Projection of Forest Fire Danger due to Climate Change in the French Mediterranean Region

Varela, Vassiliki; Vlachogiannis, Diamando; Sfetsos, Athanasios; Karozis, Stelios; Politi, Nadia; Giroud, F.

doi:10.3390/su11164284

Cited by 52 publications

(42 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Drought code was the most important variable, followed by anthropogenic features, in both the LR and RF models. These results were consistent with other studies on determinant factors for the occurrence of wildfire where climatic and anthropogenic predictors had a higher influence on the fire occurrence probability [70,[89][90][91][92]. All those models were efficiently applied at a smaller scale (such as national parks or protected areas), while our models showed similar efficacy at a larger scale.…”

Section: Discussionsupporting

confidence: 91%

Forest Fire Probability Mapping in Eastern Serbia: Logistic Regression versus Random Forest Method

Milanović

Marković

Pamučar

et al. 2020

Forests

View full text Add to dashboard Cite

Forest fire risk has increased globally during the previous decades. The Mediterranean region is traditionally the most at risk in Europe, but continental countries like Serbia have experienced significant economic and ecological losses due to forest fires. To prevent damage to forests and infrastructure, alongside other societal losses, it is necessary to create an effective protection system against fire, which minimizes the harmful effects. Forest fire probability mapping, as one of the basic tools in risk management, allows the allocation of resources for fire suppression, within a fire season, from zones with a lower risk to those under higher threat. Logistic regression (LR) has been used as a standard procedure in forest fire probability mapping, but in the last decade, machine learning methods such as fandom forest (RF) have become more frequent. The main goals in this study were to (i) determine the main explanatory variables for forest fire occurrence for both models, LR and RF, and (ii) map the probability of forest fire occurrence in Eastern Serbia based on LR and RF. The most important variable was drought code, followed by different anthropogenic features depending on the type of the model. The RF models demonstrated better overall predictive ability than LR models. The map produced may increase firefighting efficiency due to the early detection of forest fire and enable resources to be allocated in the eastern part of Serbia, which covers more than one-third of the country’s area.

show abstract

Section: Discussionsupporting

confidence: 91%

Forest Fire Probability Mapping in Eastern Serbia: Logistic Regression versus Random Forest Method

Milanović

Marković

Pamučar

et al. 2020

Forests

View full text Add to dashboard Cite

show abstract

“…It is believed that the suggested interdisciplinary research approach, at least in part dealing with all the important "P"s [12], may become successful compared to approaches where only a minor fraction of the characteristic bow-tie is investigated. Committing to the suggested research agenda is therefore expected to contribute to the recent fire research within, e.g., multi-risk assessment [91], socioeconomic changes [92,93], forest fire danger rating prediction [94,95], land-use [96], and social vulnerability [97].…”

Section: Discussionmentioning

confidence: 99%

Reducing Wooden Structure and Wildland-Urban Interface Fire Disaster Risk through Dynamic Risk Assessment and Management

Log

Vandvik

Velle

et al. 2020

ASI

View full text Add to dashboard Cite

In recent years, severe and deadly wildland-urban interface (WUI) fires have resulted in an increased focus on this particular risk to humans and property, especially in Canada, USA, Australia, and countries in the Mediterranean area. Also, in areas not previously accustomed to wildfires, such as boreal areas in Sweden, Norway, and in the Arctic, WUI fires have recently resulted in increasing concern. January 2014, the most severe wooden town fire in Norway since 1923 raged through Laerdalsøyri. Ten days later, a wildfire raged through the scattered populated community of Flatanger and destroyed even more structures. These fires came as a surprise to the fire brigades and the public. We describe and analyze a proposed way forward for exploring if and how this increasing fire incidence can be linked to concomitant changes in climate, land-use, and habitat management; and then aim at developing new dynamic adaptive fire risk assessment and management tools. We use coastal Norway as an example and focus on temporal changes in fire risk in wooden structure settlements and in the Norwegian Calluna vulgaris L. dominated WUI. In this interface, the fire risk is now increasing due to a combination of land-use changes, resulting in large areas of early successional vegetation with an accumulation of biomass, and the interactive effects of climatic changes resulting in increased drought risk. We propose a novel bow-tie framework to explore fire risk and preventive measures at various timescales (years, months, weeks, hours) as a conceptual model for exploring risk contributing factors and possibilities for risk management. Ignition is the top event of the bow-tie which has the potential development towards a fire disaster as a worst case outcome. The bow-tie framework includes factors such as changes in the built environment and natural habitat fuel moisture content due to the weather conditions, WUI fuel accumulation, possibly improved ecosystem management, contribution by civic prescribed burner groups, relevant fire risk modeling, and risk communication to the fire brigades and the public. We propose an interdisciplinary research agenda for developing this framework and improving the current risk understanding, risk communication, and risk management. This research agenda will represent important contributions in paving the road for fire disaster prevention in Norway, and may provide a model for other systems and regions.

show abstract

“…The maps for each parameter are shown in Figure 4. The final map ( Figure 5) was classified by dividing the final index into five classes of risk, as follows: very low (<22), low (23)(24)(25)(26), medium (26)(27)(28), high (28)(29)(30)(31), and very high (>31) using geometrical intervals. The study area results were equally divided between the five classes.…”

Section: Ffr Assessment For 2018mentioning

confidence: 99%

“…The same final ranking of 2018 map was applied also for 2050 assessment with the classification in 5 classes of risk to make the 2 maps comparable. The map was classified as follows: very low (<22), low (23)(24)(25)(26), medium (26)(27)(28), high (28)(29)(30)(31) and very high (>31). The map for 2050 is shown in Figure 7a and highlights two main patterns: (i) the very low and low risk will shrink over the region occupying only 1% of the territory, compared to 20% of 2018 ( Figure 7c) and (ii) the spreading of the high and very high risk over more than 80% of the territory (Figure 6c), also in zones previously characterized by medium-low risk.…”

Section: Ffr Assessment For 2050mentioning

confidence: 99%

“…On the other hand, the use of spatial data with different ages, scales, and resolutions can bring uncertainties and imprecisions in the FFR assessment [22][23][24]. In many countries, such as China [25], Spain [26], France [27], Brazil [18], Portugal, Corsica, and Greece [28,29], these tools have been successfully used and incorporated in forest management plans. Also, for the Italian peninsula, some studies have been realized.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Hybrid GIS and AHP Approach for Modelling Actual and Future Forest Fire Risk Under Climate Change Accounting Water Resources Attenuation Role

et al. 2019

View full text Add to dashboard Cite

Forest wildfires usually occur due to natural processes such as lightning and volcanic eruptions, but at the same time they are also an effect of uncontrolled and illegal anthropogenic activities. Different factors can influence forest wildfires, like the type of vegetation, morphology, climate, and proximity to human activities. A precise evaluation of forest fire issues and of the countermeasures needed to limit their impact could be satisfactory especially when forest fire risk (FFR) mapping is available. Here, we proposed an FFR evaluation methodology based on Geographic Information System (GIS) and the analytic hierarchy process (AHP). The study area is the Campania region (Southern Italy) that, for the last 30 years, has been affected by numerous wildfires. The proposed methodology analyzed 12 factors, and AHP was used for weight assignment, offering a new approach to some parameters. The method divided the study area into five risk classes, from very low to very high. Validation with fire alerts showed a good correlation between observed and predicted fires (0.79 R2). Analyzing the climate projections, a future FFR for 2040 was also assessed. The proposed methodology represents a reliable screening tool to identify areas under forest fire risk, and can help authorities to direct preventive actions.

show abstract

Projection of Forest Fire Danger due to Climate Change in the French Mediterranean Region

Cited by 52 publications

References 22 publications

Forest Fire Probability Mapping in Eastern Serbia: Logistic Regression versus Random Forest Method

Forest Fire Probability Mapping in Eastern Serbia: Logistic Regression versus Random Forest Method

Reducing Wooden Structure and Wildland-Urban Interface Fire Disaster Risk through Dynamic Risk Assessment and Management

A Hybrid GIS and AHP Approach for Modelling Actual and Future Forest Fire Risk Under Climate Change Accounting Water Resources Attenuation Role

Contact Info

Product

Resources

About