Search

International audience; Large wildfires across parts of France can cause devastating damage which puts lives, infrastructure, and the natural ecosystem at risk. In the climate change context, it is essential to better understand how these large wildfires relate to weather and climate and how they might change in a warmer world. Such projections rely on the development of a robust modelling framework linking large wildfires to present-day atmospheric variability. Drawing from a MODIS product and a gridded meteorological dataset, we derived a suite of biophysical and fire danger indices and developed generalized linear models simulating the probability of large wildfires ( > 100 ha) at 8 km spatial and daily temporal resolutions across the entire country over the last two decades. The models were able to reproduce large-wildfire activity across a range of spatial and temporal scales. Different sensitivities to weather and climate were detected across different environmental regions. Long-term drought was found to be a significant predictor of large wildfires in flammability-limited systems such as the Alpine and south-western regions. In the Mediterranean, large wildfires were found to be associated with both short-term fire weather conditions and longer-term soil moisture deficits, collectively facilitating the occurrence of large wildfires. Simulated probabilities on days with large wildfires were on average 2-3 times higher than normal with respect to the mean seasonal cycle, highlighting the key role of atmospheric variability in wildfire spread. The model has wide applications, including improving our understanding of the drivers of large wildfires over the historical period and providing a basis on which to estimate future changes to large wildfires from climate scenarios.

[Departement_IRSTEA]Territoires [ADD1_IRSTEA]Adaptation des territoires au changement global; National audience; Ce court article présente quelques applications possibles de résultats issus du Centre d'Expertise Scientifique (CES) incendie du pôle Theia pour les gestionnaires du territoire

In WUI, the ornamental vegetation can be an efficient vector of fire propagation towards the housing. Assessing this vegetation flammability and ranking ornamental species accordingly can be a way to assess the fire risk in these areas. As both live and dead surface fuels are targeted in fire propagation (due to spot fire or radiant heat), the flammability of both types of fuel was assessed for each species, as well as, at larger scale, the fire propagation from the litter to the first branch. The objectives of this work were to show if the ranking of species varied from one type of fuel to the other, but also according to the fuel scale, from the finest (particle=live leaf) to intermediate (litter samples), and to larger scale (fire propagation from litter to branch). The flammability of the 15 main ornamental species used in WUI of SE France was assessed for these different fuel scales using a fire bench to burn the litter samples and to assess the fire propagation from litter to branch (tested on 4 species) while live leaves were burned on an epiradiator. In each case, the different species were ranked from the least flammable to the most flammable according to the flammability variables recorded during the burning experiments. Then, the different rankings were compared with one another. The comparison of rankings of species for live leaf and litter showed that the ranking did not vary for only two species (regardless of fuel type, Pittosporum tobira: least flammable species and Cotoneaster franchetti: most flammable species). However, for most species, the ranking varied; some species having litters more flammable than live leaves and some presented the opposite pattern. The fire propagation from the litter to the branch significantly varied according to species (Elaeagnus being the most flammable species and Pyracantha the least flammable species), even if, on the whole, fire propagated very well vertically and horizontally, consuming the branch almost completely. In this case, the ranking of species was the same as that of live leaves for Prunus (intermediate flammability), as that of litters for Pyracantha (low flammability) and Elaeagnus (high flammability) but was in between these rankings for Ligustrum (intermediate flammability). Ranking the ornamental species according to their flammability was not that simple especially because some species did not present the same flammability at different fuel scales. For each species, the flammability assessment should take into account the most severe ranking, along with the combustibility of the whole plant, in order to avoid an underestimation of fire risk.

In WUI, the ornamental vegetation is an efficient vector of fire propagation towards the housings. One way to assess the fire risk around housing is the ranking of the flammability of the ornamental species. The objective of this work was to show if the ranking of the flammability of the species varied between live and dead fuels of the main ornamental species in SE France. Flammability variables were recorded during burning experiments using a fire bench for litters and an epiradiator for live leaves. For both types of fuel, the different species were ranked from the least flammable to the most flammable according to these variables. The rankings were compared together as well as to the “total” ranking obtained using the variables of live and dead fuels. The main results showed that a few species presented the same ranking regardless of the type of fuel. However, for most species, the ranking of litter flammability differed from that of live leaf flammability; the litter of some species being ranked more flammable than their live leaves or the opposite. For the “total” ranking of these species, the most flammable species corresponded to the species presenting the most flammable litters and the moderately flammable ones to those of moderately flammable live leaves. However, the poorly flammable species corresponded to species either with highly flammable litters or moderately flammable live leaves. Among the least flammable species, Ligustrum japonicum presented litters ranked as highly flammable. The underestimation of the flammability of some species in the “total” ranking can be an issue in the fire risk assessment.

National audience; Nous présentons rapidement les principaux résultats d'un article en cours de rédaction sur le sujet.

Dans un contexte de pression urbaine et de déprise agricole, les interfaces habitat-forêt se multiplient en France méditerranéenne et deviennent une véritable préoccupation pour la gestion des territoires et du risque d'incendie de forêt. Dans le département des Bouches-du-Rhône, les chercheurs d'Irstea ont utilisé la méthodologie de cartographie diachronique pour y mesurer le développement spatio-temporel de ces espaces afin d'aider les décideurs à cibler les priorités d'actions de prévention.

WUI are key locations in terms of fire risk management, as well for the fire occurrence (hazard) as for the fire effects and damage (vulnerability). Their characterization and mapping are taken into account in the fire risk assessment through the mapping of houses which are main stakes, thus helping the fire fighting through knowledge of the spatial organization of these areas and improving the fire prevention. Managers and end-users look forward the improvement of the methodological guide of the characterization and mapping of WUI, performed by Irstea in 2010 for MEDDTL. Improvements developed in the current report take into account the vegetation on the one hand and are related to the housing characterization on the other hand. The vegetation has been taken into account at two scales: (i) at the global scale, the horizontal structure of the wildland vegetation is assessed by the cover rate that can be calculated on large territories and separated from the agricultural areas; (ii) at the local scale, fuel classes are spatially identified from current fuel typologies. The improvement of the housing characterization at the global scale allows us: (i) to take into account, beyond the houses, all the dwellings submitted to the clear-cutting regulation (agricultural, industrial, trade dwellings, public dwellings, campgrounds, etc.); (ii) to identify relevant classes of dwelling density individualizing dwellings at the vicinity of the WUI from the dwellings at the centre of the WUI., Les interfaces habitat-forêt constituent des espaces clé en matière de gestion du risque d'incendie de forêt que ce soit en terme d'occurrence des feux (aléa) mais aussi d'effets et de dommages (vulnérabilité). La caractérisation et la cartographie des interfaces habitat-forêt participe à l'évaluation du risque d'incendie de forêt en inventoriant et localisant les enjeux primordiaux que sont les habitations, en aidant la lutte à travers la connaissance de l'organisation spatiale de ces territoires, en améliorant la prévention. Le Guide Méthodologique de caractérisation et cartographie des interfaces habitat-forêt1, réalisé par Irstea en 2010 pour le MEDDTL, suscite beaucoup d'attente de la part des gestionnaires et utilisateurs qui souhaitent que certains points méthodologiques soient améliorés. Les améliorations développées dans le présent rapport concernent la prise en compte de la végétation d'une part, la caractérisation de l'habitat d'autre part. L'amélioration de la prise en compte de la végétation se situe à deux échelles: (i) à l'échelle globale, la structure horizontale de la végétation naturelle est appréhendée à partir du calcul d'un taux de recouvrement pouvant être calculé sur de vastes territoires et les surfaces agricoles sont différenciées des surfaces correspondant à la végétation naturelle; (ii) à l'échelle locale, des classes de combustibles sont identifiées spatialement à partir de typologies de combustibles existantes. L'amélioration de la caractérisation de l'habitat permet à l'échelle globale: (i) de considérer au-delà des bâtis de type résidentiel l'ensemble des constructions soumises au débroussaillement (constructions agricoles, industrielles, commerciales et édifices publics, campings, etc.); (ii) d'identifier des classes de densités de bâtis pertinentes discriminant les bâtis situés en périphérie d'une interface des bâtis situés au centre de celle-ci.

Since the 2000s, assessing the impact of forest fires on both natural and human environments using the spatial analysis of fire ignitions and burned areas has often been related to the improvement of the knowledge of fire causes and to the development of various risk indicators. The spatial distribution of forest fires is studied through two classes of variables: the environmental variables and the socio-economic variables. The novelty of the present work is the study of the fire impact, e.g. burned area and fire frequency, according to different types of territory differing by their climate, land cover and population density. The idea was to identify the most impacted territories, especially those located in the wildland-urban interfaces (WUI). The first part of this work was carried out in the département Bouches-du-Rhône and included: (i) the building of an accurate and complete spatial database of the different environmental and socio-economic factors available at the level of the département, (ii) the study of the spatial relationships between the fire ignitions or the burned areas and each of the factors identified, (iii) the identification of territories that present the same characteristics (climate, land cover, population density) in the study area. The study of the spatial relationships between fire ignitions/burned areas and each factor showed that (i) the human activities played an important part in the location of fire ignitions (WUI, distance to communication networks), (ii) the factors related to relief especially, as well as the factors related to vegetation (vegetation continuity, vegetation types) seemed to be the most significant for the distribution of burned areas. The study of territories presenting the same characteristics was carried out by combining descriptive variables determined for each territorial entity (e.g. municipality). This combination of descriptive variables allowed the identification of 12 types of territory. From these types, the spatial relationships between forest fires and environmental and socio-economic variables will be compared in order to identify the most impacted types of territory, including in the WUI areas., L’impact des incendies de forêt sur le milieu naturel comme sur le milieu anthropique, appréhendé à partir de l'analyse spatiale de l'environnement des départs de feux ainsi que des surfaces incendiées depuis le début des années 2000, a souvent été lié à l’amélioration de la connaissance des causes de départs de feux ainsi qu'à l’élaboration de nombreux indicateurs de risque. La répartition spatiale des incendies de forêt est étudiée au travers de deux catégories de variables: les variables biophysiques, et les variables socio-économiques. L'originalité de la présente étude est d'étudier l’impact des feux de forêt en fonction de différents types de territoires contrastés – en matière de climat, d'occupation du sol, de densité de population – au travers de différents paramètres (surfaces brûlées, fréquences de feux) mis en relation avec les données environnementales / biophysiques et socio-économiques. Il s’agirait alors d’identifier des types de territoires, en particulier situés dans les interfaces habitat-forêt, les plus impactés par les incendies de forêt. La première tranche de cette étude est réalisée sur le département des Bouches-du-Rhône et s'articule autour de plusieurs points: (i) la constitution d'une base de données spatiales précises et exhaustives des différents facteurs biophysiques et socio-économiques disponibles sur le département; (ii) l'étude des relations spatiales entre les départs de feux de forêts ou les surfaces brûlées et chacun des facteurs biophysiques ou socio-économiques identifiés; (iii) l'identification sur le département des territoires présentant les mêmes caractéristiques biogéographiques (climat, occupation du sol, densité de population). L'étude des relations spatiales entre les départs de feux de forêt ou les surfaces incendiées et chacun des facteurs biophysiques ou socio-économiques montrent que: (i) les activités humaines jouent un rôle primordial dans la localisation des départs de feux (interfaces habitat-foret, distance aux réseaux); (ii) les facteurs liés au relief notamment ainsi que le facteur végétation (continuité de la végétation, type de végétation) semblent prépondérant pour la répartition des surfaces brûlées. Le regroupement des territoires présentant les mêmes caractéristiques biogéographiques a été réalisé à partir de la combinaison de variables descriptives déterminées pour chaque unité territoriale définie à partir des communes. Cette combinaison de variables descriptives a permis d'identifier 12 types de territoire. À partir de ces 12 types de territoires, les relations spatiales entre les incendies de forêt et les facteurs biophysiques et socio-économiques pourront être comparées afin d'identifier les types de territoires les plus impactés par les incendies y compris dans les zones d'interface habitat-forêt.

International audience; Fire risk is high in Wildland-Urban Interfaces (WUI) and WUI fires, burning the surrounding ornamental vegetation, are often transmitted to structures. Thus, to be able to use fire-wise species in such locations, the assessment and ranking of the flammability of ornamental plant species are required. The flammability of seven species, among those most frequently planted in hedges in SE France, was studied, under laboratory conditions, at two levels: live leaf and litter. Leaf characteristics (FMC, mass, surface, volume, S/V ratio) were measured and the flammability parameters (ignition frequency, time-to-ignition, flaming duration) were recorded using an epiradiator as burning device. The Gross Heat of Combustion (GHC) was measured by calorimetry. The flammability parameters (ignition frequency, time-to-ignition, flaming duration and initial flame propagation) as well as the bulk density of the undisturbed litter samples were recorded during burning experiments performed on a fire bench. Regarding live leaves, burning experiments showed that Phyllostachys sp. and Photinia fraseri had higher ignitability (along with Pyracantha coccinea) and sustainability and were characterized by low FMC whereas Pittosporum tobira and Nerium oleander had lower ignitability and were characterized by high GHC. Prunus laurocerasus and Cupressus sempervirens had lower sustainability and were characterized by low S/V ratio. Regarding litters, Cupressus sempervirens and Pyracantha coccinea (high proportions of scale-leaves and/or fine particles and high bulk density) presented higher sustainability, ignitability (high ignition frequency), along with Photinia fraseri, and lowest combustibility (low flame propagation) along with Pittosporum tobira. The latter, characterized by high proportions of debris, coarse particles and evergreen leaves, was the least ignitable (high time-to-ignition) along with litters of Pyracantha coccinea and Nerium oleander, and had lower sustainability along with those of Phyllostachys sp. and Nerium. Litters of Prunus laurocerasus, Photinia fraseri and Phyllostachys had better combustibility (high proportions of debris) and the latter was also quick to ignite. Hierarchical cluster analysis performed on the flammability parameters recorded during both types of experiments ranked the seven species in four distinct clusters from the most flammable (Phyllostachys sp. and Photinia fraseri) to the least flammable (Pittosporum tobira and Nerium oleander); the other species displaying two groups of intermediate flammabilities (Prunus laurocerasus-Pyracantha coccinea and Cupressus sempervirens ). Further flammability experiments on the whole plant are needed and the highly flammable species should not be used in hedges planted in WUIs in SE France.

L'analyse diachronique des interfaces habitat-forêt est réalisée sur le département des bouches-du-rhône à deux dates: 1999 et 2009. Les interfaces habitat-forêt – définies par l'environnement de 100 mètres autour des bâtis résidentiels situés à moins de 200 mètres de forêts, garrigues ou maquis – ont été cartographiées aux deux dates selon la méthode développée par Irstea dans le guide méthodologique intitulé «Caractérisation et cartographie des interfaces habitat-forêt»1. Cette méthode est fondée sur la combinaison de critères spatiaux caractérisant l'organisation de l'habitat résidentiel d'une part, la structure horizontale de la végétation d'autre part. L'outil WUImap©, développé par Irstea, aide à cartographier ces interfaces. L'objectif est de mettre en évidence les évolutions spatiales des interfaces habitat-forêt entre 1999 et 2009, soit une période de 10 ans, sur l'ensemble du département des Bouches-du-Rhône mais également pour chacun des 23 massifs du département de manière à quantifier les dynamiques de ces territoires. L'étude de la dynamique des interfaces habitat-forêt réalisée sur le département des Bouches-du-Rhône permet de mettre en évidence les territoires pour lesquels les changements sont les plus importants en termes : (i) de dynamique de la végétation (évolutions de la structure horizontale de la végétation), (ii) de dynamique du bâti et de la configuration de l'habitat résidentiel, (iii) de dynamique des interfaces habitat-forêt et des types d'interface habitat-forêt, (iv) de dynamique des territoires soumis à l'obligation légale de débroussaillement. Les analyses réalisées à l'échelle du département des Bouches-du-Rhône et à l'échelle des 23 massifs forestiers du département permettent d'identifier et de caractériser géographiquement les changements.

In laboratory conditions, reconstructed or intact litter samples can be used to assess the flammability of dead surface fuels but the sampling method could affect litter flammability results. To assess this effect, samples of litters were collected in south-eastern France, according to two different methods previously used in other studies, one keeping intact the structure of litter layers (intact litter) and the other using mainly the surface litter layer to reconstruct the litter sample (reconstructed litter). The sampling occurred in representative ecosystems of the study area: pure Pinus halepensis stands, mixed pine-oak stands, pure cork oak stands and in two types of shrubland (maquis and garrigue). Flammability experiments were carried out on fire bench, using a glowing firebrand as ignition source and a 10 kmh-1 wind. The main variables recorded were: ignition frequency, time-to-ignition, flaming duration, flame rate of spread, flame propagation, mean flame temperature, maximum flame height and rate of consumption. The comparison of flammability results showed that the sampling method had a significant effect on litter bulk-density, rate of spread and rate of consumption, intact litter being more flammable than reconstructed litter. The type of vegetation had a significant effect on litter depth, ignitability and combustibility (except on flame temperature and rate of spread) and the litter composition could explain in part this fire behaviour. The effect of the reconstruction of litters on flammability parameters and its magnitude differed according to vegetation types. The intact structure of the litter appeared to be an important driver of its flammability, especially for combustibility and consumability. The assessment of these parameters will differ when using reconstructed litter samples compared to intact litter samples. Future research on litter flammability should take into account the bias due to the litter sampling method when the litter is reconstructed.

SE France is the area the most affected by wildfires, so for a better fire prevention, it is necessary to understand wildfires variation in space and time and to improve the knowledge of their causes. Using the forest fires database Prométhée which gathers all the forest fires of SE France since 1973, the number of fires, their size and causes were investigated during two time periods (1973-1990 and 1991-2010), in three geographical areas of the study area. Two fire sizes were taken into account as the large fires (≥100 ha) represented 1% of the total number of fires but burned 72% of the total area between 1973 and 2010. There was a significant decrease in occurrence and burned area from the first period to the second, except for region 1 in which the number of fires increased. The three regions differed significantly in number of fires and burned areas, region 1 being the most affected. The knowledge of fire causes varied spatially; the proportion of unknown causes being the highest in region 1, and it increased between both periods. Fire causes varied spatially; the main causes were arson in region 1 and negligence during professional activities (especially farm works) in regions 2 and 3. There was also a temporal variation, following the LCLU changes, but also between the seasons. On the whole, the proportions of causes due to arson and to negligence during leisure activities increased between both periods, especially in summer, while that of causes due to professional activities, more frequent from fall to spring, decreased regardless of the region. Large fires were mainly due to arson, except in region 3 where negligence during farm works was the main cause, whereas smaller fires were mainly due to negligence during farm works especially. Before 1991, the main cause differed between regions, ranging from negligence during professional activities (region 3) to arson (region2) and accidental fires in rubbish dump (region 1) but since then, arson is the most frequent cause regardless of the region.

International audience; Fire is one of the environmental risks that is expected to increase in connection to climate change. Fire will be more common in regions where it is today less thought of. EU-FireSmart is a project whose objectives are to identify obstacles that hinder the effectiveness of forest fire preventive measures and to derive recommendations to integrate prevention practices into standard forest management plans. The project tackles both the European and local level of addressing forest fire prevention, where the local level has been covered through the implementation of test areas in France, Italy, Spain and Portugal. Documents containing information about different methods and practices of forest fire prevention in Europe have been gathered in a database which now contains more than 1,400 entries. The material available in this database has been analysed according to the strengths, weaknesses, opportunities and threats of different land management practices. This analysis has been performed considering different subjects. The analysis is then used to derive practical recommendations on how to turn current negative fire prevention factors into viable and proactive factors able to strengthen prevention methods. Agroforestry has been found to be a strong preventive method and the strengths and weaknesses of the method as well as the practical recommendations for increasing the preventive values of agroforestry are presented.

The submission presents a methodology to obtain a spatial total fire risk map in wildland-urban interfaces using a Geographic Information System (GIS) and spatial analysis. The study area is located in the county of Provence, South-eastern of France. The following fire risk variables were generated for this area: housing configuration, nature and structure of vegetation, topography, land cover data, etc. Combined, these variables indicate the intermediate fire risk indicators, which represent the density of fire ignition, the density of wildfire and the burned surface ratio. The coefficients for generating these intermediate indicators as well as the final index were obtained from PLS regressions, where density of fire ignition, density of wildfire and burned surface ratio data were considered as the dependent variables. A total fire risk index was also built from these three indicators bearer of information of hazard and/or vulnerability. This index was mapped and results are promising in terms of fire risk assessment. Finally spatial implications of the results, and their potential use for prevention planning appeared interesting in terms of fire management

In SE France, where most wildfires occur, the fire size has never exceeded 6744 ha whereas mega-fires have already burned huge areas in other Mediterranean countries such as Spain or Portugal. It was interesting to find out what main factors drove the ignition of the largest fires having occurred in this region. The study was carried out on the forest fires database Prométhée, recording all the forest fires in the 15 département of SE France since 1973. However, the records before 1997 being not reliable, only the period 1997-2010 was investigated. Only 1% of the fires (N=260) recorded during this period were larger than 100ha whereas 78% of the fires were smaller than 1ha. However these large fires accounted for 71% of the burned area and 66% of these fires occurred during the summer (July-August). The number of large fires and the burned area per year and per département were calculated and the proportions of fires causes were determined. In each département, the impact of different explanatory variables, such the population density, the proportion of land covers, the number of dry months and the dry severity index (de Matonne index, IM), on the number of large fires and on the burned area was investigated using a multivariate analysis (Co-inertia analysis). Results showed that high values in shrubland and pasture covers, population density and number of dry months had a positive impact on the number of large fires whereas high values in forest cover, spring and fall-winter IM had a negative impact on this parameter. High values in natural vegetation cover, especially the shrubland cover, fall-winter IM and number of dry months had a positive impact on the burned area whereas high values in summer IM, farmland and pasture covers and population density had a negative impact on this parameter. The départements the most affected by such fires are those situated at the eastern part of the region on the outskirts of the Mediterranean Sea and, in each one of them, the main fire cause is arson.

International audience; Fire incidence upon southern European forests is expected to worsen as a result of climate change; furthermore, fire prevention methods, priorities, practices and legal frames vary considerably across Europe. Hence the need to reinforce the corner stone of fire fighting, i.e.: fire prevention. However, the baseline knowledge concerning forest fire preventive practices is atomized and not universally accessible to the wide stakeholder community, particularly along the silvicultural chain. This paper presents early results of the EU Support Action EU-FireSmart whose objectives are: (i) to identify the obstacles that hinder the effectiveness of forest fire preventive measures and (ii) to derive recommendations to integrate prevention practices in regular forest management plans. The analysis tackles both the European and the local panorama, by means of test sites in Portugal, Spain, France and Italy. A scientific and technological data base (1400 entries, public by mid 2011) of forest fire prevention methods or practices has been gathered and analysed to ascertain the strengths, weaknesses, opportunities and threats of the said preventive experiences within eight defined topics (agroforestry and grazing as preventive measures, awareness rising, prevention in the wild land urban interface, etc). The analysis of the DB yields practical recommendations to turn current negative factors into pro-active forest fire prevention measures, also backed up by the preliminary conclusions of a questionnaire designed for experienced forest managers and scientists.

The submission presents a methodology to obtain a spatial total fire risk map in wildland-urban interfaces using a Geographic Information System (GIS) and spatial analysis. The study area is located in the county of Provence, South-eastern of France. The following fire risk variables were generated for this area: housing configuration, nature and structure of vegetation, topography, land cover data, etc. Combined, these variables indicate the intermediate fire risk indicators, which represent the density of fire ignition, the density of wildfire and the burned surface ratio. The coefficients for generating these intermediate indicators as well as the final index were obtained from PLS regressions, where density of fire ignition, density of wildfire and burned surface ratio data were considered as the dependent variables. A total fire risk index was also built from these three indicators bearer of information of hazard and/or vulnerability. This index was mapped and results are promising in terms of fire risk assessment. Finally spatial implications of the results, and their potential use for prevention planning appeared interesting in terms of fire management.

[Departement_IRSTEA]Territoires [TR1_IRSTEA]SEDYVIN; En région méditerranéenne, les forêts sont régulièrement la proie des flammes, avec des conséquences écologiques, économiques et sociales souvent désastreuses. Les zones où l'habitat est en contact étroit avec la forêt représentent plus particulièrement un danger et deviennent une véritable préoccupation pour les gestionnaires, qui ont besoin d'outils de localisation pour agir et limiter les risques d'incendies dans ces territoires. Cet article présente une méthode innovante d'évaluation du risque fondée sur l'observation et la description ainsi que sur l'analyse spatiale et statistique de ces zones à risque.

De toute la France métropolitaine, la Région PACA est celle qui est le plus soumise aux risques naturels. La prise en compte des risques naturels est devenue incontournable et passe par une amélioration constante de l'état des connaissances sur les phénomènes naturels et la nécessaire diffusion de ces connaissances auprès des décideurs mais aussi d'un plus large public. Ce document a pour objectif de porter à la connaissance d'un large public les actions menées dans la région au travers des résultats de projets et d'études scientifiques dans le domaine des risques naturels, réalisés notamment dans le cadre du Contrat de Projets État-Région. Le chapitre Feux de Forêts comprend les fiches suivantes : - phénomène feux de forêts - plan départemental de protection des forêts contre l'incendie PDPFCI 2006-2012 - interface habitat-forêts - guide technique : causes de départs de feux - grands feux de forêts en région PACA

Cette présentation est une description et caractérisation des interfaces habitat-forêt.

Plusieurs fiches sur la thématique forêt. Ouvrage réalisé dans le cadre de l'année internationale de la forêt.

Dans ce livre, le Cemagref, institut de recherche en sciences et technologies pour l'environnement, fait le point de ses travaux scientifiques menés sur la forêt. Pourquoi une recherche sur la forêt ? D'abord parce qu'elle occupe plus d'un quart de la surface de notre territoire métropolitain. Ensuite, parce qu'elle fournit tout un ensemble de services variés et, pour certains, essentiels à notre société : la production de bois, la protection contre les aléas naturels, la chasse et la promenade en sont de très bonnes illustrations. Enfin, parce qu'elle constitue un milieu naturel riche en biodiversité et occupe une place notable dans les débats sur les enjeux environnementaux. Forts de ce constat, les chercheurs et experts du Cemagref apportent leur contribution en abordant des problématiques majeures pour la gestion durable des forêts : état et dynamique de la forêt, de sa biodiversité, vulnérabilité des forêts aux changements globaux et stratégies d'adaptation, gestion multifonctionnelle des forêts, rôle de protection contre les aléas naturels, place de la forêt dans les territoires et enjeux sociaux. L'interaction entre forêt et gestion se situe ainsi au c½ur de nos travaux.

A method to characterize and to map wildland-urban interfaces (WUI) is proposed in the context of wildfire risk. The WUI mapping on the territory allows new spatial configurations: inside WUI and outside WUI. Establishing relationships between WUI distribution and forest fire history (fire ignitions and burned areas) types of territory appear with high levels of risk: high fire ignition density values and high wildfire density and high burned area ratio. A model of a total fire risk index has also been developed., Une méthode de cartographie des interfaces habitat-forêt est développée dans le contexte du risque d'incendie de forêt. La cartographie des interfaces habitat-forêt sur le territoire qui en résulte permet alors une nouvelle compartimentation du territoire : types de territoire interfacés et types situés en dehors des interfaces. En mettant en relation la distribution spatiale de ces types de territoire avec l'historique des incendies (départs de feu et surfaces brûlées), certains types d'espaces révèlent de haut niveau de risque d'incendie avec une forte densité de départs de feu, d'incendie et de taux de surfaces brûlées. Une modélisation du risque d'incendie est proposée de façon globale.

International audience; La méthode cube est appliquée dans différents pays pour décrire le combustible à l'échelle des individus, dans le cadre du programme Fire Paradox. Elle est décrite dans cette communication, réalisée lors du séminaire de restitution des résultats du programme au Maroc.

Assessing the risk of forest fires in wildland urban interfaces (WUIs) is crucial for wildfire prevention and land management. With the goal of developing an efficient management of fire risk in wildland urban interfaces in European Mediterranean countries, this chapter recommends methods and advices for the identification, characterization and mapping of WUIs as well as for the assessment of the fire hazard, the vulnerability and the damage potential of these areas. These tools are the result of scientific researches and fruit of past experiences analysis. More over this chapter insists on the interest of providing wildland urban interface maps in the purpose to know the real extent of theses areas and to manage their development. It briefly presents the steps of a method to identify, to characterize and to map WUI in European Mediterranean countries combining relevant criteria which are connected to the spatial organization of inhabitant dwellings and the structure of the vegetation. A WUI' typology is established. The method can be easily applied by land agencies or managers easily and is suitable both in large areas or landscape level (small scales) and in local conditions (large scale). Because of their high vulnerability, ignition probability and combustibility, it is important and efficient to focus risk assessment in the WUIs. The chapter brings a method to assess and to map fire hazard levels and vulnerability levels according to WUIs and their environment. Introducing the risk of fire and particularly the vulnerability of the territory with such maps is a way to make the inhabitants becoming aware of fire risk in WUI. This will globally decrease the risk of fire either by reducing fire propagation with biomass removal and or by reducing fire ignition probability together with less carelessness. Accomplishing this goal is strictly related to the designation of suitable prevention messages and preventive actions which can be different according to WUI types.

Lien entre structure des peuplements (hauteurs, recouvrement, ...) et comportement du feu simulé avec firetec (vitesse de propagation).

This poster presents a method for wildland-urban interface mapping described in the product P5.2.8, applied to large areas, at a large scale. This method can be adapted easily to specific situations observed in different European countries. The method is based on a combination of four housing configuration types and three types of horizontal structure of vegetation. Twelve wildland-urban interface types have been produced. It is particularly well adapted to the French environment, but the method has also been well adapted (and applied) to Spain and Greece. In tandem with a software tool WUImap it was developed and presented in guide-form to map WUIs automatically. Some examples of use of such WUI maps have been presented in the product.

Fuel mapping is a key activity for forest fire risk management. It is based on remote sensing images processing methods. These methods are versatile and validated in some particular contexts. They are usually implemented in one specific software environment. We propose a distributed solution for sharing and integration of image processing methods developed in their own computer environment and validated in some particular contexts, using specific data, to respond to specific needs. Its architecture includes a knowledge database of methods and resources, and an expert system for methods selection in relation to the user needs specification. Selected methods can then be organised into demarches. An executive engine is designed to execute the different methods of the demarche in their respective computer environment, through mediating wrappers. A research prototype called Fuel Mapping Methods Integration Platform (FMMIP) was developed., La cartographie du combustible est une activité clé pour la gestion du risque d'incendie de forêt. Elle se base sur des méthodes de traitement d'images télé-acquises. Ces méthodes sont variées, validées dans leur contexte de mise en ½uvre, et généralement implémentées dans des environnements logiciels spécifiques. Nous proposons une solution d'intégration distribuée permettant le partage et la réutilisation de méthodes de traitement d'images pour la cartographie du combustible, développées dans des environnements informatiques en utilisant des données hétérogènes afin de répondre à des besoins spécifiques dans des contextes différents. Son architecture s'articule autour d'une base de connaissance et d'un système expert permettant d'évaluer la capacité de chacune des méthodes à répondre aux besoins spécifiés par l'utilisateur. Les méthodes sélectionnées sont alors organisées dans des démarches de traitements d'images, exécutables. Un moteur d'exécution permet l'exécution séquentielle de chacune des méthodes, dans leur environnement informatique respectif, au travers d'adaptateurs logiciels de médiation. Un prototype de recherche, appelé "Fuel Mapping Methods Integration Platform" (FMMIP) a été développé.

International audience; The description and modelling of the fuels provide useful information to be used in pre-fire planning and therefore to improve forest fire management plans. But fuels are complex and variable according to ecological and bio geographical factors. They are composed of vegetal material enabling fire ignition and fire propagation. They have many properties that determine fire behaviour, which make them difficult to characterize. The aims of this paper are (i) to provide a state of the art on the different methods to characterize fuel in the laboratory and in the field, (ii) to explain a method that has been developed in different countries (iii) to give some examples from different countries in the Mediterranean Basin. Different kinds of descriptions provide information on fuel at different scales, from micro to macrostructure. Field sampling and complementary laboratory work are stored in a fuel database and are used as input in the Firetec physical model. This modelling enables the assessment of fire behaviour such as rate of fire spread and fire intensity, according to fuel types. Fuel microstructure is described at the particle level, as for example with bulk density measured according to the particle size. The main parameters for fuel particle characterization have been measured using the same method in different countries: shape, surface area to volume ratio, mass to volume ratio, high calorific value, specific heat, ash content. Different species have been described in different countries. Fuel macrostructure is described at forest stand level or landscape level, according to type, composition, density, cover and height. This paper describes the different steps of a method to describe fuel for fire behaviour assessment and provides some examples coming from Mediterranean Countries.

The paper presents the major outcomes of the produced guide. It particularly describes the method developed for mapping WUIs on large areas and at large scale in the European Mediterranean context. The method used remote sensing and spatial analysis tools. Particularly adapted to the French context, it can be easily adapted according specific contexts observed in different European countries. After specifying accurate definition of wildland-Urban Interface (WUI), a WUI typology was based on the combination of four housing configuration types and three natures of horizontal structure of vegetation determined through objective and quantitative criteria. Housing configuration reveals spatial organization of residential dwellings: isolated, scattered, dense clustered and very dense clustered housing. Horizontal structure of vegetation points out the continuous vegetation, the sparse vegetation or the no existing vegetation in contact with houses. Twelve wildland-urban interface types were produced.In parallel a software tool was developed and briefly presented in the paper in order to map WUIs as automatically as possible. The production of WUI maps presents high interest for quantification of the extension of WUIs on the territory, main characteristics of WUIs, assessment and mapping of fire risk levels in WUIs. The main objective of the guide is to help end-users such as land managers, foresters and fire-fighters to locate WUIs on the ground with a view to develop specific actions for wildfires prevention according to WUI types, to create public awareness programs for inhabitants faced with wildfire risk, and to assess fire fighting difficulties due to urbanization.

National audience

Knowledge about wildland fuel is essential to fire prevention and to both biodiversity conservation and reduction of the negative impacts of fire. Little is known of flammability of dead fuels, especially for non-reconstructed samples that are hypothesized to represent the field conditions.To assess the impact of vegetation types and of fire regime on the flammability of dead fuels, samples of litter were collected undisturbed (i.e. keeping the structure of litter layers) in 30 study sites spread over the limestone Provence. The sampling plan comprised the most representative ecosystems of the study area: pure Pinus halepensis stands, mixed pine-oak stands and shrublands. Three classes of litter depth were studied (low, medium and high) to account the variability existing in the field. Sampling also included the number of fires in each site since 1960 (no fire, 1 or 2 fires and ≥3 fires) and the time interval since the last fire (> 40 years, interval 40-15 years and ≤15 years).Flammability experiments were carried out in laboratory using a glowing firebrand and a 10 km.h-1 wind. The main variables recorded were: ignition frequency, time-to-ignition, flaming duration, flame rate of spread, flame propagation, mean flame temperature, mean flame height and rate of consumption.Results showed that the ignition frequency was higher in mixed stands than in pure pine stands whereas the time-to ignition, flaming duration and rate of consumption were the highest in pure pine stands. The mean flame height and the flame propagation decreased with the increase of the number of fires and the time-to-ignition was the highest when the interval since the last fire was the shortest. The increase of the litter depth entailed an increase of the mean flame temperature, the mean flame height and the flame propagation. The characteristics of litter samples explained in part these results.

Characterizing the flammability of litter fuels is of major importance to assess the wildland fire hazard. We present here a comparison of the flammability of litter fuels for the main ecosystems of the Maures massif (southeastern France) forming a mosaic of oak woodlands (Quercus suber, Q. ilex and Q. pubescens) and shrublands dominated by Erica arborea, Calicotome spinosa and Cistus spp. This mosaic is partly driven by the historic fire regime, and each vegetation type corresponds to a specific range of fire recurrence. We thus tested the hypothesis that, as the fire regime impacts the composition and biomass of live fuels, it may subsequently affect the characteristics and the flammability of dead fuels. We studied undisturbed litter samples (diameter 38 cm) with at least 30 replicates per vegetation type. Each sample was submitted to an experimental burning in laboratory conditions using a glowing firebrand, in order to assess the four main components of flammability (i.e. ignitibility, sustainability, combustibility, and consumability).

La typologie du combustible est un outil utilisé par les gestionnaires des milieux naturels soumis aux incendies dans de nombreux pays et massifs subissant ce risque. Cet outil permet d'intégrer le comportement du feu dans la gestion des milieux et dans la planification, à l'échelle d'un massif forestier. Ces typologies caractérisent la répartition verticale et horizontale de la végétation, ainsi que sa biomasse, éléments déterminants du comportement du feu, en milieu naturel. Le Cemagref d'Aix-en Provence, sur la base de près de 150 descriptions de végétation situées en Basse Provence calcaire, a bâti une typologie des structures de végétation et a pu caractériser le comportement du feu, dans chacun des onze types de combustibles construits.

International audience; Simple risk model based on cellular automat for an application on WUI maps developed in the South of France; Développement d'un modèle de risque fondé sur les automates cellulaires pour une mise en application sur une cartographie des interfaces habitat-forêt dans le sud de la France

The objective of this manual is to improve the knowledge of the fire ignition causes and their accurate localisation. This manual made by Cemagref within the Forest Focus* framework (FF2004-06 convention about the monitoring of the forest continuum relating to the forest fire) was performed with the French partnerships (forest managers, policemen, firefighters) and the Spanish experts (EIMFOR*). It computes different methods used in Spain, Portugal, California and Australia to fit them to the French Mediterranean context. This approach relies on the method of the physical evidences which is very objective and organized in order to leave nothing to chance., Ce guide a pour objectif de participer à l'amélioration de la connaissance des causes de départ de feu ainsi qu'à leur localisation précise. Il a été réalisé par le Cemagref d'Aix-en-Provence dans le cadre de l'étude Forest Focus* (Programme FF2004-06 orienté sur le suivi continu des forêts dans le domaine des incendies de forêt), en collaboration avec les partenaires français (forestiers, gendarmes, pompiers) et espagnols (EIMFOR*) reconnus pour leur compétence dans le domaine. Il synthétise les différentes méthodes utilisées en Espagne, au Portugal mais aussi en Californie et en Australie et propose une adaptation de ces méthodes au contexte français. La démarche préconisée s'appuie sur la méthode des preuves physiques, démarche qui se veut objective et très hiérarchisée, de manière à ne rien laisser au hasard.

Method to characterize and map wildland-urban interfaces (WUI) in the South of France, definition of WUIs, spatial analysis of dwelling structure and vegetation. Typology of WUI combining vegetation structure and dwelling structure. Link with fire risk., Méthode de cartographie des interfaces habitat-forêt développée dans le sud de la France dans un contexte de prévention des risques d'incendie de forêt: définition des interfaces, caractérisation de la structure des habitats et de la végétation. Typologie d'interface et lien avec l'incendie de forêt.

Knowledge about fuel vegetation is a fundamental part of the fire management that is necessary to both conserve biodiversity and reduce the negative impacts of wildland fires. On the framework of the European programme FireParadox, flammability characteristics of dead fuels were described in Southern France on both acidic and calcareous soils according to fire modalities (no fire since 1960, 1 and more fires old and recent), to assess if they were influenced by the past fire regime, influencing in turn the vegetation dynamics. Samples of disturbed litters have been collected in representative ecosystems of southern France : (i) on calcareous soils, in high stands of Pinus halepensis and of mixed oak-pine, in medium-sized stands of P. halepensis and mixed oak-pine with understorey and in garrigue (shrubland); and (ii) on acidic soils, in high Quercus suber stands, high maquis and lower maquis. The low and medium-sized stands come from a post-fire succession, the highest biodiversity, in terms of specific richness occurring between 4 and 11 years after the fire. To assess the flammability of these litters, several series of tests were conducted in laboratory, using a glowing firebrand and a wind speed of 10 km.h-1. The main parameters recorded were time to ignition, flaming duration, flame propagation. For the low (garrigue and low maquis) and the medium-sized (pure pine or mixed pine-oak stands with understorey and higher maquis) stands where biodiversity is maximum, the comparison between litters collected on calcareous soils and on acidic soils shows no significant difference for all the parameters. For the high mature stands, where biodiversity is the lowest, especially in Pine stands, the comparison between the both sets of litters shows that there is a significant difference only for the time to ignition which is much higher on acidic soils (73s) than on calcareous soils (17s). These results suggest that the Quercus suber litter collected on acidic soils is less flammable than the litter collected on calcareous soils.

Knowledge about fuel vegetation is a fundamental part of the fire management that is necessary to both conserve biodiversity and reduce the negative impacts of wildland fires. On the framework of the European programme FireParadox, live and dead fuels were described in Calcareous Provence (Southern France) to assess if their characteristics were influenced by the past fire regime, influencing in turn the vegetation dynamics. The sampling was performed on three types of vegetation representative of the ecosystems of Southern France (Pinus halepensis stands, pine-oak mixed stands and garrigues) according to four fire modalities (no fire since 1960, 1 to 2 fires old and recent, and 3 or more fires). The ecological description took into account as well individualized trees or shrubs as vegetation patches of a same species or vegetation layers including different species of a same height, underlining so that these various structures are also part of the landscape diversity. Thickness and coverage (%) were the variables recorded on the dead fuel while variables such as the height and coverage of the vegetation were recorded on the live fuel. Results show that both fire modalities and vegetation types have a significant influence on the thickness and the coverage of the dead fuel. Concerning the live fuel, the type of vegetation has a significant influence on : (i)height of shrubs, vegetation patches and layer (ii)coverage of the different vegetation height strata (iii)coverage of the trees, shrubs and vegetation patches (iv)tree density per hectare and the fire modalities on : (i)height of trees and vegetation layer (ii)stratum coverage (iii)tree coverage (iv)tree volume per hectare Vegetation types and fire modalities induce variations of these parameters, and so influence the structure of stands. As diverse fire regimes within ecological limits are essential for maintaining biodiversity, this variation of structure may play a part in the diversity at local and landscape scale.

While most wildfire ignition points are aggregated at the vicinity of roads due to human-caused ignitions (Prométhée database), the probability of ignition and initial fire propagation at the interface between road and woodlands have rarely been studied. Vegetation at these interfaces includes dead fuels (litter) and live fuels (grasses and shrubs) which present specific features influencing ignition probability: composition, annual growth cycle, variation of moisture content. Moreover, vegetation is modified by management practices such as mowing and shrub clearing. We present here the results of ca. 900 lab burning experiments of small (18x20 cm) undisturbed vegetation samples combining five dominant vegetation types (graminae, dicots, pine litter, pine litter+ graminae, pine litter+ graminae+ shrubs), two management practices (mowed vs. unmowed vegetation), two modes of ignition (flaming vs. glowing standard wood cube), two wind speeds (3 vs. 10 km/h), and a wide range of fuel moisture content (fresh, air-dried and oven-dried). Results indicate clear differences of flammability variables (ignition probability, time to ignition, propagation) among the five vegetation facies. The combination of ignition mode and wind speed strongly affects the probability of ignition, with contrasted ignition efficiency according to vegetation facies. It also affects the propagation' and sustainability' of fire. Fuel moisture and vegetation types relate well to fire sustainability, propagation and intensity. Multiple and logistic regression were established between ignition probability and the major causal variables of ignition. These experimental data will help modelling fire hazard, especially by implementing a spatially-explicit simulator of fire ignition and initial propagation. The final aim is to help simulating management practices to reduce fire risk at interfaces., La plupart des points d'ignition d'incendies de forêts sont situés à proximité des réseaux routiers, du fait des activités humaines intenses, et pourtant les probabilités d'ignition et de propagation initiale des végétations typiques de ces interfaces (herbacées, litières, végétation débroussaillée...) sont mal connues. Nous présentons ici le résultat d'environ 900 brûlages expérimentaux d'échantillons de végétaiton non perturbés, combinant cinq faciès de végétation, deux types de modes d'ignition, deux vitesses de vent, trois niveaux d'humidité de la végétation et deux modes de gestion de la végétation (débroussaillée / non débroussaillée). Les résultats montrent des différences d'inflammabilité nettes entre les cinq faciès végétatux (probabilité d'ignition, délai et durée d'ignition, propagation). La combinaison du mode d'ignition et de la vitesse du vent modifient fortement ces variables, et l'humidité dans une moindre mesure. Des régressions logistiques et multiples ont permis d'établir des modèles prédictifs de l'ignition. Ces données expérimentales vont permettre de mieux connaître le risque d'incendie, et vont alimenter un simulateur d'ignition et de propagation initiale destiné à tester différents scénarios, et à réduire le risque dans ces interfaces.

Peu d'études ont été conduites sur l'origine des incendies de forêt. Les causes de départs de feu font certes l'objet de suivi mais alimentent de façon plus ou moins exhaustive des bases de données. Celles-ci sont souvent incomplètes et ne permettent de dégager que des tendances ou de cartographier des points d'ignition sur des carroyages. Ces travaux ont déjà été élaborés aux Etats Unis avec plus ou moins de succès (Hardy et Bunnell, 1999). Dans certains départements français, une première estimation de la répartition spatiale des éclosions en fonction de la proximité d'une voie d'accès ou d'une zone urbanisée a pu être obtenue. Ainsi dans les Bouches-du-Rhône, où la très grande majorité des feux démarrent en bordure de voie d'accès ouverte à la circulation publique, on a pu démontrer que le nombre d'éclosions dans chaque commune était très bien corrélé au linéaire de voies traversant les massifs forestiers de la commune (lui-même corrélé au nombre d'habitants de la commune). La connaissance généralement partielle de cette répartition spatiale et insuffisante des causes fait qu'il n'y a pas de corrélation fiable entre la position spatiale et le type d'origine des feux et il demeure de grandes incertitudes sur certains types d'espaces. Trois objectifs sont poursuivis : - Diminuer le nombre de départs de feu. Ceci suppose de mieux connaître la répartition spatiale des points d'éclosion. - Proposer une méthode pour obtenir des informations venant compléter les données du socle commun sur les départs de feu, notamment l'identification des causes. - Développer un système d'information performant en matière de recueil de données concernant les points d'éclosion : localisation précise, description de l'environnement immédiat (végétation, relief,) et de l'occupation du sol au point d'éclosion, détermination de la cause ...

Large areas initially consisted of contiguous forests, particularly in Europe, have been influenced by human activities to a large extent. This influence contributed to the fragmentation of the rural landscape and forested areas were found surrounded by or intermixed with urban development. Urban and economic development in or near wildland vegetation poses a major threat to the environment (Johnson 2001, Radeloff et al. 2005). They are areas of humanenvironment conflicts, such as the destruction of homes by wildfires, natural habitat fragmentation, introduction of exotic species, and biodiversity decline (Radeloff et al. 2005). These areas that characterized of increased human activities and land use conversion make up the Wildland/Urban Interfaces. The significance of Wildland/Urban Interfaces (WUIs) has grown in recent years mainly because WUI, as a landscape unit, has grown worldwide (Steward et al. 2003). Essentially in USA, Canada and Australia interest for WUI study appeared after the huge fires of 1985 in the WUI (Davis, 1990) and in Europe this interest appeared ten years ago with increasing of damages on goods and people due to WUI fire. Large efforts aiming at the identification and mapping of existing or potential wildland urban interface areas have already been recorded in North America but they have to be developed in Europe. In the European project Fireparadox, several research and engineering teams are working together with a view to develop a methodology to characterize and to map WUI through spatial analysis of the territory according to regional and local approach. The objective at term is to be able to assess and to map vulnerability level or index according to each WUI types and also damages in case of fire. Theses results will allow to develop appropriated prevention actions, to help firefighters adapting their fighting strategy (concentration or dispersal of crews and facilities according to the stakes, positioning on the most vulnerable zones of interface ). The presentation presents an application of the method of WUI characterization and mapping on a case study area in Sardinia.

Le Cemagref a développé pour les gestionnaires, les bureaux d'études, les services d'aménagement des collectivités, les acteurs de la lutte (chargés de l'élaboration ou de la mise en ½uvre de plans de prévention contre les incendies, mais aussi de la lutte en cas d'incendie), une méthode pour cartographier les interfaces habitat-forêt sur de grandes surfaces et à grande échelle ; en utilisant des données de bâtis et de végétation et en s'intéressant à leur organisation spatiale, l'outil permet de définir à quel type d'interface appartient toute zone du territoire considéré ; ces informations permettent de mieux définir l'aménagement de ces espaces en fonction de leur sensibilité au feu et d'en maitriser le développement.

Southern Europe. These fires arose in or spread to human habitats at the so-called wildland-urban interface (WUI). Within the conterminous United States alone, the WUI occupies just under 10 percent of the surface and contains almost 40 percent of all housing units. Recent dry spells associated with climate variability and climate change make the impact of such catastrophic fires a matter of urgency for decision makers, scientists and the general public. In order to explore the qualitative influence of the presence of houses on fire spread, we considered only uniform landscapes and fire spread as a simple percolation process, with given house densities d and vegetation flammabilities p. Wind, topography, fuel heterogeneities, firebrands and weather affect actual fire spread. The present theoretical results would therefore, need to be integrated into more detailed fire models before practical, quantitative applications of the present results. Our simple fire-spread model, along with housing and vegetation data, shows that fire-size probability distributions can be strongly modified by the density d and flammability of houses. We highlight a sharp transition zone in the parameter space of vegetation flammability p and house density d. The sharpness of this transition is related to the critical thresholds that arise in percolation theory for an infinite domain; it is their translation into our model's finite-area domain, which is a more realistic representation of actual fire landscapes. Many actual fire landscapes in the United States appear to have spreading properties close to this transition zone. Hence, and despite having neglected additional complexities, our idealized model's results indicate that more detailed models used for assessing fire risk in the WUI should integrate the density and flammability of houses in these areas. Furthermore, our results imply that fire proofing houses and their immediate surroundings within the WUI would not only

Les litières de cinq types de peuplements sont décrites. Leurs paramètres d'inflammabilité sont estimés à partir d'expérimentations en laboratoire.

[Departement_IRSTEA]GT [TR1_IRSTEA]SET / RICOMED; Caractérisation des interfaces habitat/forêt par l'utilisation d'outil de traitement d'images et d'indices d'écologie du paysage, dans le cadre de la prévention.

Typologie du combustible à partir de données décrivant la structure et la biomasse de la végétation.

[Notes_IRSTEA]La revue n'a publié que le résumé de l'article. [Departement_IRSTEA]GT [TR1_IRSTEA]SET / RICOMED; International audience; Présentation d'une méthode de modélisation de l'aléa d'éclosion. Résumé provenant de la conférence de Figueira da Foz, Portugal.

Cet article résume les objectif du programme européen Fireparadox (6ème PCRD). Il donne également une interview d'Olivier Chandioux sur la base scientifique de Tivissa, en juillet 2006.