[go: up one dir, main page]
Next Article in Journal
Post-Fire Vegetation (Non-)Recovery across the Edges of a Wildfire: An Unexplored Theme
Previous Article in Journal
Thirty Years of Progress in Our Understanding of the Nature and Influence of Fire in Carboniferous Ecosystems
Previous Article in Special Issue
Optimizing Wildfire Prevention through the Integration of Prescribed Burning into ‘Fire-Smart’ Land-Use Policies
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Fire
Volume 7
Issue 7
10.3390/fire7070249
fire-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Review

A Systematic Approach to Map and Evaluate the Wildfire Behavior at a Territorial Scale in the Northwestern Iberian Peninsula

by
Thais Rincón
1,
Laura Alonso
1,2,
Juan Picos
1,
Domingo M. Molina-Terrén
3 and
Julia Armesto
1,2,*
1
Escola de Enxeñaría Forestal, Universidade de Vigo, 36005 Pontevedra, Spain
2
Grupo Xestión Segura e Sostible de Recursos Minerais (XESSMin), CINTECX, Universidade de Vigo, 36310 Vigo, Spain
3
Department of Forest and Agriculture Science and Engineering, University of Lleida, Alcalde Rovira Roure 191, 25198 Lleida, Spain
*
Author to whom correspondence should be addressed.
Fire 2024, 7(7), 249; https://doi.org/10.3390/fire7070249
Submission received: 4 June 2024 / Revised: 3 July 2024 / Accepted: 11 July 2024 / Published: 13 July 2024
(This article belongs to the Special Issue Nature-Based Solutions to Extreme Wildfires)
Browse Figures
Figure 1
<p>Study area: the SW corner of Galicia (northwestern Spain).</p> "> Figure 2
<p>Map of the distribution of Rothermel fuel models in the study area (<b>left</b>) and a detailed view of a portion of the study area (<b>right</b>).</p> "> Figure 3
<p>Workflow followed to perform the study.</p> "> Figure 4
<p>Distribution of ignition points, represented as gray dots, in the study area.</p> "> Figure 5
<p>Output simulation data obtained for a sample ignition point.</p> "> Figure 6
<p>Graphical representation of wildfire size, FS (in hectares), for each ignition point considered.</p> "> Figure 7
<p>Wildfire behavior aggregated maps of the study area: (<b>A</b>) ROS in typical conditions; (<b>B</b>) FLI in typical conditions; (<b>C</b>) ROS in atypical conditions; (<b>D</b>) FLI in atypical conditions.</p> ">
Versions Notes

Abstract

:
In the current context of extreme wildfires, understanding fire behavior at a territorial level has proven crucial for territory planning. This type of analysis is usually conducted by analyzing past wildfire statistics. In this study, we forego the past information related to wildfires and analyze, instead, the behavior of the entire territory in the face of wildfires. This allows for the distribution of ignition points to be systematized and for typical and atypical weather scenarios to be considered. This analysis relies on the use of wildfire simulation software. Ignition points used for the simulations were distributed using a systematic 1 × 1 km grid throughout the whole study area. Wildfires were simulated for each ignition point using eight different weather scenarios representing both typical and atypical weather conditions. The fire behavior on the territory was analyzed using rate of spread and intensity parameters for each simulated wildfire. It was observed that this territory is extremely prone to large wildfires both in typical and atypical weather conditions and that there is a tendency for extreme behaviors to develop. Some features were identified as prevention issues that ought to be addressed. This study develops a strategy to evaluate, in a systematic manner, the response of the territory to the threat of wildfires.

1. Introduction

Over the last several decades, wildfires all over the world have caused devasting consequences such as the loss of cultural and natural heritage, negative health effects, declines in biodiversity, and drastic changes in territory [1]. It is estimated that between 2001 and 2018, 72 million km2 of land burned worldwide [2], which corresponds to half of the emerged land on earth. According to Abatzoglou et al. [3] and Turco et al. [4] burned areas are strongly projected to increase in the context of climate change. Climatic change is already affecting countries that have been frequently affected by wildfires, resulting in longer periods of fire danger and increasing the frequency of extreme fire weather [5,6,7]. These new climate conditions also affect wildfires in that they lead to prolonged heat periods, drier plant fuels, and, in some regions, higher wind speeds [8]. These factors result in more extreme wildfires in terms of area burned, duration, intensity, severity, cost of suppression, and loss of life and property [9].
In this context of large wildfire occurrence and extreme behavior, traditional wildfire management strategies like direct extinction and the creation of firebreaks seem to be insufficient [10,11]. In this line, Molina-Terren [12] pointed out that the operational capacity of extinguishing services is sometimes overwhelmed. Although firefighting efforts could be increased, they would be strongly limited by the vastness of the area affected by a fire and the rate of spread [13]. In fact, according to [14], once a wildfire reaches critical values, it becomes unfeasible to be suppressed by standard direct attack; wildfires that originated in extreme weather conditions might only by suppressed when the weather situation improves or when there is no additional fuel left to burn [15]. Numerous studies have shown that adopting an integrated approach that combines conventional fire defense measures with advanced territory and territorial planning can help to anticipate and mitigate the risks associated with this type of wildfire [16]. For example, Urza et al. [17] and Ott et al. [18] have shown that having heterogeneous landscape configurations helps to reduce fire spread. This type of fire prevention planning includes the analysis of fire behavior in the terrain in terms of area affected, intensity, and rate of spread [19]. This kind of analysis can help in promoting effective preventive and control measures [20] but also to better plan wildfire suppression activities. In fact, it has been reported that an increasing number of land managers are demanding information on wildfire behavior to design specific management actions [21,22].
In the current literature, the most common approach to analyzing fire behavior is to use fire spread models in wildfire simulation software [23]. This kind of software generates spatially explicit, predictive estimations of the fire behavior in a certain area [24]. This software can be used to understand the behavior of wildfires in a certain territory considering different starting points, weather conditions, fuels, and topography. Spatial simulations of fire spread can provide different outputs like the probability that an area has to burn once a fire has started [25], the fire spread rate, or the linear fire intensity. Conclusions about wildfire danger can be drawn by integrating this information [14]. Additionally, these output data can be used to determine the difficulty of fire suppression [26], that is, the difficulty involved in controlling and extinguishing a wildfire by firefighting personnel.
The most common strategy found in the literature for distributing ignition points in software simulations is by using past wildfire data. For example, Jahdi [27] and Thompson [28] followed this approach to perform fire simulations oriented to design fuel treatments. It was also used for simulations in the wildfire risk assessment studies by Alcasena [29], Bertomeu [30], and Sakellariou [31]. This approach can be used when past wildfire data are available and enough to allow the design of complete ignition model. However, it might be unfeasible in areas where these data are not enough or they are not available. Further, in the current context of climate change, natural ignitions (i.e., lightning) are beginning to overcome arson as the primary driver of extreme wildfires [32]. For these reasons, studies like that by Massada [33] pointed out the importance of selecting adequate ignition location models for every case. They specifically compared the effect of the empirical and random approaches in the spatial pattern of fire occurrence and spread. Some other examples are Quílez [34], Massada [33], and Ager’s [35] studies, which used random points to specifically simulate potential wildfires occasioned by lightning.
Of the fire weather variables, air temperature, relative humidity and wind all have a direct influence on the moisture content of fine dead fuel and, therefore, on fire line intensity and fire spread [36]. The extent to which these environmental factors dictate the growth of large wildfires and the development of extreme fire in a particular territory needs to be understood [37]. The weather conditions that are most commonly used for the simulations are taken from past statistical data. In fact, the previously mentioned studies use past weather statistics as the basis for their simulations. There are studies that focus exclusively on the most extreme weather conditions since they have the potential to lead to extreme wildfires [34,38]. Nevertheless, studies such as those by Bertomeu [30] and Massada [33] distinguish between typical and extreme weather conditions, since disentangling fire behavior in each scenario may help understand the wildfire risk in a certain area.
Forest fuel accumulation and fuel continuity are associated with rapidly spreading, high-intensity wildfires [39,40]. Spatially explicit information about the distribution of forest fuel within a study area is essential for performing software simulations and producing quantitative variables to describe wildfire behavior. Most of the commercial wildfire simulation software relies on Rothermel’s [41] fire spread model [42]; this is one of the most standard fuel models and is used all over the world [43,44]. Traditionally, maps reflecting fuel distribution in a study area were created using information from field surveys [45,46]. However, currently, remote sensing techniques are becoming more and more common for performing this task [47,48]. These techniques allow for the acquisition of up-to-date maps at broad scales, which is essential for obtaining reliable predictions of fire behavior and for designing management practices at large scales [46,47].
In this study, an in-depth geospatial analysis of fire behavior in the territory is carried out using simulation techniques for typical (normal, mean values) and atypical (extreme or abnormal values) weather conditions. A systematized ignition point approach is defined consisting of a grid of starting points that is regularly distributed throughout the study area. This approach allows us to evaluate the response of the territory to wildfires. The obtained simulations are aggregate (distinct ignition points and weather conditions), and the results are evaluated numerically and geospatially.

2. Materials and Methods

2.1. Case Study

The study area is in Galicia, a region in the northwest of Spain. Sixty-nine percent of the surface area of Galicia is covered by forests [49], and it is one of the Spanish regions that is most prone to forest fires. The official report published for the period between 2006 and 2015 indicates that Galicia accounts for 23% of the total burned area in Spain. In recent years, there have been several extreme fire events in the region. In 2022, according to the Department of the Rural Environment of the regional government, a total of 51,642 ha burned, making that year, along with 2017, one of the most devastating in history in terms of wildfire behavior [50]. The study area selected is shown in Figure 1. Official forest fire statistics reveal that forest fires have been occurring more frequently and to a greater extent in recent years [51].
The study area is defined by natural and artificial barriers, bounded by the sea to the north and west, a river mouth to the south, and valley covered mostly by anthropogenic areas. The topographical relief of the area includes two features: the Argallo range, with a maximum altitude of 416 m; and A Groba, with a maximum altitude of 648 m. The two features are separated by a valley. The eastern side is dominated by disperse settlements. In its entirety, the study area comprises a total of 34,612 ha.
The study area has an Atlantic climate [52]. Forested areas prevail, especially shrub and broadleaf forests (Quercus sp. and riparian forests) and forest plantations (Eucalyptus sp. and Pinus sp.) [53]. The study area is not densely populated (with a mean of 132 inhabitants per square kilometer [54]; however, the inhabitants are dispersed into small settlements and isolated constructions, making the study area a sensitive area in terms of its WUI (Wildland–Urban Interface).

2.2. Materials

2.2.1. Source Data

The input data used to perform the simulations are summarized in Table 1. A digital terrain model (DTM) was used to obtain topographic-related information that is required to perform the simulations [55]. The DTM was obtained in an open-access regime from the National Geographic Institute of Spain [55]. It has a resolution of 5 × 5 m.
The fuel model map used in this study was that developed by Solares-Canal [56]. That study describes a methodology for producing updated fuel model maps (Rothermel fuel models) for Atlantic regions using remotely sensed data. The study used Sentinel-2 multispectral images and airborne LiDAR data, both acquired in 2019, to identify the types of fuels present in the territory and their vertical structures. In order to establish relationships between remote sensing data, Rothermel fuel models and the fuel situations present in the study area of the Lourizan Research Center Photo guide [57] were used. This photo guide describes in detail the fire behavior for the different fuel situations present in Galicia. The established relationships were used at the pixel level to identify and map the corresponding Rothermel fuel models at a spatial resolution of 5 m and thus obtain a fuel model map of the study area (Figure 2). The accuracy of the map was estimated through two different procedures. The overall accuracy resulted in 98% for models 7 and 10 and 90% for the other models (see [56] for details).
The most frequent fuel models are crops and pasture (Rothermel fuel model 2) and tree stands with continuous vertical structure (Rothermel fuel model 7), with relative extents of 30.43% and 23.64%, respectively. A complete list of the models in the study area and their extensions in absolute and relative values is presented in Table 2.
The weather data were obtained from a weather station in the municipality of O Rosal, Pontevedra, which lies within the study area. This weather station belongs to the weather Observation and Prediction Unit of Galicia [52], a public organization affiliated with the regional government. The weather stations in this network record several variables daily. For this study, the maximum temperature, minimum relative humidity, prevailing wind speed, speed of wind gusts, prevailing wind direction, and direction of wind gusts were downloaded for the months of June, July, August, and September for the years 2011–2021.

2.2.2. Software

The fire simulation software used in this study was the Wildfire Analyst 2.9 © (WFA) software [58]. It is a simulator that is commonly documented in the scientific literature and in operational management services and has been used both in real firefighting operations and in fire prevention analyses [59]. The geospatial processes were performed in ArcGIS [60].

2.3. Methodology

The workflow followed in this study is presented in Figure 3. As shown in the figure, the methodology consisted of the following steps:
-
Territory modeling: The topographic parameters and the fuel model map were generated and introduced into the simulation software.
-
Definition of weather scenarios: Different scenarios were built to model the habitual weather conditions in the study area as well as conditions that can occur in extreme weather scenarios.
-
Fire modeling: Ignition points were defined as well as the duration of simulations and the absence of fire extinction.
-
Geospatial analysis of wildfire behavior.

2.3.1. Territory Modelling

The main variables that affect wildfire behavior are those related to topography and forest fuels. The topographic parameters were derived from the DTM raster file. The ArcGIS tools hill shade, aspect, and slope were used to obtain the exposure, aspect, and slope raster layers, respectively. These layers were obtained at a 5 × 5 m spatial resolution. They were used as input data for the WFA 2.9 simulation software.
The fuel model map is a raster layer where the digital value of each pixel corresponds to a Rothermel fuel model. The spatial resolution of this map is also 5 m. The map was also used directly as input data for the WFA 2.9 simulation software.

2.3.2. Definition of Weather Scenarios

Eight weather scenarios were constructed to model the typical and atypical weather conditions that may be expected in the study area. The scenarios were built using the weather records described in the materials section for the period between 2011 and 2021. Four predominant wind directions were identified in the study area for this period. One scenario was built for each wind direction using two different criteria: typical weather conditions and atypical weather conditions. As a result, two sets of four scenarios were built. The scenarios are summarized in Table 3. The criteria were as follows:
-
Typical conditions: The daily values for maximum temperature, minimum relative humidity, and prevailing wind speed were averaged to obtain four monthly values. These values were used for the scenarios that might correspond to the typical weather conditions in the study area.
-
Atypical conditions: The most adverse single record from the daily values was selected for maximum temperature, minimum relative humidity, and maximum wind speed. These values were used to build four additional scenarios that might correspond to anomalous or extreme conditions in the study area.

2.3.3. Fire Modeling

Fire simulation requires the establishment of fire-starting points, also known as ignition points. The past wildfires’ available data in the study area are not enough in number to build a complete ignition model. However, the lack of past wildfires does not mean that they could not occur in the future. In this study, the ignition points were distributed systematically across a 1000 × 1000 m grid to obtain a homogenous density of points in the study area. As a result, 355 points were defined, regularly dispersed (see Figure 4). Fire simulations were carried out individually for each ignition point and for each weather scenario. All the simulations were initialized at 13:00 h (ignition detection time) and executed allowing for eight hours of propagation time without fire-fighting actions. The weather values were considered to remain constant throughout this eight-hour propagation time frame. The simulations were executed with a 33 m cell.
Each simulation results in a vector polygon that corresponds to the perimeter that the fire reaches after the propagation time frame. The attribute table of the vector layer contains the total area of the polygon in hectares, which corresponds to the fire size (FS). Additionally, two raster layers with a 33 m spatial resolution are obtained: one corresponding to the rate of spread (ROS) and the other to the front intensity (FLI). The digital value of each pixel in these layers corresponds to the quantitative value of the variable for that pixel; the rate of spread is expressed in kilometers per hour and the front intensity in Kilowatts per meter. As a result of all the simulations run, eight ROS raster layers were obtained for each ignition point, one per weather scenario, and eight FLI raster layers, also one per weather scenario.

2.3.4. Geospatial Analysis of Wildfire Behavior

The simulation results were analyzed separately and jointly to provide a detailed overview of the behavior of wildfires in the study area. First, the size of the resulting wildfires was evaluated. The area of each resulting contour polygon was assigned to its corresponding starting point. Then, the four values for the typical weather conditions were averaged to obtain the mean size fire generated at each starting point given typical weather. The four values corresponding to atypical weather conditions were also averaged to obtain the mean size fire for each starting point given extreme weather. The graphic representation of these results depicts the geographic distribution of the most critical ignition points. Fires started at critical points have the potential to develop into large wildfires, given the weather conditions considered.
Second, the ROS raster layers for each ignition point and for each scenario were analyzed. Given the extension of the simulated fires, and the relative proximity of the starting points, close starting points frequently resulted in overlapping fire contours; consequently, the ROS raster layers overlapped as well.
To obtain a single ROS raster layer for each weather scenario, the maximum ROS value for each pixel was selected from among all the ROS values generated for that pixel throughout the simulations using all of the different starting points. Each of the eight resulting ROS raster layers, one corresponding to each weather scenario, was a compilation of these maximum values at each pixel. Then, the four layers corresponding to the typical weather conditions were averaged to obtain the typical weather mean ROS for each pixel, and the four layers corresponding to the atypical weather conditions were averaged as well to obtain the atypical weather mean ROS value for each pixel.
Third, the FLI raster layers were analyzed. This was carried out through an analogous process. For each weather scenario, the maximum FLI value for each pixel was selected from among all the FLI values generated for that pixel throughout the simulations using all the different starting points. Each of the eight resulting FLI raster layers, one corresponding to each weather scenario, was a compilation of the maximum FLI values at each pixel. Then, the four layers corresponding to the typical weather conditions were averaged to obtain the typical weather mean FLI value for each pixel, and the four layers corresponding to the atypical weather conditions were averaged as well to obtain the atypical weather mean FLI value for each pixel.
Finally, the ROS and FLI maps were used to evaluate suppression capabilities. According to [26,61], in low intensity and low or medium ROS situations, personnel can easily contain fires, suggesting that fuel treatment can be obviated [62]. However, high-intensity fires and high rates of spread can exceed suppression capabilities, causing significant damage in dense forests and threatening communities [63]. Alexander and Lanoville [64] proposed FLI threshold values that could be used as references to establish whether a wildfire be classified as very difficult, extremely difficult, or virtually impossible to suppress.

3. Results

3.1. Weather Scenarios

The evaluation of wind direction frequencies in the time series analyzed revealed that south (180°), southwest (225°), northeast (45°), and northwest (315°) directions are the predominant directions in the study area. These four directions were used to build the scenarios for the typical and atypical weather conditions.

3.2. Fire Modelling

A total of 2840 simulations were run for the study area. As a result of the proposed simulation design, each pixel with forest fuel was burned between 1 and 8 times. Starting points that correspond to pixels with no fuels (Model 0) do not generate any output. Figure 5 shows the output data obtained for a sample ignition point: isochrone lines, ROS, and FLI.

3.3. Geospatial Analysis of Wildfire Behavior

The fire sizes (FS) resulting from the simulation of each weather scenario are summarized in Table 4 as the proportion of ignition points that generated fires of each of the different sizes. The corresponding graphical representations, one per weather scenario, are shown in Figure 6. These results established the geographic distribution of the ignition points where, if a fire should start, it could potentially develop into a large wildfire under the given weather conditions.
The FS quantitative results show that 20.3% to 22.3% of the ignition points, depending on the scenario considered, do not generate a wildfire, as they do not have sufficient surrounding fuel. When forest fuels are available, 38.3% of the ignition points considered generate large wildfires (FS > 500 ha) in typical weather conditions. The minimum proportion (31.8%) corresponds to the NE wind direction (scenario II), which is the most frequent wind direction. The other three scenarios present similar ratios, ranging between 40.2% and 41.2%. The graphic representation reveals that ignition points in the northern valley, which is the most densely populated in the area, along with the riverside and the coastal boundary are the least prone to generating large wildfires.
In atypical conditions, the proportion of ignition points that result in large wildfires is 71.5%, with values ranging between 66.5% and 73.8%. The most favorable scenario corresponds to the NW wind direction (scenario VII). It is worth noting that in three of the simulated scenarios, more than 52% of the ignition points generate a wildfire that affects more than 2000 hectares. In these atypical conditions, the geographic representation of the FS results does not show any topographic pattern.
The ROS and FLI results are presented in Table 5, Table 6 and Table 7. The first one shows the distribution of ROS and FLI values of the wildfires simulated in each weather scenario. Table 6 shows the behavior of each fuel model, in terms of ROS and FLI values reported, in the simulations using typical weather conditions. According to these values, wildfires propagate with velocities over 2 km/h in 8.5% of the study area in typical conditions. Shrubs are the main landcover where fires propagate faster than this threshold value: in 53.4% and 41.0% of the study area, for fuel models 4 and 5, respectively. In these conditions, 34.4% of the area analyzed presents FLI values over 10,000 kW/m, which is considered the extreme threshold. Again, shrubs are the main land cover that generate FLI values above this threshold when they burn: in 98.9% and 86.0% of the area, for fuel models 4 and 5, respectively. No differences, in this term, can be appreciated between the two weather scenarios.
Table 7 presents the behavior of each fuel model, in terms of ROS and FLI values reported, in the simulations using atypical weather conditions. According to the results for these conditions, the simulated wildfires propagate with velocities over 2 km/h in 89.9% of the study area. The main fuel models that propagate wildfires faster than this extreme ROS threshold are 1, 4, and 5, in 99.3%, 99.9%, and 95.5% of the simulations, respectively. FLI values over 10,000 kW/m are reported in 37.8% of the study area. As in the previous case, shrubs are the main landcover that generate FLI values above this value when burned: in 98.8% and 88.8% of the simulations, for models 4 and 5, respectively. In this case, the SW wind direction (Scenario VIII) is more favorable than the others, meaning that it yields lower values for both ROS and FLI.
The resulting ROS and FLI maps for typical and atypical weather conditions are shown in Figure 7. They show that in typical weather conditions, the simulated wildfires present more favorable behavior in the northern valley, which is the most densely populated. The highest ROS values are found in the Serra da Groba. The highest FLI values are found both in the Serra da Groba and in the Serra do Argallo. The aggregated maps of the simulations in atypical conditions show that most of the area analyzed would propagate the wildfires at an extreme velocity, and the greatest FLI values would be reported mainly in the central areas with topographic relief.
Considering the FLI thresholds provided by Lanoville et al. [64], in atypical conditions, wildfire control would be very or extremely difficult in 59.0% of the study area, and fires would exceed extinction capacity in 37.8% of the study area. In typical weather conditions, the ratios would be 1.5% and 34.4%, respectively.

4. Discussion

One of the novelties of this study is the use of systematic grid-based criteria to distribute the ignition points, which has allowed us to completely saturate the study area, obtaining a fire modeling domain of the territory. In contrast, wildfire simulation studies usually use past registers or random distribution of ignitions points [30,34]. However, studies like Massada et al. [33,65] or Zigner et al. [66] showed that the distribution approach significantly affects the resulting surface of burned areas and spatial predictions of fire spread patterns. The systematic criteria provided a regular distribution of ignition points, allowing to equalize the potential burn probability in the whole territory (except the boundary area). Since the weather conditions have been kept constant during the simulations, the burning of a certain pixel in this approach depends on its topography and fuel; consequently, their incidence in the final burned area and behavior could be analyzed.
The geospatial cross analysis of wildfire behavior variables and forest fuels and the DMT has shown that the areas more prone to wildfires with high FLI and ROS values correspond to areas with models 4 and 7 and areas with steeper slopes. This result is especially evident on the slopes of the Serra da Groba and the Serra do Argallo (see Figure 7). In fact, these ranges have been affected by large wildfires in the past years, while the surroundings have not [51,67]. Models 4 and 7 correspond to shrubs and forests with a vertical continuous structure in the study area. These broad categories include varying species and fuel loads according to the Photo Guide for Galician Vegetation Situations [57]. A more detailed analysis of the vegetation in those areas that are prone to critical wildfire behavior would be interesting; it could be accomplished through field work, remote sensing sources, and finer fuel models like [68] or [69].
The numerical results demonstrate that, as expected, under atypical weather conditions, a great proportion of ignition points lead to large wildfires (71.5%) and extreme behaviors. However, it should be mentioned that even in typical weather conditions, 38.3% of the ignition points lead to large wildfires. Additionally, these conditions can also lead to high rates of spread and high fire line intensities. These results attest to the need to integrate typical conditions into the study of territory behavior in the face of wildfires. This result aligns with the studies of Bertomeu et al. [30] and Massada et al. [33] that also indicate the importance of distinguishing between typical and extreme weather conditions rather than considering only atypical conditions, as is suggested by other authors such as Quílez [34] and Botequim [38]. Furthermore, since the results are reported in a geospatially explicit format, the areas identified as the most vulnerable in terms of wildfire behavior in typical weather conditions should be considered potential priority targets for carrying out fire prevention activities.
Analyzing FLI and ROS at the territory scale has also shed light on the reality of fire control and suppression capabilities. Fires in a large percentage of the area would be insuppressible both in typical and atypical weather conditions, with rates of 34.4% and 37.8%, respectively. These percentages reveal that the area is quite vulnerable to wildfires. The specific areas that are most vulnerable could be further explored by incorporating geospatial data related to population distribution and sites of natural and socioeconomic interest to be protected. This would provide a more comprehensive vulnerability analysis that would help in the design of wildfire prevention measures [70].
Multiple studies have used fire simulations to evaluate the behavior of the territory, validate spread models, and establish intervention zones in areas that have been traditionally affected by wildfires, such as ref. [71] in Catalonia, [72] in Valencia, and [73] in California. The methodology presented in this study could be especially interesting in areas where there is lack of past data or past wildfire. The results could be useful to support the planning of the wildfire prevention strategies in those areas, such as establishing the location and designing the management of strategic management areas (SMAs) described in [70], which are mandated by Spanish legislation [74].

5. Conclusions

The proposed methodology provides an in-depth analysis of fire behavior in the study area through simulation techniques considering typical and atypical weather conditions. A grid-based distribution of ignition points was used to simulate wildfires. This approach allowed for a comprehensive analysis of the entire study area.
The proposed methodology has revealed that the territory in question is highly prone to large wildfires and that extreme fire behaviors can be expected in both typical and atypical weather conditions. This is relevant in the current context of climate change. It highlights the importance of evaluating wildfire behavior in both normal and exceptional weather circumstances. Furthermore, factors were identified, including fuel types and topographic features, that are correlated with high values of wildfire behavior variables (FS, ROS, FLI). Since these wildfire behavior variables were recorded in a geospatially explicit format and the results of the simulations were integrated into single maps, specific areas of the territory were identified that could potentially be prioritized when it comes to designing wildfire prevention measures.
Future studies should apply the proposed workflow to evaluate wildfire prevention measures oriented at managing forest fuels. A comparison of the results with the current fuel conditions and with potential changing fuel conditions, along with a cross analysis including other geospatial data sources, would be useful for supporting the design of wildfire prevention plans, prioritizing management areas, and improving the resilience of the territory after wildfires.

Author Contributions

Conceptualization, J.P. and J.A.; methodology, J.P., J.A. and T.R.; validation, T.R., J.P., J.A., D.M.M.-T. and L.A.; formal analysis, T.R., L.A., J.A. and J.P.; research, T.R., L.A., J.A. and J.P.; resources, J.P. and J.A.; writing—writing of the original draft, T.R., L.A., J.A., J.A., J.P. and D.M.M.-T.; writing—revising and editing, T.R., L.A., J.A., J.P. and D.M.M.-T.; visualization, T.R. and L.A.; supervision, J.P., J.A. and D.M.M.-T.; project administration, J.P. and J.A.; fundraising, J.P. and J.A. All authors have read and agreed to the published version of the manuscript.

Funding

This research is part of the project “Prevention and management of the landscape exposed to large forest fires in cross-border rural areas between Spain and Portugal”, Program EP-INTERREG V, grant FIREPOCTEP+, and the project “Paleointerface: Strategic element for the prevention of forest fires. Development of multispectral and 3d analysis methodologies for integrated management” project, funded by MICIU/AEI/10.13039/501100011033, Spanish Ministry of Sciences and Innovation, Grant code PID2019-111581RB-I00; it is also supported by an FPU grant from the Spanish Ministry of Sciences and Innovation under Grant FPU19/02054.

Data Availability Statement

The raw data supporting the conclusions of this article will be made available by the authors on request.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Lannom, K.O.; Tinkham, W.; Smith, A.; Abatzoglou, J.; Nwingham, B.; Hall, T.; Morgan, P.; Strand, E.; Paveglio, T.; Anderson, J.; et al. Defining extreme wildland fires using geospatial and ancillary metrics. Int. J. Wildland Fire 2014, 23, 322–337. [Google Scholar] [CrossRef]
  2. Food and Agriculture Organization of the United Nations (FAO). Evaluación de los recursos forestales mundiales 2020. In Informe Principal; FAO: Roma, Italy, 2021. [Google Scholar] [CrossRef]
  3. Abatzoglou, J.T.; Williams, A.P.; Barbero, R. Global Emergence of Anthropogenic Climate Change in Fire Weather Indices. Geophys. Res. Lett. 2019, 46, 326–336. [Google Scholar] [CrossRef]
  4. Turco, M.; Rosa-Cánovas, J.J.; Bedia, J.; Jerez, S.; Montávez, J.P.; Llasat, M.C.; Provenzale, A. Exacerbated fires in Mediterranean Europe due to anthropogenic warming projected with non-stationary climate-fire models. Nat. Commun. 2018, 9, 3821. [Google Scholar] [CrossRef] [PubMed]
  5. Dowdy, A.J. Climatological Variability of Fire Weather in Australia. J. Appl. Meteor. Clim. 2018, 57, 221–234. [Google Scholar] [CrossRef]
  6. Couto, F.T.; Santos, F.L.M.; Campos, C.; Andrade, N.; Purificação, C.; Salgado, R. Is Portugal Starting to Burn All Year Long? The Transboundary Fire in January 2022. Atmosphere 2022, 13, 1677. [Google Scholar] [CrossRef]
  7. Wang, X.; Thompson, D.K.; Marshall, G.A.; Tymstra, C.; Carr, R.; Flannigan, M.D. Increasing frequency of extreme fire weather in Canada with climate change. Clim. Chang. 2015, 130, 573–586. [Google Scholar] [CrossRef]
  8. Intergovernmental Panel on Climate Change (IPCC). Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change. 2021. Available online: https://www.ipcc.ch/report/ar6/wg1/downloads/report/IPCC_AR6_WGI_SummaryVolume.pdf (accessed on 13 February 2023).
  9. Modugno, S.; Balzter, H.; Cole, B.; Borrelli, P. Mapping regional patterns of large forest fires in wildland–urban interface areas in Europe. J. Environ. Manag. 2016, 172, 112–126. [Google Scholar] [CrossRef] [PubMed]
  10. Ortega, M.; Silva, F.R.; Molina, J.R. Modeling fuel break effectiveness in southern Spain wildfires. Fire Ecol. 2024, 20, 40. [Google Scholar] [CrossRef]
  11. San-Miguel-Ayanz, J.; Moreno, J.M.; Camia, A. Analysis of Large Fires in European Mediterranean Landscapes: Lessons Learned and Perspectives. For. Ecol. Manag. 2013, 294, 11–22. [Google Scholar] [CrossRef]
  12. Molina-Terrén, D.M.; Xanthopoulos, G.; Diakakis, M.; Ribeiro, L.; Caballero, D.; Delogu, G.M.; Viegas, D.X.; Silva, C.A.; Cardil, A. Analysis of Forest Fire Fatalities in Southern Europe: Spain, Portugal, Greece and Sardinia (Italy). Int. J. Wildl. Fire 2019, 28, 85–98. [Google Scholar] [CrossRef]
  13. Palaiologou, P.; Kalabokidis, K.; Ager, A.A.; Day, M.A. Development of Comprehensive Fuel Management Strategies for Reducing Wildfire Risk in Greece. Forests 2020, 11, 789. [Google Scholar] [CrossRef]
  14. Hirsch, K.G.; Martell, D.L. A review of initial attack fire crew productivity and effectiveness. Int. J. Wildl. Fire 1996, 6, 199–215. Available online: https://www.publish.csiro.au/wf/WF9960199 (accessed on 12 July 2024). [CrossRef]
  15. Finney, M.; Grenfell, I.; McHugh, C. Modeling containment of large wildfires using generalized linear mixed-model analysis. For. Sci. 2009, 55, 249–255. Available online: https://www.fs.usda.gov/research/treesearch/33778 (accessed on 12 July 2024). [CrossRef]
  16. Fernandes, P.M. Empirical Support for the Use of Prescribed Burning as a Fuel Treatment. Curr. For. Rep. 2015, 1, 118–127. [Google Scholar] [CrossRef]
  17. Urza, A.K.; Hanberry, B.B.; Jain, T.B. Landscape-scale fuel treatment effectiveness: Lessons learned from wildland fire case studies in forests of the western United States and Great Lakes region. Fire Ecol. 2023, 19, 1. [Google Scholar] [CrossRef]
  18. Ott, J.E.; Kilkenny, F.F.; Jain, T.B. Fuel treatment effectiveness at the landscape scale: A systematic review of simulation studies comparing treatment scenarios in North America. Fire Ecol. 2023, 19, 10. [Google Scholar] [CrossRef]
  19. Beverly, J.; McLoughlin, N. Burn probability simulation and subsequent wildland fire activity in Alberta, Canada Implications for risk assessment and strategic planning. For. Ecol. Manag. 2019, 451, 117490. [Google Scholar] [CrossRef]
  20. Salis, M.; Del Giudice, L.; Arca, B.; Ager, A.A.; Alcasena-Urdiroz, F.; Lozano, O.; Bacciu, V.; Spano, D.; Duce, P. Modeling the effects of different fuel treatment mosaics on wildfire spread and behavior in a Mediterranean agro-pastoral area. J. Environ. Manag. 2018, 212, 490–505. Available online: https://www.sciencedirect.com/science/article/abs/pii/S0301479718301191?via%3Dihub (accessed on 12 July 2024). [CrossRef] [PubMed]
  21. Lozano, O.; Salis, M.; Ager, A.; Arca, B.; Alcasena, F.; Monteiro, A.; Finney, M.; Del Giudice, L.; Scoccimarro, E.; Spano, D. Assessing Climate Change Impacts on Wildfire Exposure in Mediterranean Areas. Risk Anal. 2017, 37, 1898–1916. Available online: https://onlinelibrary.wiley.com/doi/abs/10.1111/risa.12739 (accessed on 12 July 2024). [CrossRef]
  22. Alcasena, F.; Ager, A.; Salis, M.; Day, M.; Vega-Garcia, C. Optimizing prescribed fire allocation for managing fire risk in central Catalonia. Sci. Total Environ. 2018, 621, 872–885. Available online: https://www.sciencedirect.com/science/article/abs/pii/S0048969717333545?via%3Dihub (accessed on 12 July 2024). [CrossRef]
  23. Miller, C.; Ager, A. A review of recent advances in risk analysis for wildfire management. Int. J. Wildl. Fire 2013, 22, 1–14. Available online: https://www.researchgate.net/publication/273741060_A_review_of_recent_advances_in_risk_analysis_for_wildfire_management (accessed on 12 July 2024). [CrossRef]
  24. Alcasena, F.; Salis, M.; Ager, A.; Castell, R.; Vega-García, C. Assessing wildland fire risk transmission to communities in Northern Spain. Forests 2017, 8, 30. Available online: https://www.mdpi.com/1999-4907/8/2/30 (accessed on 12 July 2024). [CrossRef]
  25. Ager, A.; Vaillant, N.; Finney, M. Integrating Fire Behavior Models and Geospatial Analysis for Wildland Fire Risk Assessment and Fuel Management Planning. J. Combust. 2011, 2011, 572452. [Google Scholar] [CrossRef]
  26. Wotton, W.; Flannigan, M.; Marshall, G. Potential climate change impacts on fire intensity and key wildfire suppression thresholds in Canada. Environ. Res. Lett. 2017, 12, 095003. [Google Scholar] [CrossRef]
  27. Jahdi, R.; Del Giudice, L.; Melis, M.; Lovreglio, R.; Salis, M.; Arca, B.; Duce, P. Assessing the effects of alternative fuel treatments to reduce wildfire exposure. J. For. Res. 2023, 34, 373–386. [Google Scholar] [CrossRef]
  28. Thompson, M.P.; Vogler, K.C.; Scott, J.H.; Miller, C. Comparing risk-based fuel treatment prioritization with alternative strategies for enhancing protection and resource management objectives. Fire Ecol. 2022, 18, 26. [Google Scholar] [CrossRef]
  29. Alcasena, F.J.; Salis, M.; Vega-García, C. A fire modeling approach to assess wildfire exposure of valued resources in central Navarra, Spain. Eur J. For. Res 2016, 135, 87–107. [Google Scholar] [CrossRef]
  30. Bertomeu, M.; Pineda, J.; Pulido, F. Managing Wildfire Risk in Mosaic Landscapes: A Case Study of the Upper Gata River Catchment in Sierra de Gata, Spain. Land 2022, 11, 465. [Google Scholar] [CrossRef]
  31. Sakellariou, S.; Sfougaris, S.; Christopoulou, O.; Tampekis, S. Integrated wildfire risk assessment of natural and anthropogenic ecosystems based on simulation modeling and remotely sensed data fusion. Int. J. Disaster Risk Reduct. 2022, 78, 103129. [Google Scholar] [CrossRef]
  32. Rodrigues, M.; Cunill Camprubí, A.; Balaguer-Romano, R.; Coco Megía, C.J.; Castañares, F.; Ruffault, J.; Fernandes, P.M.; de Dios, V.R. Drivers and implications of the extreme 2022 wildfire season in Southwest Europe. Sci. Total Environ. 2023, 859, 160320. [Google Scholar] [CrossRef]
  33. Bar-Massada, A.; Syphard, A.D.; Hawbaker, T.J.; Stewart, S.I.; Radeloff, V.C. Effects of Ignition Location Models on the Burn Patterns of Simulated Wildfires. Environ. Model. Softw. 2011, 26, 583–592. [Google Scholar] [CrossRef]
  34. Quílez, R.; Valbuena, L.; Vendrell, J.; Uytewaal, K.; Ramirez, J. Establishing Propagation Nodes as a Basis for Preventing Large Wildfires: The Proposed Methodology. Front. For. Glob. Chan. 2020, 3, 548799. [Google Scholar] [CrossRef]
  35. Ager, A.; Day, M.; Finney, M.; Vance-Borland, K.; Vaillant, N. Analyzing the transmission of wildfire exposure on a fire-prone landscape in Oregon, USA. For. Ecol. Manag. 2014, 334, 377–390. [Google Scholar] [CrossRef]
  36. Duane, A.; Brotons, L. Synoptic weather conditions and changing fire regimes in a Mediterranean environment. Agric. For. Meteorol. 2018, 253, 190–202. [Google Scholar] [CrossRef]
  37. Liu, Y.; Stanturf, J.; Goodrick, S. Trends in global wildfire potential in a changing climate. Int. J. Wildl. Fire 2010, 259, 685–697. [Google Scholar] [CrossRef]
  38. Botequim, B.; Fernandes, P.; Borges, J.; González-Ferreiro, E.; Guerra, J. Improving silvicultural practices for Mediterranean forests through fire behaviour modelling using LiDAR-derived canopy fuel characteristics. Int. J. Wildl Fire 2019, 28, 823–839. [Google Scholar] [CrossRef]
  39. Moreira, F.; Ascoli, D.; Safford, H.; Adams, M.A.; Moreno, J.M.; Pereira, J.; Catry, F.X.; Armesto, J.; Bond, W.; González, M.E.; et al. Wildfire management in Mediterranean-type regions: Paradigm change needed. Environ. Res. Lett. 2020, 15, 011001. [Google Scholar] [CrossRef]
  40. Montiel, M.C.; Karlsson, M.O.; Galiana, M.L. Regional fire scenarios in Spain: Linking landscape dynamics and fire regime for wildfire risk management. J. Environ. Manag. 2019, 233, 427–439. [Google Scholar] [CrossRef] [PubMed]
  41. Rothermel, R. A Mathematical Model for Predicting Fire Spread in Wildland Fuels; INT-115; Department of Agriculture, Intermountain Forest and Range Experiment Station: Ogden, UT, USA, 1972; 40p. [Google Scholar]
  42. Simons, N.E. Improving Decision Making during Wildland Fire Events (Order No. 3602215). Ph.D. Thesis, University of California, Santa Barbara, Berkeley, CA, USA, 2013. Available online: https://www.proquest.com/dissertations-theses/improving-decision-making-during-wildland-fire/docview/1468440376/se-2 (accessed on 12 July 2024).
  43. Ascoli, D.; Vacchiano, G.; Motta, R.; Bovio, G. Building Rothermel fire behavior fuel models by genetic algorithm optimization. Int. J. Wildl. Fire 2015, 24, 317–328. [Google Scholar] [CrossRef]
  44. Vacchiano, G.; Ascoli, D. An Implementation of the Rothermel Fire Spread Model in the R Programming Language. Fire Technol. 2015, 51, 523–535. [Google Scholar] [CrossRef]
  45. Anderson, H. Aids to Determining Fuel Models for Estimating Fire Behavior; INT-GTR-122; U.S. Department of Agriculture, Forest Service, Intermountain Forest and Range Experiment Station: Ogden, UT, USA, 1982; 22p. [Google Scholar]
  46. Arroyo, L.; Pascual, I.; Manzanera, J. Fire models and methods to map fuel types: The role of remote sensing. For. Ecol. Manag. 2008, 256, 1239–1252. [Google Scholar] [CrossRef]
  47. Matthew, G.; Geoffrey, J.; Van Dijk, A.; Yebra, M. Forest fire fuel through the lens of remote sensing: Review of approaches, challenges and future directions in the remote sensing of biotic determinants of fire behavior. Remote Sens. Environ. 2021, 255, 112282. [Google Scholar] [CrossRef]
  48. Szpakowski, D.M.; Jensen, J.L.R. A Review of the Applications of Remote Sensing in Fire Ecology. Remote Sens. Environ. 2019, 11, 2638. [Google Scholar] [CrossRef]
  49. Ministerio para la Transición Ecológica y el Reto Demográfico (MITECO). Anuario de Estadística Forestal. 2020. Available online: https://www.miteco.gob.es/content/dam/miteco/es/biodiversidad/estadisticas/anuario_ef2020_tcm30-559705.pdf (accessed on 12 July 2024).
  50. de Galicia, X. Plan de Prevención y Defensa Contra Incendios Forestales de Galicia (PLADIGA). 2023. Available online: https://mediorural.xunta.gal/sites/default/files/temas/forestal/pladiga/2023/01_Memoria_Pladiga_Castellano_2023.pdf (accessed on 12 July 2024).
  51. Regos, A. Cartografía de áreas queimadas en Galicia no século XXI: Presentación do produto e aplicación web mapping. NACC 2018, 25, 45–53. Available online: https://revistas.usc.gal/index.php/nacc/article/view/5150 (accessed on 12 July 2024).
  52. METEOGALICIA. Red Meteorological. Available online: https://www.meteogalicia.gal/observacion/estacionshistorico/historico.action?idEst=10091# (accessed on 10 September 2022).
  53. Ministerio Para la Transición Ecológica y el Reto Demográfico (MITECO). Mapa Forestal de España (MFE) de Máxima Actualidad. 2011. Available online: https://www.miteco.gob.es/es/cartografia-y-sig/ide/descargas/biodiversidad/mfe.aspx (accessed on 12 July 2024).
  54. Instituto Galego de Estadística. Censo de Población y Vivienda. Available online: https://www.ige.gal/web/mostrar_actividade_estatistica.jsp?codigo=0201001001 (accessed on 9 December 2022).
  55. Organismo Autónomo Centro Nacional de Información Geográfica (CNIG). Modelo Digital del Terreno (MDT05); Centro de descargas, Instituto Geografico Nacional. España. Available online: https://centrodedescargas.cnig.es/CentroDescargas/index.jsp (accessed on 5 November 2022).
  56. Solares-Canal, A.; Alonso, L.; Rincón, T.; Picos, J.; Molina-Terrén, D.M.; Becerra, C.; Armesto, J. Operational fuel model map for Atlantic landscapes using ALS and Sentinel-2 images. Fire Ecol. 2023, 19, 61. [Google Scholar] [CrossRef]
  57. Arellano, S.; Vega, J.A.; Ruiz, A.D.; Arellano, A.; Álvarez, J.G.; Vega, D.J.; Pérez, E. Foto-Guía de Combustibles Forestales de Galicia y Comportamiento del Fuego Asociado; Andavira Editora, S.L.: Santiago de Compostela, Spain, 2017; Available online: https://lourizan.xunta.gal/es/transferencias/foto-guia-de-combustibles-forestales-de-galicia (accessed on 8 June 2023).
  58. Technosylva. Wildfire Analyst Software, 2.9; Technosylva: La Jolla, CA, USA, 2014.
  59. Alan, A.; Finney, M.; Mark, A. Application of wildfire simulation models for risk analysis. Geophys. Res. Abstr. 2009, 11, EGU2009–EGU5489. Available online: https://www.fs.usda.gov/research/treesearch/42278 (accessed on 12 July 2024).
  60. ESRI. ArcGIS Desktop, version Release 10; Environmental Systems Research Institute: Redlands, CA, USA, 2011.
  61. Alcasena, F.; Ager, A.; Bailey, J.; Pineda, N.; Vega-García, C. Towards a comprehensive wildfire management strategy for Mediterranean areas: Framework development and implementation in Catalonia, Spain. J. Environ. Manag. 2019, 231, 303–320. [Google Scholar] [CrossRef]
  62. Mitsopoulos, I.; Mallinis, G.; Zibtsev, S.; Yavuz, M.; Saglam, B.; Kucuk, O.; Bogomolov, V.; Borsuk, A.; Zaimes, G. An integrated approach for mapping fire suppression difficulty in three different ecosystems of Eastern Europe. J. Spat. Sci. 2017, 62, 139–155. [Google Scholar] [CrossRef]
  63. Dillon, G.; Menakis, J.; Fay, F. Wildland fire potential: A tool for assessing wildfire risk and fuels management needs. In Proceedings of the Large Wildland fires Conference, Missoula, MT, USA, 19–23 May 2014; Proc. RMRS-P-73. Keane, R.E., Jolly, M., Parsons, R., Riley, K., Eds.; U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station: Fort Collins, CO, USA; pp. 60–76. Available online: https://www.fs.usda.gov/research/treesearch/49429 (accessed on 12 July 2024).
  64. Alexander, M.E.; Lanoville, R.A. Predicting Fire Behavior in the Black Spruce-Lichen Woodland Fuel Type of Western and Northern Canada—Poster; Forestry Canada, Northern Forestry Center: Edmonton, AB, Canada; Government of the Northwest Territories, Department of Renewable Resources, Territorial Forest Fire Center: Fort Smith, NT, Canada, 1989; 16p. [Google Scholar]
  65. Bar Massada, A.; Radeloff, V.; Stewart, S.; Hawbaker, T. Wildfire risk in the wildland-urban interface: A simulation study in northwestern Wisconsin. For. Ecol. Manag. 2009, 258, 1990–1999. [Google Scholar] [CrossRef]
  66. Zigner, K.; Carvalho, L.M.V.; Jones, C.; Benoit, J.; Duine, G.-J.; Roberts, D.; Fujioka, F.; Moritz, M.; Elmquist, N.; Hazard, R. Wildfire Risk in the Complex Terrain of the Santa Barbara Wildland–Urban Interface during Extreme Winds. Fire 2022, 5, 138. [Google Scholar] [CrossRef]
  67. Fernández-Alonso, J.M.; Vega, J.A.; Jiménez, E.; Ruiz-González, A.D.; Álvarez-González, J.G. Spatially modeling wildland fire severity in pine forests of Galicia, Spain. Eur. J. For. Res 2016, 136, 105–121. [Google Scholar] [CrossRef]
  68. Fernandes, P.; Loureiro, C. Modelos de Combustível Florestal para Portugal—Documento de Referência, Versão de. 2021. Available online: https://www.researchgate.net/publication/357812218_Modelos_de_combustivel_florestal_para_Portugal_-_Documento_de_referencia_versao_de_2021 (accessed on 24 February 2024).
  69. Molina-Martinez, J.R.; Rodriguez y Silva, F. Modelos de Combustible Forestales UCO40. In Herramientas de Nueva Generación en Defensa contra Incendios Forestales. In Proceedings of the 5th Spanish Forestry Congress, Ávila, Spain, 21–25 September 2009. [Google Scholar]
  70. Madrigal, J.; Romero-Vivó, M.; Rodríguez y Silva, F. Definición y Recomendaciones Técnicas en el Diseño de Puntos Estratégicos de Gestión. Generalitat Valenciana; Conselleria d’Agricultura, Medi Ambient, Canvi Climatic i Desenvolupament Rural: Valencia, Spain, 2019; ISBN 978-84-941695-4-0. [Google Scholar]
  71. Krsnik, G.; Busquets Olivé, E.; Piqué Nicolau, M.; Larrañaga, A.; Cardil, A.; García-Gonzalo, J.; González Olabarría, J.R. Regional Level Data Server for Fire Hazard Evaluation and Fuel Treatments Planning. Remote Sens. 2020, 12, 4124. [Google Scholar] [CrossRef]
  72. Gómez-González, J.L.; Cantizano, A.; Caro-Carretero, R.; Castro, M. Leveraging national forestry data repositories to advocate wildfire modeling towards simulation-driven risk assessment. Ecol. Ind. 2024, 158, 111306. [Google Scholar] [CrossRef]
  73. Cardil, A.; Monedero, S.; SeLegue, P.; Navarrete, M.Á.; de-Miguel, S.; Purdy, S.; Marshall, G.; Chavez, T.; Allison, K.; Quilez, R.; et al. Performance of operational fire spread models in California. Int. J. Wildl. Fire 2023, 32, 1492–1502. [Google Scholar] [CrossRef]
  74. Real Decreto-Ley 15/2022, de 1 de Agosto, por el que se Adoptan Medidas Urgentes en Materia de Incendios Forestales, Ministerio de la Presidencia, Justicia y Relaciones con Las Cortes, Gobierno de España. Available online: https://www.boe.es/eli/es/rdl/2022/08/01/15/con (accessed on 12 July 2024).
Fire 07 00249 g001
Figure 1. Study area: the SW corner of Galicia (northwestern Spain).
Figure 1. Study area: the SW corner of Galicia (northwestern Spain).
Fire 07 00249 g001
Fire 07 00249 g002
Figure 2. Map of the distribution of Rothermel fuel models in the study area (left) and a detailed view of a portion of the study area (right).
Figure 2. Map of the distribution of Rothermel fuel models in the study area (left) and a detailed view of a portion of the study area (right).
Fire 07 00249 g002
Fire 07 00249 g003
Figure 3. Workflow followed to perform the study.
Figure 3. Workflow followed to perform the study.
Fire 07 00249 g003
Fire 07 00249 g004
Figure 4. Distribution of ignition points, represented as gray dots, in the study area.
Figure 4. Distribution of ignition points, represented as gray dots, in the study area.
Fire 07 00249 g004
Fire 07 00249 g005
Figure 5. Output simulation data obtained for a sample ignition point.
Figure 5. Output simulation data obtained for a sample ignition point.
Fire 07 00249 g005
Fire 07 00249 g006
Figure 6. Graphical representation of wildfire size, FS (in hectares), for each ignition point considered.
Figure 6. Graphical representation of wildfire size, FS (in hectares), for each ignition point considered.
Fire 07 00249 g006
Fire 07 00249 g007
Figure 7. Wildfire behavior aggregated maps of the study area: (A) ROS in typical conditions; (B) FLI in typical conditions; (C) ROS in atypical conditions; (D) FLI in atypical conditions.
Figure 7. Wildfire behavior aggregated maps of the study area: (A) ROS in typical conditions; (B) FLI in typical conditions; (C) ROS in atypical conditions; (D) FLI in atypical conditions.
Fire 07 00249 g007
Table 1. Summary of the input data.
Table 1. Summary of the input data.
Type of DataSourceFormatResolution
(Spatial/Temporal)		Source Date
Digital Terrain Model (DTM)IGN [55]RasterSpatial: 5 × 5 m2009
Fuel model mapSolares-Canal et al. [56] RasterSpatial: 5 × 5 m2019
Weather recordsObservation and Prediction Unit of Galicia [52]AsciiTemporal: Daily2019
Table 2. Distribution of Rothermel fuel models in the study area in absolute and relative values.
Table 2. Distribution of Rothermel fuel models in the study area in absolute and relative values.
Fuel ModelDescriptionArea (ha)Percentage (%)
Model 0Anthropogenic areas, bodies of water1708.865.6
Model 1Rocky areas3525.3311.6
Model 2Crops and pastures9240.8430.4
Model 4High shrubs2818.499.3
Model 5Low shrubs1140.13.8
Model 7Tree-covered areas with continuous vertical structure7179.2123.6
Model 10Tree-covered areas with discontinuous vertical structure4322.214.2
Model 12Harvested areas431.591.4
TOTAL 30,366.62100.0
Table 3. Weather scenarios. WD: wind direction, WS: wind speed, T: temperature, RH: relative humidity.
Table 3. Weather scenarios. WD: wind direction, WS: wind speed, T: temperature, RH: relative humidity.
ConditionID ScenarioWD (°)WS (km/h)T (°C)RH (%)
TypicalI180 (S)12.52564.9
II45 (NE)11.728.053.1
III315 (NW)12.431.053.1
IV225 (SW)12.428.063.0
AtypicalV180 (S)58.835.724.0
VI45 (NE)46.038.921.0
VII315 (NW)54.741.514.0
VIII225 (SW)35.434.762.0
Table 4. Proportion of ignition points (in percentages) that lead to different sizes of fires for each weather scenario.
Table 4. Proportion of ignition points (in percentages) that lead to different sizes of fires for each weather scenario.
FS (ha)Typical ConditionsAtypical Conditions
IIIIIIIVAver.VVIVIIVIIIAver.
No fuel21.120.620.620.820.7721.121.122.320.321.2
<1003.72.32.52.52.70.00.31.40.60.6
100–50034.945.435.836.338.16.25.69.95.46.8
500–100030.124.833.029.929.47.36.58.27.97.5
1000–150010.17.08.210.48.93.96.810.47.97.3
1500–20000.00.00.00.00.04.84.86.55.65.4
>20000.00.00.00.00.056.654.941.452.451.3
Total100100100100100100100100100100
Table 5. Distribution of wildfire behavior variables in typical and atypical weather scenarios. Percentage of simulated wildfires corresponding to each category of ROS and FLI values for each weather scenario.
Table 5. Distribution of wildfire behavior variables in typical and atypical weather scenarios. Percentage of simulated wildfires corresponding to each category of ROS and FLI values for each weather scenario.
Variable Values I II III IV Typical V VI VII VIII Atypical
ROS<0.12.62.82.72.82.72.82.72.62.72.7
(km/h)0.1–0.643.045.539.341.342.30.00.10.00.40.1
0.6–246.447.147.345.546.50.61.20.527.17.3
2–57.54.210.210.08.034.239.735.920.132.5
>50.50.50.50.40.562.456.361.049.857.4
Total100100100100100100100100100100
FLI<50012.617.012.611.813.52.82.72.72.72.7
(kW/m)500–200052.147.150.552.450.50.00.30.11.40.5
2000–40000.60.71.61.21.03.48.23.243.814.6
4000–10,0000.40.90.20.50.552.853.654.117.344.4
10,000–30,00034.033.934.633.734.07.00.14.70.13.0
30,000–100,0000.20.20.20.20.21.020.14.334.415.0
>100,0000.20.30.20.20.232.915.031.00.419.8
Total 100100 100 100 100 100 100 100 100 100
Table 6. Behavior of each fuel model in typical weather simulations. Percentage of simulated wildfires corresponding to each category of ROS and FLI values.
Table 6. Behavior of each fuel model in typical weather simulations. Percentage of simulated wildfires corresponding to each category of ROS and FLI values.
Fuel Model
ParameterValue124571012
ROS (km/h)<0.10.10.00.00.10.00.00.2
0.1–0.61.236.00.76.339.138.925.3
0.6–273.153.246.152.544.745.654.2
2–524.310.552.840.715.915.019.5
>51.20.30.60.30.40.50.8
Total100100100100100100100
FLI (kW/m)<50010.65.60.00.20.00.10.5
500–200048.562.01.013.553.349.928.3
2000–40001.41.30.10.13.74.111.0
4000–10,0000.20.20.00.00.20.20.3
10,000–30,00038.030.597.985.842.345.259.0
30,000–100,0000.70.20.50.00.30.20.3
>100,0000.60.20.50.20.20.30.6
Total100100100100100100100
Table 7. Behavior of each fuel model in atypical weather simulations. Percentage of simulated wildfires corresponding to each category of ROS and FLI values.
Table 7. Behavior of each fuel model in atypical weather simulations. Percentage of simulated wildfires corresponding to each category of ROS and FLI values.
Fuel Model
ParameterValue124571012
ROS (km/h)<0.10.10.00.00.10.00.00.2
0.1–0.60.00.00.00.00.00.00.0
0.6–20.00.00.00.00.00.80.5
2–50.619.20.14.348.345.721.6
>599.380.899.995.551.753.577.7
Total100100100100100100100
FLI (kW/m)<5000.10.00.00.10.00.00.2
500–20000.00.00.00.00.00.00.0
2000–400011.80.10.00.10.00.40.1
4000–10,00035.161.81.111.154.951.733.2
10,000–30,00014.517.40.33.42.92.56.6
30,000–100,0003.46.22.53.85.45.46.7
>100,00035.114.596.081.636.839.953.2
Total100100100100100100100
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Rincón, T.; Alonso, L.; Picos, J.; Molina-Terrén, D.M.; Armesto, J. A Systematic Approach to Map and Evaluate the Wildfire Behavior at a Territorial Scale in the Northwestern Iberian Peninsula. Fire 2024, 7, 249. https://doi.org/10.3390/fire7070249

AMA Style

Rincón T, Alonso L, Picos J, Molina-Terrén DM, Armesto J. A Systematic Approach to Map and Evaluate the Wildfire Behavior at a Territorial Scale in the Northwestern Iberian Peninsula. Fire. 2024; 7(7):249. https://doi.org/10.3390/fire7070249

Chicago/Turabian Style

Rincón, Thais, Laura Alonso, Juan Picos, Domingo M. Molina-Terrén, and Julia Armesto. 2024. "A Systematic Approach to Map and Evaluate the Wildfire Behavior at a Territorial Scale in the Northwestern Iberian Peninsula" Fire 7, no. 7: 249. https://doi.org/10.3390/fire7070249

Article Metrics

Back to TopTop