnement, Nature, Paysage, document 441, mis en ligne le 18 mars 2009, consulté le 25 janvier 2017. URL : https://journals.openedition.org/cybergeo/22013. https://doi.org/10.4000/ cybergeo.22013. Vinchon C., Baron-Yelles N, Berthelier E., Hérivaux C., Lecacheux S., Meur-Ferec C., Pedreros R., Rey-Valette H., Rulleau B., 2011, MISEEVA: Set up of a transdisciplinary approach to assess vulnerability of the coastal zone to marine inundation at regional and local scale, within a global change context, Littoral 2010 – Adapting to Global Change at the Coast: Leadership, Innovation, and Investment, London 21-23 September 2010, 10p. Wassner T.S., Sauvagnargues, P.A. Ayral, F. Tena-Chollet, N. Frealle, 2016. Retour sur l’exercice de simulation de crise au sein du simulateur de l’école des mines d’Alès, rapport WP4 - D4 - 3a, projet ANR SPICy "Système de prévision des inondations côtières et fluviales en contexte cyclonique", 142 p. Yates-Michelin M., Le Cozannet G., Balouin Y., 2011, Etat des connaissances sur les effets potentiels du changement climatique sur les aléas côtiers en région Languedoc-Roussillon, rapport final BRGM/RP 58 872- FR, 83p., 26 III., 2 annexes. Zijlema M., Stelling G., Smit P., 2011, “SWASH: An operational public domain code for simulating wave fields and rapidly varied flows in coastal waters”.

Journal

Geoenvironmental Disasters – Springer Journals

Published: Oct 29, 2018

There are no references for this article.