Climate Risk Management 44 (2024) 100595

Contents lists available at ScienceDirect

Climate Risk Management

journal homepage: www.elsevier.com/locate/crm

Enhancing infrastructure resilience in wildfire management to

face extreme events: Insights from the Iberian Peninsula
Erica Arango a, Pilar Jiménez b, Maria Nogal c, *, Hélder S. Sousa a, Mark G. Stewart d,
José C. Matos a
a
Dept. of Civil Engineering, ISISE, ARISE, University of Minho, Guimarães 4800-058, Portugal
b
Dept. of Mining and Civil Engineering, Technical University of Cartagena, Cartagena 30.203, Spain
c
Dept. of Materials, Mechanics, Management, and Design, Technical University of Delft, Delft 2628CN, the Netherlands
d
School of Civil and Environmental Engineering, University of Technology Sydney, NSW 2007, Australia

A R T I C L E I N F O A B S T R A C T

Keywords: Factors such as human activity and climate change are contributing to an increase in the fre­
Wildfire management quency and intensity of wildfires. This problem has challenged society’s knowledge, response
Resilience capacity, and resilience, revealing its inadequacy to cope with the new wildfire regime charac­
Iberian Peninsula
terized by extreme wildfire events (EWE). Policies on wildfire management mainly focus on
Cross-border policies
GIS-based tool
suppression and managing emergencies, which may be insufficient to reduce EWE’s incidence and
cope with its impact. Consequently, there is a lack of tools to support decision-making in wildfire
management in other important aspects, such as prevention and protection. This study examines
global wildfire policies specifically in the Iberian Peninsula (Portugal and Spain), including cross-
border policies. A GIS-based tool to evaluate different normal and extreme wildfire management
policies is applied to a cross-border case study, paying attention to the impact on critical land-
based transport systems. A relevant outcome of the tool application is that suppression must be
complemented with other wildfire management strategies in the analyzed area. The gained in­
sights can help stakeholders to improve decision-making in wildfire management to successfully
address EWE.

1. Introduction

Climate change is causing significant changes in the frequency, intensity, and duration of wildfire events around the world.
Consequently, we are experiencing a new regimen of wildfires, characterized by uncontrollable behavior that burns through large
areas, resulting in severe and unexpected impacts, i.e., destroying homes, wildlife, and natural resources as never before. For instance,
the Australian bushfires of 2020 destroyed over 10 million hectares of forest and caused dozens of fatalities (BBC, 2020). Over the last
20 years, wildfires in the European Union have cost more than €55 billion (Gonzalez, 2021; EC, 2018), significantly affecting the
Southern Member States, such as Spain and Portugal. In 2017, wildfires severely affected the municipality of Pedrógão Grande in
Portugal, causing the deaths of 65 people. It resulted in over €523 million in direct costs, with an estimated total social cost of €613
million (CTI, 2017). The wildfire of 2023 in Maui, Hawaii, stands as one of the deadliest catastrophes in the United States, claiming
approximately 100 lives. Furthermore, it resulted in the displacement of nearly 6.000 people, with an estimated $5.5 billion required

* Corresponding author.
E-mail address: [email protected] (M. Nogal).

https://doi.org/10.1016/j.crm.2024.100595
Received 15 May 2023; Received in revised form 26 February 2024; Accepted 27 February 2024
Available online 11 March 2024
2212-0963/© 2024 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY license
(http://creativecommons.org/licenses/by/4.0/).
E. Arango et al. Climate Risk Management 44 (2024) 100595

Fig. 1. The holistic view of wildfire management based on resilience.

for the repair of damaged areas in West Maui (Hassan and Betts, 2023). In this context, the existing wildfire policies are facing
increasing pressure, and their adequacy is being questioned in light of the new conditions induced by climate change and inadequate
land and forest management. The latter is critical in the Iberian Peninsula due to the land abandonment (Duane et al., 2021).
For the sake of clarity, in this paper, Extreme Wildfire Events (EWE) are defined following Tedim et al. (2018), as fires with
exceptionally high values in critical aspects, such as intensity and rate of spread, leading to unpredictable behavior. These pose sig­
nificant challenges to suppression efforts and result in significant societal costs.
Effective wildfire management is a complex multifaceted process that integrates proactive and reactive approaches. Reactiveness
involves responding to fires with a focus on extinguishing them rather than addressing their underlying causes. In contrast, proac­
tiveness encompasses actions aimed at preventing fire occurrences and impacts (Mourao and Martinho, 2019). These approaches
encompass various strategies, such as prevention, protection, detection, suppression, and recovery strategies. Prevention aims to
reduce the risk of wildfires by addressing the root causes, such as reducing fuel loads, managing forests, and implementing fire-safe
building codes. Protection involves mitigating the potential damage from wildfires by creating defensible spaces around homes and
communities and developing evacuation plans for residents. Detection is also an important aspect of wildfire management, as early
warning systems can provide crucial information about the location, size, and potential impact of a fire. Rapid response is critical and
suppression policies and resources such as firefighters, helicopters, and equipment are critical to this effort. Once a wildfire has been
contained, recovery policies are necessary to help communities and ecosystems rebuild and heal. This includes supporting displaced
residents, restoring damaged infrastructure, and rehabilitating affected ecosystems. Effective wildfire management requires a
comprehensive and integrated approach that encompasses the five strategies from a resilience perspective. Fig. 1 presents a global view
of wildfire management based on resilience theory. This framework aligns with the Integrated Fire Management framework proposed
in the European Commission report (EC, 2018). One notable difference lies in the incorporation of the learning process within the
holistic vision presented here. Note that mitigation is implicit in prevention and protection strategies. While adaptation is primarily
associated with the learning process, it can also be implemented at any of the other stages.
Although wildfire management should include the five strategies, the main focus of wildfire-related policies is suppression and
emergency management (Fernandes et al., 2020; Leone et al., 2020; Rossi et al., 2022). This approach has been adequate for normal
wildfires, i.e., fires that do not fall under the classification of EWE which are more controllable. However, as fires become EWE, i.e.,
more intense, and faster spreading, suppression capacity is overwhelmed (Fernandes et al., 2016; Moreira et al., 2020).
Wildfire management and policies are generally approached from the perspective of risk management and suppression (Noble and
Paveglio, 2020). Countless methods of all kinds have been developed to study the wildfire problem and support decision and policy
making. For instance, qualitative methods attempt to incorporate the different factors related to the complexity of the fire phenomenon
(Asori et al., 2020; Schultz et al., 2021). The quantitative methods include but are not limited to, statistics-based models that correlate
various types of data (e.g., Costafreda-Aumedes et al., 2017; Alcasena et al., 2021). Data analysis and simulation tools to extract in­
formation from large databases are used by Wang et al. (2017); McNamee et al. (2022), logistic models or random forest algorithms are
applied by Galizia and Rodrigues (2019); Salis et al. (2021); whereas Naderpour et al. (2021) implement automatic learning methods.
These methods are based on historical records and forecasts, usually data intensive. Since EWE are characterized by unexpected
behavior, and increasingly exceed expectations in terms of intensity, frequency, and magnitude, using past data to predict EWE may
not be suitable (Pinto et al., 2018). The uncertainty associated with the new wildfire regime makes predicting its behavior extremely
difficult. Therefore, risk estimation and the consequent decision-making are subjected to larger uncertainty. On the other hand, many
of these methods are scenario-based, often becoming highly categorical for specific scenarios. Generalizing results to other scenarios
and conditions requires careful consideration. Consequently, these methods are unable to properly capture the effects of the new
generation of EWE (Arango et al., 2023a).
In conclusion, risk assessment relies on accurate and updated hazard information. Thus, wildfire management is based on known
and measurable data. This approach is successful in suppressing wildfires and protecting society in highly predictable scenarios
(Castellnou et al., 2019) but not facing EWE. Therefore, experts suggest shifting the paradigm from suppression to prevention and
adopting a resilient approach (IPCC, 2022; Ganteaume et al., 2021; Moreira et al., 2020). It results in several studies aiming to increase
the resilience of wildfire emergency response, (e.g., Talebi et al., 2022; Sakellariou et al., 2023; Castellnou et al., 2019); and others to
promote the transition towards more resilient landscapes, as Duane et al. (2022). However, there are some limitations to be covered
such as the lack of available tools to support the implementation of proactive strategies around prevention and protection. These tools
are key to improving policies. Existing tools generally focus on supporting only one of the phases of wildfire management, either

2
E. Arango et al. Climate Risk Management 44 (2024) 100595

reactive (e.g., Andrade and Hulse, 2023) or proactive (e.g., Song et al., 2023; Subramaniam et al., 2023). However, from a resilience-
based perspective, policies should aim at finding a sustainable balance between reactivity and proactivity in the long term.
To overcome the lack of available tools, Arango et al. (2023a) propose a GIS-based fire analysis tool, GIS-FA, to quantify the
exposure level to wildfires of infrastructure or buildings, based on natural and man-made environmental conditions. This tool also
overcomes the described limitations of the risk-based models. Furthermore, it allows assessing different strategies in their capacity to
reduce wildfire impact, including normal and extreme wildfires (Arango et al., 2024).
The novelty of this paper lies in showcasing how to enhance decision-making in resilient wildfire management in transboundary
contexts. Despite the importance of cross-border collaboration for defending communities against the increased risk of wildfires, this
subject is little analyzed. This has been achieved by leveraging the capabilities of the GIS-FA tool. The paper highlights the unique
ability of this tool to effectively address fire management in situations involving multiple borders, thus marking a significant advance
in integrated fire management in border areas. It not only identifies and overcomes limitations in individual country-level fire policies
but also promotes bilateral cooperation, optimizing management and emphasizing joint actions to address common challenges in
resilience and safety in cross-border areas. This facilitates a balanced approach between reactive and proactive measures, supporting
long-term decision-making and emergency planning. This paper presents a discussion on the policies and strategies addressed in
wildfire management, putting under scrutiny the prevention and protection capacity of communities and approaches to manage the
problem of the new fire regime. Special attention is paid to wildfire management policies in the Iberian Peninsula (Portugal and Spain),
analyzing a cross-border case study. It shows an example of cross-border cooperation of countries with vast experience in wildfire
management but yet little experience in EWE. In this context, the land-based transportation system is analyzed given its critical role in
the different phases of wildfire management. The GIS-FA tool is used for this purpose providing interesting insights on the challenges of
cross-border cooperation and potential consequences of centering the efforts on suppression under EWE. The discussion presented in
this work can help to find the keys to future wildfire management and research.
The rest of the document is organized as follows. In Section 2, policies and practices related to wildfire management are analyzed at
the international level with a particular focus on the Iberian Peninsula. Then, the GIS-FA tool is explained in Section 3. In Section 4 the
cross-border case study is presented, whereas the discussion around the limitations of the cross-border cooperation and insights from
the case study are presented in Section 5. Finally, some conclusions are drawn in Section 6.

2. Review of policies on wildfire management

2.1. A global approach based on suppression

This section reviews the existing policies on wildfire management, paying attention to those countries with more experience in fire
fighting. The United States had a policy of total fire suppression until the year 2000. Then, they approved the National Fire Plan
(NWCG, 2006), with a clear shift towards ecosystem adaptation, recovery, and economic assistance to rural communities (McCaffrey
et al., 2015). After 2009, it was strengthened through the Federal Land Management and Land Assistance Improvement Act (WFLC,
2014). The foundation of the USA wildfire policy is risk management, which mainly incorporates fuel treatment such as firebreaks or
prescribed burning. However, property and people protection, early warnings, and better management of natural resources to coexist
with wildfires are poorly considered. Therefore, at present, the USA fire management mostly focuses on suppression.
Current Canadian practices are more resilience-oriented. They include zoning based on the five strategies of wildfire management.
However, most resources are dedicated to the suppression phase, despite studies pointing to the need for increased protection capacity
(EMPOD, 2017). In terms of prevention, they use measures such as forest area closures, burning permits, and the prohibition of fires (e.
g., bonfires, fireworks, and explosive targets). Forecasts of future EWE suggest that current wildfire management policies in Canada are
insufficient to address the growing wildfire threat (Tymstra et al., 2020).
In the case of Australia, due to its long history of wildfires, they have implemented more advanced management policies in terms of
prevention and protection. They include the preliminary stages of training and education of communities to reduce the likelihood of
fire-related losses and improved building standards to reduce vulnerability (Reynolds, 2017). People took up the focus on community
safety and protection as an incentive to stay during fires to defend their properties (Whittaker et al., 2013). However, after the Black
Saturday wildfires (2009), there was a shift towards early self-evacuation as the safest option for the community (McLennan et al.,
2019). Despite these social and research development efforts to reduce wildfire risk, the current management program, the Enhanced
Bushfire Management Program (EBMP), is more focused on supporting improved wildfire suppression and risk reduction (OEH, 2016).
European policy focuses on fire suppression too, with evacuation being the last option. However, in recent years, preventive
guidelines, projects, and funding focused on fire-resistant landscapes through fuel management or preventing arson have been mainly
encouraged in some of the countries in the southern part of Europe (Salis et al., 2016; EC, 2018). However, protection initiatives
remain ineffective and financially under-resourced (Cardoso et al., 2018). The European Union (EU) has created a Civil Protection
Mechanism to strengthen cooperation among the EU countries and participating states regarding firefighting, thus, improving response
to disasters. Member States are required to make as many resources available as the Commission may request, however, they are not
obliged to offer their assistance.
This reality shows that society is aware of the importance of wildfire prevention and protection, however, actual changes in wildfire
management policies towards prevention and protection although significant, are insufficient and efforts continue to focus on wildfire
suppression.

3
E. Arango et al. Climate Risk Management 44 (2024) 100595

Fig. 2. Strategic coordination of the Integrated Management System for rural fires - Portugal.

2.2. Cross-border policies: the Iberian Peninsula

It is frequent in Europe to find fires initiated in one country that cross the border and cause serious damage in the neighboring
country. Especially in the border area between Portugal and Spain, which is one of the longest land borders in Europe with about 1.200
km. Both countries have similar landscapes and orography, and most of the border territory is predominantly rural, especially in
Portugal. The urban areas are poorly developed with small settlement concentrations. Moreover, given the geographical situation, they
are prone to wildfires, thus, their collaboration is essential for the defense of their communities against the increased risk of large forest
fires. This section describes the legislative framework related to the wildfire policies of each country and analyses their cross-border
wildfire management policies and cooperation instruments. This section also serves to contextualize the case study.

2.2.1. Wildfire legislation and policies

Related to wildfire legislation and policies, Portugal and Spain present a similar framework. The fire management strategy is based
on national legislation that governs the forest area and its management. The Portuguese policy is based on Law 76/2017 (Diario da
República, 2017), Sistema de Defesa da Floresta Contra Incêndios, an updated of Decree-Law 124/2006. The forestry policy in Spain is
based on Law 43/2003 (BOE-A-2033-21339), Ley de Montes. Each of the District Centres in Portugal and each regional Spanish
community have legislative autonomy to establish laws and guidelines for their territory (always complying with the guidelines set at
the national level). Designated government entities of each country are responsible for wildfire management throughout the national
territories and their districts to support prevention and suppression tasks.
In recent decades, Portugal has been the subject of several analyses by wildfire specialists, especially after the wildfires in 2017 and
2018 (CTI, 2017; CTI, 2018). The recommendations were relatively consistent and pointed to four main domains that require
enhancement: (1) preventing negligent human ignitions, (2) creating a structural fire defense system consisting of fuel management
strips and reducing fuel loads in critical areas, (3) improving firefighting capabilities by implementing perimeter control tactics and
comprehensive firefighting strategies, and (4) restructuring firefighting organization (Beighley and Hyde, 2018). Therefore, based on
the lessons learned from the 2017 and 2018 wildfires, the Portuguese National Forest Fire Defence Plan was modified. The new plan,
developed by the Agency for Integrated Rural Fire Management (AGIF) and approved by the Portuguese Congress (resolution No. 45-
A/2020), contains a fuel management program for 2020–2030, that aims to reduce exposure to wildfires. Unlike its predecessor plan
(the one that ruled from 2006 to 2018 whose main objective was suppression), this plan recognizes that rural fires cannot be prevented
entirely, so it is necessary to prepare the society. Therefore, the plan is based on two central axes: managing rural wildfires and
protecting against rural wildfires. Prevention measures mainly consist of fuel treatment mosaics or linear firebreaks designed with
local stakeholders (i.e., free of waste, spontaneous scrub, and dry vegetation). These prevention measures are an important step but are
still insufficient. For example, the firebreak strips were officially approved by Portuguese Decree-Law 17/2009 and the forensic report
of the 2017 fire (CTI, 2017) found no evidence of fuel management for the municipality of Pedrógão Grande.
The Spanish forestry law establishes common guidelines on training in prevention and suppression. Related to fire prevention, the
fragmentation of forest areas is a solution to regulate the exercise of all the activities that can increase fire risk. Another safety standard
is to provide a minimum of 25-meter-wide firebreak strips for homes and buildings in forest areas. Similarly, the ditches and easement
areas of roads and railways that cross forest areas shall be kept clean for a minimum width of 2 meters. It is recommended that
preventive work has to be carried out throughout the year. Related to wildfire suppression, firefighting services prioritize using public
infrastructures such as roads, telephone lines, airports, reservoirs, seaports, and all those necessary for the communication and supply
of these services.

4
E. Arango et al. Climate Risk Management 44 (2024) 100595

Fig. 3. Entities involved in the coordination of wildfire management emergency - Spain.

Fig. 4. Mutual Aid Plan for wildfire emergencies in border areas between Portugal – Spain.

With the advance of large wildfires, despite the increase in the investment of equipment and resources for suppression, it has not
been enough to control this type of event. Therefore, initiatives have arisen to increase investment in wildfire prevention programs, i.
e., the promotion of long-term preventive actions. The target is to move from a management model mainly based on specific actions
and maintenance of infrastructures (such as firebreaks and water points) to a long-term value forestry and agricultural model in rural
areas to create an agroforestry territory resilient to forest fires (WWF, 2021).

5
E. Arango et al. Climate Risk Management 44 (2024) 100595

Therefore, both countries are guiding their efforts towards prevention with fuel-treatment measures such as agricultural zoning
resilient to forest fires and creating networks of linear firebreaks (i.e., linear fuel-cutting) designed with local stakeholders. However,
these plans are recent, and the implementation of these measures is very slow, with few resources that are not able to achieve an
effective response.

2.2.2. Agents involved in the coordination of wildfire management

Regarding the coordination, the Agency for the Integrated Management of Rural Fires, I.P. (AGIF, I.P.), created by Decree-Law No.
12/2018, is responsible for the strategic coordination of the integrated wildfire management system (SGIFR) in Portugal. The success
of this system depends on the harmonious integration of the responsibilities assigned to two main public institutions, the Institute for
Conservation and Forests (ICNF, I.P.), and the National Authority for Emergency and Civil Protection (ANEPC). Both are in charge of
the prevention and protection of fire risk, in rural areas and the built and industrial surroundings, respectively. These two entities
count on the support of the local government, Republican National Guard (GNR), and the Armed Forces, as well as other entities from
civil society, to manage wildfires from planning to recovery, that is, through planning, prevention, protection, suppression, and post-
event recovery. Fig. 2 summarizes the hierarchical coordination for compliance with the Portuguese SGIFR.
With three levels of hierarchy, Spain shows similar coordination (see Fig. 3). However, each Autonomous Community manages
wildfires under different directorates, examples of wildfire management in each community can be found in González and Vallejo
(2023); Typsa (2023). When the fire affects different areas, the three levels (national, autonomous, and local) collaborate through the
national coordination committee, which supports the entities involved in wildfire management, such as firefighters, medical services,
national police, and the Civil Guard.
Because of the similarities between the management of wildfire policies and conditions, the first steps to cross-border cooperation
Portugal-Spain were established in 1980. Since then, there has been a technical cooperation and mutual assistance agreement between
the Portuguese and Spanish firefighting and rescue services with an additional protocol on forest fires in border areas. It was
formalized through the so-called Évora protocol, signed by the two countries on March 9th, 1992 (Martín et al., 2018) and its last
update was in 2019 (BOE-A-2019-12928). It aims to allow rapid intervention and mutual assistance in the event of wildfires in the
border area, whenever the urgency of the situation requires. Within border areas, special attention will be given to fires that are less
than 25 km from the border and whose propagation conditions (i.e., wind, topography, fuel patterns) suggest that there is a very high
probability that the fire crosses the border in a short period. Fig. 4 summarizes the mutual aid plan for cross-border in Portugal - Spain.
Cross-border cooperation involves policies, legal instruments, such as bilateral agreements and treaties, as well as financing
programs. Among the most important funding programs is the “Interreg” program, which has been implemented between Portugal and
Spain since 1989. The current version is Interreg VI, which covers the period from 2021 to 2027 (Interreg Spain-Portugal, 2022). This
program finances projects focused on (i) training and equipping the staff for cross-border suppression operations, (ii) educating the
rural population about the risk of large forest fires, good preventive practices, and self-protection, and (iii) improving the coordination
of the emergency action units with common elements of mobile emergency communication, among others –FIREPOCTEP (FIRE­
POCTEP 2020), Interlumes (Interreg Spain-Portugal, 2019), CILIFO (Interreg, 2019), ARIEM-112 (ARIEM, 2015)–.
The Cross-Forest project (CEF, 2018) aims to develop a Digital Service Infrastructure (DSI) focused on predicting forest fire
behavior and spread. It includes forest maps and forest fire spread models at the country level based on GIS datasets. This project is co-
financed by the European Union’s Innovation and Networks Executive Agency (INEA), through the Connecting Europe Facility (CEF).
Also, the VOST platform (Sousa et al., 2018), formed by a large group of volunteers, whose objectives ranges from monitoring, col­
lecting, and disseminating official information during emergencies of any kind on social networks, to combating the phenomenon of
online disinformation that can cause a social alarm during emergencies.
It is noted that the main focus of the long-standing collaboration has been on suppression and emergency actions. On the other
hand, despite the wide variety of entities and projects aiming at cross-border cooperation, further work must be done on integration,
coordination, and sharing of data and experiences to make the actions more effective. For instance, to avoid overlaps and duplication
of responsibilities in wildfire management in all phases. In this sense, tools to inform stakeholders and support the decision-making
process can be used by both countries to identify the need for interventions and the division of responsibilities, thus improving the
coordination of wildfire management. In this sense, wildfire management can be oriented towards a more holistic view based on
resilience (see Fig. 1).

3. GIS-FA tool to inform policies on wildfire management

Arango et al. (2023a) propose the GIS-FA tool that allows the assessment of the priority level for the interventions of infrastructure
systems in wildfire management, supporting the decision-making and prioritization of resources. The tool includes the impact of
normal fires and EWE at a system level and considers inter-dependencies and redundancies between system components. It is
developed from a resilience-based perspective, focusing on the system’s capabilities rather than wildfire occurrence probability. The
tool considers the system’s exposure to different types of wildfires as a consequence of the natural environment and human activities,
including those aiming at preventing and protecting the system. However, it does not consider the wildfire occurrence probability,
thus, it avoids the unpredictability associated with the EWE problem.
The priority level for intervention is calculated using two criteria, the exposure of the infrastructure to wildfires and the criticality
of an asset within the system. In this work, exposure assessment is highlighted as an individual tool to support decision-making. It uses
an exposure metric denominated Fire Arrival Time (FIRAT), which considers the environment in its ability to spread fire under
different fire conditions. As a result, an exposure map is obtained providing the average time for a random fire, either normal or EWE,

6
E. Arango et al. Climate Risk Management 44 (2024) 100595

Table 1
Characteristic ROS values of grassland for the different wildfire categories according to Tedim et al. (2018).
Fire category Normal fires Extreme Wildfires

1 2 3 4 5 6 7

ROS (m/min) 5–15 15–30 20–50 50–100 150–250 250–300 >300

Fig. 5. Cross border case study: Elvas Portugal (PT)-Badajoz Spain (SP) transportation network. The enumeration of railway segments is listed in
yellow, and road segments are listed in white.

to reach a given asset of the system (e.g., roads, stations). FIRAT calculation considers the Rate of Spread (ROS) associated with
different burning sources and fire barriers for a different wildfire intensity, as well as the Equivalent Fire Distance, EFD. The EFD allows
the aggregation of different sources and barriers around the asset. Seven wildfire categories are included in the analysis based on the
corresponding ROS values, as summarized in Table 1.
Therefore, to determine the exposure level of a given infrastructure system using this tool, the following steps have to be followed:
(1) Definition of the studied system through its components or assets (e.g., buildings or roads) and fire propagation sources and barriers
that can influence the exposure of the studied target. (2) Establishing the ROS values for the different burning sources and fire barriers
considering their characteristics and locations. (3) Selection of one reference object from the fire propagation sources previously
identified (e.g., grassland). (4) Normalization of the ROS values of the sources and barriers using the reference object. (5) Measurement
of the distances between the fire propagation sources and barriers and the components or assets. The most critical distance for each
source and barrier concerning each component or asset is considered. (6) Equivalent Fire Distance (EFD) is calculated as the mean
value of the critical distances for each component or asset, multiplied by the normalized ROS rations. (7) FIRAT estimation, which
considers the EFD and the characteristic ROS value of the reference object associated with each wildfire category (see Tedim et al.,
2018).
Exposure assessment introduces aspects related to the environmental conditions of the system, thus reflecting the impact of eco­
nomic, social, and political activities. In this way, the social dimensions of wildfires are considered. For a more detailed description of
the parameters, procedure, and results the reader is referred to Arango et al. (2023a).

4. Cross-border case study

GIS-FA tool has been presented and validated in Arango et al. (2023a) and Arango et al. (2023b), respectively. This section presents
a cross-border case study using the GIS-FA tool and how it can support the decision-making process at all stages of wildfire
management.

7
E. Arango et al. Climate Risk Management 44 (2024) 100595

Fig. 6. Fire propagation sources and barriers considered for Elvas - Badajoz.

8
E. Arango et al. Climate Risk Management 44 (2024) 100595

Fig. 7. Analysis of the land cover in the cross border case study: Elvas Portugal - Badajoz Spain.

Fig. 8. Differences between the forest classification for Portugal and Spain.

4.1. Case study description

The study area is located between the Portuguese city of Elvas, in the district of Portalegre in the Alentejo region, and the Spanish
city of Badajoz, the capital of the namesake province and the largest economic and commercial center of the autonomous community
of Extremadura (see Fig. 5). This area was legally constituted as Euro-City Elvas-Badajoz in 2013 after the abolition of border controls
by the countries and the Schengen Agreement. Since the agreement, it has become the area with the largest population within the
Portuguese-Spanish border area (Castanho et al., 2017). This case study was selected due to its high fire risk probability, ranging from
60 to 70%, as indicated by information from the European Forest Fire Information System (EFFIS, 2022). Additionally, this area is
crucial for land connectivity between the two countries. The area is connected by road and railway networks, as depicted in Fig. 5. The
principal roads crossing the region are the A5-A6 motorway connecting the capitals of both countries, Madrid-Lisbon, which enters the
region from the northeast and exits through Badajoz, in the west, in the direction of Lisbon. The railway network, which transports

9
E. Arango et al. Climate Risk Management 44 (2024) 100595

Fig. 9. Fire Approach Time – FIRAT (in minutes) for the Elvas Portugal (PT)-Badajoz Spain (SP) transport network. (a) Normal Fire Category 2, and
(b) extreme Fire Category 6.

users and goods, has connections to Madrid, Lisbon, and other large cities such as Seville, Huelva, and Ciudad Real via the Madrid-
Cáceres-Mérida-Badajoz-Lisbon corridors. These rail and road routes are the most direct transportation lines connecting the capitals of
both countries.

4.2. Results from exposure analysis in the cross-border case study

The transportation system under study encompasses 123 km of road network and 28 km of railway network. To conduct the
analysis, the roads and railways have been segmented in order to provide a more detailed exposure assessment. Both networks are split
at the border level, to better support decision-making because actions rely on one of the two countries, see Fig. 5. The road network is
segmented into 74 road sections according to intersections with other roads and railways. The rail network is also divided into 18
segments regarding intersections and then at equal distances, in this case, 1.7 km long.
The sources of fire spread and barriers considered are shown in Fig. 6. The area consists of wide and softly undulating plains cut by
river branches. One of the most important rivers of the Iberian Peninsula, the Guadiana, flows through the zone, which is used for
irrigation and electricity generation. The Portuguese area has two special-nature protection areas, San Vicente and Torre da Bolsa, as
shown in Fig. 5.
Crops constitute 78% of the land cover in the case study, olives and plums being the most important crops. Grasslands constitute 8%
of the land cover, and forests 4%. For the entire analysis of the land cover composition, refer to Fig. 7. Note that the predominant land
cover in the case study is comprised of crops, with forested areas playing a lesser role. While forests are recognized as significant
contributors to fire spread, crops can also play a substantial role. An example of this case is the Pedrógão Grande fire in Portugal in
2017, where crops, poorly managed due to invasive species, emerged as a main source of fire spread (CTI, 2017). The information on
sources and barriers is obtained for QGIS, through Open Street Maps, and complemented with the land cover/use maps of each country
(DGT, 2022; CNIG, 2020). The modeling process and visualization of the results are automated using the Geographic Information
System (GIS).
When applying the methodology to both countries, differences in the quality of the information are apparent because the databases
are not homogenized. In this case, the biggest difference is in the forest categorization, displayed in Fig. 8. The classification of forests
in Portugal is given in up to 5 levels of detail (left area in green color). Compared to the Spanish forest classification, which has only one
level of detail (the right area in beige color). These differences will influence the accuracy of the outputs, i.e., the exposure assessment.
For instance, if a more detailed classification would reveal the presence of high-exposure species in Spain, such as eucalyptus, the
exposure would be higher for those areas.
The ROS values for all sources and barriers are considered as in Arango et al. (2023a) with updates to this specific case regarding
crop detail. The ROS values considered are shown in Tables A-1 and A-2 in Appendix A.

10
E. Arango et al. Climate Risk Management 44 (2024) 100595

Fig. 10. Percentage of km affected for the different exposure levels according to each wildfire category of the Elvas-Badajoz transport network. a)
Railway network, b) Road network.

The aggregation of the different sources and barriers is done considering the grassland ROS value as the reference source for
normalization. The grassland ROS value is used for convenience following Tedim et al., (2018), which provides the classification of
ROS characteristic values for different wildfire categories, as specified in Table 1. However, with the proper information, any source
can be used as a reference. Then, the average value of the normalized sources and barriers is considered for evaluating the EFD. The
calculated EFD and the ROS associated with the different wildfire categories are combined to obtain the exposure level of the
transportation network. Fig. 9(a) shows the exposure map of the road and railway networks for a normal wildfire category (Category 2,
ROS of 30 m/min). In contrast, Fig. 9(b) shows the exposure level of the networks for an extreme wildfire (Category 6) with a
propagation speed of 300 m/min. Each FIRAT interval is associated with a specific exposure level; i.e., FIRAT less than 30 min cor­
responds to a very high exposure level. For wildfire Category 2, the level of exposure of both transport networks is mainly low for both
countries. Except for Road 24, with 177 min for FIRAT or reaction time, which has a moderate exposure level. It is due to being
surrounded by diverse sources of high exposure, including commercial and industrial sources (see Fig. 6b). This implies that in case of
low wildfire, that would be the first road that can be reached by the fire, and consequently, it would be the main road that could be
proactively intervened upon to reduce its exposure. In contrast, for wildfire Category 6, it can be observed that the transport network’s
exposure is higher. Especially in Elvas Portugal, for which 22% of the roads have reaction times of less than one hour, as shown in Fig. 9
(b). This short reaction time can lead to road closures severely hampering evacuation processes and consequently to high societal
impacts. Meanwhile, the level of exposure of the transportation networks in Badajoz Spain is lower. This can be attributed to more
barriers in the Spanish area, such as the river. As a result, most roads have moderate to intermediate levels of exposure.
Fig. 10 shows the difference in the level of exposure of the transportation networks analyzed for all wildfire categories. It is evident
from the data that the railway network exhibits low exposure to normal fires (Categories 1 to 4). A high exposure level occurs in
Category 7, encompassing 40% of the railway network segments. In contrast, the road network, under the same category, has 12% of
its roads classified at a very high exposure level and 28% at a high level. Further details regarding exposure levels can be found in
Fig. 10. Additionally, Table B-1 in Appendix B provides detailed exposure levels categorized for each country.

4.3. Discussion of results on decision-making support using the GIS-FA tool

This tool supports decision-making for medium- and long-term planning (i.e., proactively) and during emergence (i.e., reactively).
For medium- and long-term planning, decision-makers should define the degree of exposure they are able to accept under specific
wildfire conditions. For instance, for normal-category fires, their goal is to guarantee that none of the assets of the analyzed system has
an exposure level larger than 30 min. This time may be related to the time for emergency services to reach the area. However, setting
this condition for the entire system under extreme events may be unrealistic because the fire ratio of spread can be 60 times larger than
normal wildfires. Therefore, for this type of wildfire intensity, the focus should be on guaranteeing some critical assets of the system,
for instance, the roads of an evacuation route to safeguard civilians.
The GIS-FA tool allows the assessment of the intervention’s impact on the existing exposure of the system. These interventions can
aim at improving vegetation management, creating firebreaks, installing temporal fire stations during the most critical months, or any
other type of infrastructure adaptation. Then, the tool can translate these interventions into geographically located sources and
barriers thus, modifying the exposure of the system. To illustrate how this tool can be used to support medium- and long-term decision-
making, some examples are presented below. Based on the exposure map for wildfire Category 2 (Fig. 9a), one of the actions could be
investment in reducing the level of exposure of Road 24, as it is the most exposed road. Examples of actions to reduce the exposure of
this road can be guaranteeing adequate maintenance of poorly irrigated crops and unattended vegetation in the surroundings. For
more details on vegetation management actions see Menon and Vishnu-Menon (2022); Herbert et al. (2022); Baudena et al., (2023).
From the exposure map for wildfire Category 6 (Fig. 9b), Roads 1, 21, 37, 44, and 45 could be intervened as they are part of the most
important road within the studied area, which connects the capitals of both countries. Also, intervention measures could be

11
E. Arango et al. Climate Risk Management 44 (2024) 100595

implemented to reduce the exposure of the most affected railway sections, guaranteeing the safety of the users. The above implies a
greater effort on the part of Portugal compared to Spain, as the exposure level for both railway segments and roads is higher in
Portugal. Moreover, additional interventions can be considered in the Portuguese zone, for the area near Roads 17 and 9, since their
exposure level implies that the special protection zone Torre da Bolsa can also be compromised. It should be clarified that these results
are susceptible to a more detailed mapping of information, especially in the case of Spain.
On the other hand, the tool can support decision-making for emergency planning. In this case, since the study focuses on transport
networks, it can support traffic management in terms of evacuation routes and ensure users’ safety. The exposure maps of the different
wildfire categories serve as a basis for this purpose, in which the main objective is ensuring users’ safety by avoiding potentially
disconnected areas in a wildfire event. Thus, the primary assumption is that depending on the wildfire category, the roads or railway
segments with a high level of exposure will be likely closed. For instance, as one of the first routes to be affected is Road 24, evacuation
routes must be defined in such a way that this road is avoided. However, the more intense the wildfire, the more difficult it is to define
safe routes to reach the origin–destination pairs. As shown in Fig. 9(b) for EWE, network connectivity is compromised in different
areas. At the road network level, to go from one country to another, the main route could not be used after 120 min, as fire may reach
Roads 1, 21, 37, 44, and 45. Alternative routes such as the Northern route consisting of Roads 56, 55, 38, 35, 27, 21, and 1 present the
same problem. It can be rightly argued that in the event of a southbound wildfire, the Northern route can be more dangerous. During an
emergency, decision-makers should combine the real-time information with the exposure map corresponding to the given wildfire or
larger intensity if worse conditions are expected. Even under very critical conditions and with a very intense wildfire, the entire
transport network may be affected simultaneously. For this reason, the constant improvement of all means of fire suppression and
evacuation cannot be neglected.
Cases such as Roads 32 and 34 should also be carefully considered for timely closure. To better explain this, look at Road 34 which
is intercepted at both ends by roads with higher exposure levels. Thus, if Road 34 is not closed in time, users may be trapped in the
flames and smoke (CTI, 2017 mentioned this effect in the case of Pedrógão Grande). One way to prevent users from being isolated is to
improve network connectivity. For example, Road 51 has only one end (node) connected. Therefore, extending the connectivity of
Road 51, for instance to Road 38, could reduce the possibility of that area becoming isolated.
In any case, increasing the redundancy of the network facilitates evacuation routes by adding alternative routes for the same
origin–destination pair. For this reason, the railway network is more problematic than the road network in terms of emergency
planning. Fig. 9(b) shows that any affected segments from 1 to 8 would directly affect the overall performance of the network under
consideration. For extreme wildfire categories, the rail network would lose its connectivity in less than two hours, as Segments 5, 6,
and 7 have higher exposure level i.e., less than 120 min of FIRAT. Consequently, evacuation and closure of the rail network must be a
priority in case of wildfires.

5. Insights from the cross-border case study

Cross-border collaboration brings several challenges for the countries involved. These challenges are associated with different
areas, stakeholders, and resources, which include but are not limited to:

• Governance. It represents one of the most challenging aspects of cross-border cooperation. It is because both countries have high
institutional complexity and poor joint actions between the entities in charge, making articulation difficult, and creating over­
lapping and duplication of responsibilities. For instance, in Portugal, the national forestry authority has changed its institutional
structure six times in the last twenty years and there are more than 50 derived entities engaged in the prevention and fight against
wildfires (CTI, 2018). Consequently, it is not only difficult to coordinate cooperation at the national level in each country, but even
more at the cross-border level.
• Infrastructure and mobility. There is a weak development of jointly managed cross-border public transport services and deficiencies
in the density of railway lines, especially in areas such as Extremadura-Spain and Alentejo-Portugal. This is an obstacle to con­
nectivity and mobility between both sides of the border. This limitation becomes more relevant when it comes to emergency
management, since the lower the redundancy of the networks, the more likely it is that an area will be secluded. For instance, if the
emergency plan fails and transport networks collapse, some Portuguese areas could be isolated from Spain and other parts of
Europe.
• Digitalization. There are differences between the two countries regarding their connectivity and digital services, as well as the use
of cloud services and Big Data, which is lower than the EU average for both countries (Interreg Spain-Portugal, 2022). An example
of a digitalization issue is the difference in geographic data detail, explained in Section 4.3. This challenge is being studied by
projects such as Cross-forest (CEF, 2018), which aims a combine Forest Inventory Datasets from Portugal and Spain to support
forest management and forest protection.
• Hazard management. Wildfires are particularly frequent in the area, and both sides of the border present a high fire risk level. This
means that the cooperation between Portugal and Spain, which only covers emergency plans, must also be extended to prevention
and protection projects. In the analyzed case, the suppression capacity during EWE should guarantee response times lower than 60
min. If the capabilities of the emergency systems cannot make it possible, this implies that prevention and protection strategies
must complement wildfire management, thus reducing the hazard’s risk.

The countries sharing a border must implement actions to overcome these and other limitations in their cooperation. Adopting a
more resilient approach is also important, paying attention to all the wildfire management strategies, from prevention to suppression

12
E. Arango et al. Climate Risk Management 44 (2024) 100595

Table A-1
ROS values for spread and barrier sources – assumptions or observations, according to Arango et al. (2023a).
Spread sources ROS (m/ Assumptions - Comments
min)

Basics
Grassland 50 All values were taken in 10% Humidity and 20 km/h Wind condition.
Shrublands 25
Dry eucalypt forest 12
Pine plantations 7
Wet eucalypt forest 1
Fire Barriers
Wastelands 0.5 They are assumed as elements that do not spread fire, i.e., fulfill the function of a barrier.
Rivers and water bodies 1
Swimming pools 0.5
Firefighters 0.5
Forest
Invasive species 50 It is assigned the same ROS as grass due to the existence of Cortaderia Selloana, a species with high propagation speed.
Shrublands 35 This vegetation type is assumed to be large-scale plantations and therefore can increase fine flammable fuels at a
Eucalypt (plantation) 25 landscape scale. The mean values of models built with previous information in this area were used.
Maritime pine 30
(plantation)
Stone pine (plantation) 33 These values are established from the base values and expert judgment.
Cork oak 12
Other oaks 12
Chestnut 7
Hardwoods trees 7
Farming
Poor irrigated farming 25 Farming is assumed to be combustible and therefore spread the fire.
Olive farming 15 These values are established from the base value ”Poor irrigate farming” and expert judgment.
Constantly irrigated 10
farming
Wine farming 10
Rice farming 1
Other activities
Gas stations 200 Assuming an average damage radius of 550 m and a 60s time to reach it.
Industrial 450 Assuming 3 km of heavy damage recorded on the Beirut (Vancouver) blast and 400s to reach that blast radius.
Subestations and Power 250 An average explosion radius of 250 m is assumed and 1 min is considered to reach that radius for the most critical
plants scenario.
Commercial 170 A blast radius of 804 m according to the reference and time of 284 s is assumed.
Sports 25 Assuming that they are mostly grass fields or with green areas such as pastures.
Buildings 7.2 Assuming fire-spread mechanisms between buildings with non-combustible claddings through non-fire-rated roofs or
wall openings is use a critical separation distance a broken window of 18 m and 2.5 min to reach that length.

Table A-2
Additional ROS values for spread and barrier sources – assumptions or observations.
Spread sources ROS (m/ Assumptions/ Observations
min)

Poor irrigated farming 25 It is assumed to be fuel due to the presence of weeds and poor irrigation.
Olive farming 15 Derived from the other values.
Irrigated farming (constantly 10 They are assumed to be more careful crops regarding weeds and the constant presence of water prevents a rapid
irrigated) fire spread.
Wine farming 10 Derived from the other values.
Rice farming 1 It is assumed as irrigated rice crops, paddies filled with water (Kraehmer et al., 2017). It, therefore, acts more as a
barrier than as a propagation source.
Water Bodies 1 It is assumed as an element that does not spread the fire. It is a barrier.

and recovery. In addition, cross-border collaboration should be done around all the strategies and not only during emergencies. For
instance, differences in the level of prevention and protection may have implications in the long term for both countries. At the
emergency level, any problems in the coordination of the emergency services of the two countries can have serious societal conse­
quences. In this sense, the presented GIS-FA tool may help to identify and overcome such limitations and support decision-making at all
stages of wildfire management.

6. Conclusions

Adequately addressing the new wildfire regime requires reconsidering the policies in wildfire management. The persistence of
applying only reactive and short-sighted wildfire management policies results in fighting the symptoms of the problem but not the

13
 E. Arango et al.
Table B-1
Exposure level of the Elvas-Badajoz transport networks for different wildfire categories.
Country Network Wildfire Category → Cat 1 Cat 2 Cat 3 Cat 4 Cat 5 Cat 6 Cat 7

Exposure Level ↓ km % km % km % km % km % km % km %

Portugal Railway Very High

High 11 49%
Intermediate 11 49% 11 49% 6,7 30%
Moderate 3,5 15% 6,7 30% 4,9 22%
Low 22,6 100% 22,6 100% 22,6 100% 22,6 100% 8,1 36% 4,9 22%

Spain Railway Very High

High
Intermediate 5,2 100%
14

Moderate 3,4 65% 3,4 65%

Low 5,2 100% 5,2 100% 5,2 100% 5,2 100% 1,8 35% 1,8 35%

Portugal Roads Very High 0,9 1% 4,7 7% 14,8 22%

High 0,9 1% 14,5 21% 10,0 15% 33,0 48%
Intermediate 0,9 1% 6,4 9% 33,0 48% 38,3 56% 17,8 26%
Moderate 0,9 1% 3,9 6% 7,5 11% 12,4 18% 11,6 17% 2,9 4%
Low 68,5 100% 67,6 99% 63,7 93% 53,7 78% 8,4 12% 3,8 5%

Spain Roads Very High

High 2,0 4%
Intermediate 2 4% 5,3 10% 51,9 95%
Moderate 35 64% 45 83% 0,5 1%

Climate Risk Management 44 (2024) 100595

Low 54,5 100% 54,5 100% 54,5 100% 54,5 100% 18 32% 3,7 7%
E. Arango et al. Climate Risk Management 44 (2024) 100595

underlying problem. There is a pressing need to evolve towards more resilient policies that are effective in combating fire problems and
are sustainable in the long term. This change of vision does not imply diminishing wildfire suppression efforts, but it does require more
attention and investment in other aspects such as prevention and protection. In that sense, a balance should be found between these
strategies.
One of the barriers to achieving this policy shift is the lack of adequate decision-support tools for the various levels of wildfire
management. Those tools should be formulated from a resilience point of view. This paper shows how the GIS-FA tool can be effec­
tively used to support decision-making at the protection and emergency management levels. Due to its ease of application, it can be
implemented for different components of the built environment including in cross-border areas. It can be applied at the asset level such
as buildings, factories, and bridges, or the infrastructure level such as roads, railways, and electric networks. Also, it can help to
support decision-making policies regarding land management, e.g., crops, for which some tool adjustment is necessary.
There are two main specifications of applying this tool to cross-border case studies: (i) the determination of assets or targets, such as
railway and road transport, must be defined according to the boundaries of the respective countries. This is because each country
makes the decisions about these assets; (ii) the accuracy of the exposure assessment depends on the quality of the database. For
instance, the uncertainty of the exposure assessment for the Spanish area is higher than for the Portuguese area, due to the simplified
classification of the forest. Hence, it is important that countries have standardized (or homogeneous) databases to allow the same
assessment quality and better cooperation between countries. As demonstrated, the tool can support long-term decision-making, as
well as emergency planning. The obtained exposition maps can be combined with real-time information, such as fire ignition points to
identify the optimal evacuation routes and road closures.
The GIS-FA tool can be also used to determine the extent of the reliance on suppression activities. In the analyzed case, Fig. 9(b),
suppression actions may not be sufficient to cope with EWE in either country. Both transport networks exhibit reaction times of less
than 180 min. It also establishes the areas of priority and responsibility to be considered by each country.
The new wildfire management policies should be aligned at the cross-border level. Poor wildfire management in one country can
affect the neighboring country because, although the fire may not reach the neighboring country, its consequences will. Therefore,
cooperation between two border countries such as Portugal and Spain is essential for the defense of their population against the
increased risk of extreme wildfire events. However, achieving successful coordination between two countries for the management of
hazards is extremely complex. In this regard, there are several constraints ranging from digitization to governmental and adminis­
trative barriers. One of the major limitations of cross-border coordination between Portugal and Spain is that it exists only for
emergency management. Therefore, it is important to extend cross-border coordination to other aspects of wildfire management such
as prevention and protection.

Acknowledgements/Funding

This work was partly financed by FCT/ MCTES through national funds (PIDDAC) under the R&D Unit Institute for Sustainability
and Innovation in Structural Engineering (ISISE), under reference UIDB/ 04029/2020 (doi.org/10.54499/UIDB/04029/2020), and
under the Associate Laboratory Advanced Production and Intelligent Systems ARISE under reference LA/P/0112/2020.
This work is financed by national funds through the FCT, Foundation for Science and Technology, under grant agreement
2020.05755.BD attributed to the first author.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to
influence the work reported in this paper.

Data availability

Data will be made available on request.

Appendix A. Input data: ROS Values

ROS values considered for the exposure analysis are given in Table A-1 according to Arango et al. (2023a). Additional ROS values
are detailed in Table A-2.

Appendix B. Outputs: Exposure level of the Land Transportation System

see Table B-1

15
E. Arango et al. Climate Risk Management 44 (2024) 100595

References

Alcasena, F., Ager, A., Le Page, Y., Bessa, P., Loureiro, C., Oliveira, T., 2021. Assessing wildfire exposure to communities and protected areas in Portugal. Fire 4 (4).
https://doi.org/10.3390/fire4040082.
Andrade, S.R., Hulse, D.E., 2023. Evaluation and improvement of system-of-systems resilience in a simulation of wildfire emergency response. IEEE Syst. J. 17 (2),
1877–1888. https://doi.org/10.1109/JSYST.2022.3169125.
Arango, E., Nogal, M., Sousa, H.S., Matos, J.C., Stewart, M.G., 2023a. GIS-based methodology for prioritization of preparedness interventions on road transport under
wildfire events. International Journal of Disaster Risk Reduction 99, 104126. https://doi.org/10.1016/j.ijdrr.2023.104126.
Arango, E., Nogal, M., Jiménez, P., Sousa, H.S., Stewart, M.G., Matos, J.C., 2023b. Policies towards the resilience of road-based transport networks to wildfire events.
The Iberian case. Transportation Research Procedia 71, 61–68. https://doi.org/10.1016/j.trpro.2023.11.058.
Arango, E., Nogal, M., Sousa, H.S., Matos, J.C., Stewart, M.G., 2024. Improving societal resilience through a GIS-based approach to manage road transport networks
under wildfire hazards. Transportation Engineering 15, 100219. https://doi.org/10.1016/j.treng.2023.100219. URL: https://www.sciencedirect.com/science/
article/pii/S2666691X23000593.
ARIEM, Inter-regional Reciprocal Assistance in Emergencies (2015). URL: https://ariemplus.eu/pt/projeto.
Asori, M., Emmanuel, D., Dumedah, G., 2020. Wildfire hazard and risk modelling in the northern regions of Ghana using GIS-based multi-criteria decision making
analysis. J. Environ. Earth Sci. 10 (11), 2224–3216. https://doi.org/10.7176/JEES/10-11-02.
Baudena, M., Vallejo, V.R., Baeza, J., Moghli, A., Valdecantos, A., Santana, V.M., 2023. Long-term management actions of fire-prone Mediterranean ecosystems under
climate change using fuel reduction and post-fire restoration. EGU General Assembly 2023. https://doi.org/10.5194/egusphere-egu23-9283.
BBC, 2020. Australia fires: A visual guide to the bushfire crisis. URL: https://www.bbc.com/news/world-australia-50951043.
Beighley, M., Hyde, A.C., 2018. Portugal wildfire management in a new era assessing fire risks, resources and reforms, Tech. rep., Beighley and Hyde (2018). URL:
https://www.isa.ulisboa.pt/files/events/pub/2018_Portugal-Wildfire-Management-in-a-New-Era_Portuguese.pdf.
BOE-A-2019-12928, Protocolo adicional entre el Reino de España y la República Portuguesa sobre ayuda mutua en zonas fronterizas, BOE núm. 217, de 10 de
Septiembre de 2019, Tech. rep., Ministerio de Asuntos Exteriores, Unión Europea y Cooperación (2019). URL: https://www.boe.es/eli/es/ai/2018/11/21/(1).
BOE-A-2033-21339, Ley 43/2003, de 21 de Noviembre, de montes. Boletín oficial del Estado Español, Tech. rep., Ministerio de Agricultura y Ganadería/ Ministry of
Agriculture and Livestock, Spain (2003). URL: https://www.boe.es/eli/es/l/2003/11/21/43/con.
Cardoso, F., Moreno, J., Vallejo, V., Xanthopoulos, G., 2018. Forest fires - parking firesmart polices in the EU. Tech. rep., European commission, Brussels (11. URL
https://research-and-innovation.ec.europa.eu/knowledge-publications-tools-and-data/publications/all-publications/forest-fires-sparking-firesmart-policies-eu_
en.
Castanho, R.A., Loures, L., Cabezas, J., Fernández-Pozo, L., 2017. Cross-border cooperation (CBC) in Southern Europe—an Iberian case study. The eurocity Elvas-
Badajoz. Sustainability 9 (3). https://doi.org/10.3390/su9030360.
Castellnou, M., Prat-Guitart, E., Arilla, N., et al., 2019. Empowering strategic decision-making for wildfire management: avoiding the fear trap and creating a resilient
landscape. Fire Ecology 15 (31). https://doi.org/10.1186/s42408-019-0048-6.
CEF, CROSS Harmonization & HPC modelization of forest Datasets (2018). URL: https://crossforest.eu/presentation/.
CNIG, Corine land cover (2020). doi:10.7419/162.09.2020. URL: http://centrodedescargas.cnig.es/CentroDescargas/index.jsp.
Costafreda-Aumedes, S., Comas, C., Vega-Garcia, C., 2017. Human-caused fire occurrence modelling in perspective: A review. Int. J. Wildland Fire 26 (12), 983–998.
https://doi.org/10.1071/WF17026.
CTI, análise e apuramento dos factos relativos aos incêndios que ocorreram em Pedrógão Grande, Castanheira de Pera, Ansião, Alvaiázere, Figueiró dos Vinhos,
Arganil, Góis, Penela, Pampilhosa da Serra, Oleiros e Sertã, entre 17 e 24 de junho de 2017, Tech. rep., Comissao Técnica Independente, Assembleia da República,
Portugal (2017). URL: https://mysl.nl/oYQB.
CTI, Avaliação dos incêndios ocorridos entre 14 e 16 de Outubro de 2017 em Portugal continental, Tech. rep., Comissao Técnica Independente, Assembleia da
República, Portugal (2018). URL: https://mysl.nl/fxuf.
DGT, National geographic information system, land use and occupation charter (2022). URL: https://lc.cx/8ZSqjV.
Diario da República n.◦ 158/2017, Law 76/2017, Sistema de defesa da floresta contra incêndios. Diário da república n.◦ 158/2017, Série I de 2017–08-17, Tech. rep.,
Assambleia da República, Portugal (2017). URL: https://dre.pt/dre/detalhe/lei/76-2017-108010872.
Duane, A., Castellnou, M., Brotons, L., 2021. Towards a comprehensive look at global drivers of novel extreme wildfire events. Climatic Change. https://doi.org/
10.1007/s10584-021-03066-4.
Duane, A., Trasobares, A., Górriz, E., Casafont, L., Maltoni, S., 2022. The FIRE-RES project: Innovative technologies and socio-ecological and economic solutions for
fire resilient territories in Europe. Environmental Sciences Proceedings 17 (1). https://doi.org/10.3390/environsciproc2022017100.
EC - European Commission.Cardoso Castro Rego, F., Moreno Rodríguez, J., Vallejo Calzada, V., Xanthopoulos, G., 2018. Forest fires – Sparking firesmart policies in
the EU, Publications. Office. doi:10.2777/181450.
EFFIS, EFFIS Wildfire Risk Viewer (2022). URL: https://effis.jrc.ec.europa.eu/apps/fire.risk.viewer/.
EMPOD, An Emergency Management Framework for Canada, Tech. rep., Emergency Management Policy and Outreach Directorate Public Safety Canada, Ottawa, ON
(2017). URL: https://www.publicsafety.gc.ca/cnt/rsrcs/pblctns/2017-mrgnc-mngmnt-frmwrk/2017-mrgnc-mngmnt-frmwrk-en.pdf.
Fernandes, P.M., Pacheco, A.P., Almeida, R., Claro, J., 2016. The role of fire-suppression force in limiting the spread of extremely large forest fires in Portugal. Eur. J.
Forest Res. 253–262. https://doi.org/10.1007/s10342-015-0933-8.
Fernandes, P.M., Delogu, G.M., Leone, V., Ascoli, D., 2020. 10 - wildfire policies contribution to foster extreme wildfires. In: Tedim, F., Leone, V., McGee, T.K. (Eds.),
Extreme Wildfire Events and Disasters, Elsevier, 2020, pp. 187–200. doi: 10.1016/B978-0-12-815721-3.00010-2.
Interreg Spain-Portugal, FIREPOCTEP project (2020). URL: https://www.cartif.es/firepoctep/.
Galizia, L., Rodrigues, M., 2019. Modeling the influence of eucalypt plantation on wildfire occurrence in the Brazilian savanna biome. Forests 10 (10). https://doi.org/
10.3390/f10100844.
Ganteaume, A., Barbero, R., Jappiot, M., Maillé, E., 2021. Understanding future changes to fires in Southern Europe and their impacts on the wildland-urban interface.
Journal of Safety Science and Resilience 2 (1), 20–29. https://doi.org/10.1016/j.jnlssr.2021.01.001.
Gonzalez, F., 2021. Sixth-generation fires: Public-policy change toward prevention required to address large-scale events - International Association of Wildland Fire
(2021). URL: https://www.iawfonline.org/article/sixth-generation-fires-public/.
C.A. González, J.M.I. Vallejo, Código forestal 3: Incendios forestales, Tech. rep., Ministerio de la presidencia, justicia y relaciones con las cortes. Agencia Estatal
Boletín Oficial del Estado, Spain (2023). URL: https://www.boe.es/biblioteca_juridica/codigos/codigo.php?id=270_Codigo_Forestal_3_Incendios_
Forestales&modo=2.
Hassan, A., Betts, A., 2023. Maui Wildfires Latest: Lahaina Reopens to Residents (2023). URL: https://www.nytimes.com/article/maui-wildfires-hawaii.html.
Herbert, C., Haya, B.K., Stephens, S.L., Butsic, V., 2022. Managing nature-based solutions in fire-prone ecosystems: Competing management objectives in California
forests evaluated at a landscape scale. Frontiers in Forests and Global Change 5. https://doi.org/10.3389/ffgc.2022.957189.
Interreg Spain-Portugal, Forest fire defense on the Spanish-Portuguese border (2019). URL: https://poctep.eu/es/2014-2020/defensa-contra-incendios-forestales-en-
la-frontera-hispanolusa.
Interreg Spain-Portugal, Cooperation cross-border Spain - Portugal (POCTEP) 2021-2027 (2022). URL:https://interreg.eu/programme/interreg-spain-portugal-
poctep/.
Interreg, Centro Ibérico de Investigação e Combate aos Incêndios Florestais (2019). URL: https://cilifo.eu/project?lang=pt-pt.
IPCC, AR6 Synthesis Report: Climate Change 2022, Tech. rep., The Intergovernmental Panel on Climate Change (2022). URL: https://www.ipcc.ch/report/sixth-
assessment-report-cycle/.
Kraehmer, H., Thomas, C., Vidotto, F., 2017. Rice Production in Europe, Springer International Publishing. Cham 93–116. https://doi.org/10.1007/978-3-319-47516-
5_4.

16
E. Arango et al. Climate Risk Management 44 (2024) 100595

Leone V.,, Tedim, F., 2020. How to create a change in wildfire policies. In: Tedim, F., Leone, V., McGee, T.K. (Eds.). Extreme Wildfire Events and Disasters, Elsevier,
2020, pp. 217–232. doi: 10.1016/B978-0-12-815721-3.00012-6.
J.P. Martín, M.C. Ribau, A.J.S.R. da Silva, Cross-border cooperation in forest fire prevention and extinction in the Atlantic axis (2018). URL: https://www.
eixoatlantico.com/images/informes/secretaria_general/informe-incendios-ea_pt.pdf.
McCaffrey, S., Toman, E., Stidham, M., Shindler, B., 2015. Social Science Findings in the United States. Wildfire Hazards, Risks, and Disasters 15–34. https://doi.org/
10.1016/B978-0-12-410434-1.00002-6.
McLennan, J., Ryan, B., Bearman, C., Toh, K., 2019. Should we leave now? Behavioral factors in evacuation under wildfire threat, Fire Technology 55 (2), 487–516.
https://doi.org/10.1007/s10694-018-0753-8.
McNamee, M., Pagnon Eriksson, C., Wahlqvist, J., Johansson, N., 2022. A methodology for assessing wildfire hazard in sweden – the first step towards a multi-hazard
assessment method. Int. J. Disaster Risk Reduct. 83, 103415. https://doi.org/10.1016/j.ijdrr.2022.103415.
Menon, A., Vishnu-Menon, R., 2022. Management strategies for prevention of forest fire and environmental degradation in tropics with special reference to western
ghats of Kerala region, India. Journal of Tropical Forest Science 34 (1), 24–33.
Moreira, F., Ascoli, D., Safford, H., Adams, M.A., Moreno, J.M., Pereira, J.M., Catry, F.X., Armesto, J., Bond, W., González, M.E., Curt, T., Koutsias, N., McCaw, L.,
Price, O., Pausas, J.G., Rigolot, E., Stephens, S., Tavsanoglu, C., Vallejo, V.R., Van Wilgen, B.W., Xanthopoulos, G., Fernandes, P.M., 2020. Wildfire management
in Mediterranean-type regions: paradigm change needed. Environ. Res. Lett. 15 (1), 011001. https://doi.org/10.1088/1748-9326/AB541E.
Mourao, P.R., Martinho, V.D., 2019. Forest fire legislation: Reactive or proactive? Ecol. Ind. 104, 137–144. https://doi.org/10.1016/j.ecolind.2019.04.080.
Naderpour, M., Rizeei, H.M., Ramezani, F., 2021. Forest fire risk prediction: A spatial deep neural network-based framework. Remote Sensing 13 (13). https://doi.
org/10.3390/rs13132513.
Noble, P., Paveglio, T.B., 2020. Exploring adoption of the wildland fire decision support system: end user perspectives. J. Forest. 118 (2), 154–171. https://doi.org/
10.1093/jofore/fvz070.
NWCG, Introduction to Wild-land Student Workbook Fire Behavior S-190, Tech. rep., National Wildfire Coordination Group, The USA (2006). URL: https://mysl.nl/
TIDm.
OEH, NPWS Cultural Fire Management Policy, Tech. rep., Office of Environment and Heritage, State of New South (2016). URL: www.environment.nsw.gov.au.
Pinto, M.M., DaCamara, C.C., Trigo, I.F., Trigo, R.M., Turkman, K.F., 2018. Fire danger rating over Mediterranean Europe based on fire radiative power derived from
Meteosat. Nat. Hazards Earth Syst. Sci. 18 (2), 515–529. https://doi.org/10.5194/nhess-18-515-2018.
Reynolds, B.T., 2017. A History of the Prepare, Stay and Defend or Leave Early Policy in Victoria. Ph.D. thesis. RMIT University, Melbourne. URL: https://core.ac.uk/
download/pdf/98662407.pdf.
Rossi, D., Kuusela, O.-P., Dunn, C., 2022. A microeconometric analysis of wildfire suppression decisions in the western United States. Ecol. Econ. 200, 107525.
https://doi.org/10.1016/j.ecolecon.2022.107525.
Sakellariou, S., Sfougaris, A., Christopoulou, O., Tampekis, S., 2023. Spatial resilience to wildfires through the optimal deployment of firefighting resources: Impact of
topography on initial attack effectiveness. Int. J. Disaster Risk Sci. 98–112. https://doi.org/10.1007/s13753-023-00464-3.
Salis, M., Laconi, M., Ager, A.A., Alcasena, F.J., Arca, B., Lozano, O., Fernandes de Oliveira, A., Spano, D., 2016. Evaluating alternative fuel treatment strategies to
reduce wildfire losses in a Mediterranean area. For. Ecol. Manage. 368, 207–221. https://doi.org/10.1016/J.FORECO.2016.03.009.
Salis, M., Arca, B., Del Giudice, L., Palaiologou, P., Alcasena-Urdiroz, F., Ager, A., Fiori, M., Pellizzaro, G., Scarpa, C., Schirru, M., Ventura, A., Casula, M., Duce, P.,
2021. Application of simulation modeling for wildfire exposure and transmission assessment in Sardinia, Italy. Int. J. Disaster Risk Reduct. 58, 102189. https://
doi.org/10.1016/j.ijdrr.2021.102189.
Schultz, C.A., Miller, L.F., Greiner, S.M., Kooistra, C., 2021. A qualitative study on the us forest service’s risk management assistance efforts to improve wildfire
decision-making. Forests 12 (3). https://doi.org/10.3390/f12030344.
Song, J., Cannatella, D., Katsikis, N., 2023. Landscape-based fire resilience: Identifying interaction between landscape dynamics and fire regimes in the Mediterranean
region. In: Gervasi, O., Murgante, B., Rocha, A.M.A.C., Garau, C., Scorza, F., Karaca, Y., Torre, C.M. (Eds.), Computational Science and Its Applications – ICCSA
2023 Workshops. Springer Nature Switzerland, Cham, pp. 328–344.
Duarte Sousa and Francisco Soares and Hugo Neves and Isabel Silva and João Pina and Jorge Gomes and Marco Maia and Patrício Torres, Voluntários digitais em
situações de emergência (2018). URL: https://vost.pt/.
A. Subramaniam, A.S. Rao, A. Vajrapu, A.D.R.V., V.H.R, Real time monitoring of forest fires and wildfire spread prediction, in: 2023 2nd International Conference on
Applied Artificial Intelligence and Computing (ICAAIC), 2023, pp. 772–779. doi:10.1109/ICAAIC56838.2023.10141168.
Talebi, S., Vakilian, M., Bahrami, M., Lehtonen, M., 2022. Equipment hardening strategies to improve distribution system resilience against wildfire. in: 2022
International Conference on Smart Energy Systems and Technologies (SEST), pp. 1–6. https://doi.org/10.1109/SEST53650.2022.9898416.
Tedim, F., Leone, V., Amraoui, M., Bouillon, C., Coughlan, M.R., Delogu, G.M., Fernandes, P.M., Ferreira, C., McCaffrey, S., McGee, T.K., Parente, J., Paton, D.,
Pereira, M.G., Ribeiro, L.M., Viegas, D.X., Xanthopoulos, G., 2018. Defining extreme wildfire events: Difficulties, challenges, and impacts. Fire 1 (1). https://doi.
org/10.3390/fire1010009.
Tymstra, C., Stocks, B.J., Cai, X., Flannigan, M.D., 2020. Wildfire management in Canada: Review, challenges and opportunities. Progress in Disaster Science 5,
100045. https://doi.org/10.1016/J.PDISAS.2019.100045.
T.G. TYPSA, Estudio básico para la protección contra incendios forestales en la interfaz urbano-forestal: Legislación, Tech. rep., Ministerio para la Transición
Ecológica y el Reto Demográfico, Spain (2023). URL: https://www.miteco.gob.es/es/biodiversidad/temas/incendios-forestales/estudio_interfaz.html.
Z. Wang, H.T. Vo, M. Salehi, L.I. Rusu, C. Reeves, A. Phan, A large-scale spatio-temporal data analytics system for wildfire risk management, in: Proceedings of the
Fourth International ACM Workshop on Managing and Mining Enriched Geo-Spatial Data, GeoRich ’17, Association for Computing Machinery, New York, NY,
USA, 2017, p. 6. doi:10.1145/3080546.3080549.
WFLC, The National Strategy: The final Phase in the Development of the National Cohesive Wildland Fire Management Strategy, Tech. rep., Wildland Fire Leadership
Council (WFLC), Washington, DC (2014). URL: https://www.forestsandrangelands.gov/documents/strategy/strategy/CSPhaseIIINationalStrategyApr2014.pdf.
Whittaker, J., Haynes, K., Handmer, J., McLennan, J., 2013. Community safety during the 2009 Australian ’Black Saturday’ bushfires: an analysis of household
preparedness and response. International Journal of Wildland Fire 22 (6), 841–849. https://doi.org/10.1071/WF12010.
WWF - Association for the defense of nature. Informe sobre incendios forestales 2021 paisajes cortafuegos, https://www.wwf.es/informate/biblioteca_wwf/?57700/
Informe-sobre-incendios-forestales-2021-Paisajes-Cortafuegos#, accessed: 2022-09-19 (2021).

17