Fire-smart risk assessment is needed to tackle the scale of wildfire destruction, which is a growing reality across the globe. Hazardous fires are more intense and more frequent, fuelled both by climate change and by the no less significant human footprint on landscapes.
Wildfire data outlines a clear trend: we are facing increasingly devastating events that trigger disasters of previously unknown proportions. According to the European Environmental Agency 3,770 km² of land is burnt yearly on average, with 45,000 people displaced due to wildfires from 2008 to 2023, leading up to annual losses estimated at €2.5 billion in the European Union.
In the summer of 2025, Europe experienced its most extreme wildfires in the past two decades in terms of area burned. Intense fires in the Iberian Peninsula scorched 6,720 km2 of land, with 3,930 km2 in Spain alone, resulting in a tragic toll of eight fatalities.
At the other end of the globe, Chile has also seen staggering figures, leading to particularly painful disasters. In February 2024, the Valparaíso–Viña del Mar fire claimed 136 lives and destroyed nearly 7,000 homes. Similarly, this past January in Concepción–Penco, another blaze killed 21 people and levelled more than 2,000 residences.
Examining pyrogeography
Understanding the potential of these fires to devastate communities and ecosystems is vital. Consequently, recent pyrogeographical research focuses on analysing fire behaviour across diverse spatial and temporal scales. In this context, two technological approaches stand out as the most effective for evaluating fire impacts: remote sensing resources like satellite imagery, thermal sensors, and airborne platforms, which are essential for reconstructing past impacts, as well as for early detection and real-time monitoring of active fires; simulation and predictive tools allow us to identify landscape configurations that facilitate the ignition and spread of fire and comprehend the complexity of forest fires.
By applying these insights to land-use planning, we aim to confront a reality that day by day increases the wildfire risk within our communities.
Remote sensing for mapping the scars of past wildfires
To quantify the magnitude of these events, we harness satellite imagery and advanced analytical tools to assess two primary variables: intensity and severity.
Intensity measures the fire’s power, i.e. the rate at which energy is released during combustion and helps pinpoint thermal hotspots.
Severity, by contrast, gauges the aftermath: the physical damage left in the fire’s wake.
By analysing specific spectral ranges, we can quantify the drastic drop in vegetation productivity, effectively measuring the ecosystem’s struggle to recover.
The recent Barroca Grande fire (Portugal, August 2025) and the Trinitarias fire (Chile, January 2026) serve here as case studies. By combining NASA’s FIRMS thermal data with European Space Agency(ESA)‘s Copernicus Sentinel-3 imagery, we can visualise the fire crisis in both space and time. These images reveal a staggering reality: smoke plumes stretching for hundreds of kilometres across the atmosphere.
The intensity data reveals that over 95% of the total eventual footprint was burnt in a single day in Chile on January 18. This is the definition of ‘explosive fire behaviour’ – events so rapid they outpace traditional suppression efforts.
Beyond the immediate heat, our severity analysis provides metrics that are essential for recovery efforts. 57,782 hectares were scorched in Portugal.
By cross-referencing these damage levels with fuel types, local weather, and topography, we can design precise ecological restoration plans and help the agricultural sector rebuild in a way that is hopefully more resilient to future fires.
Author provided (no reuse)
Fire behaviour modelling and risk assessment
In the world of fire risk assessment and management, there are two prevalent strategies.
The most common uses short-term assessments: the daily fire risk indices we see on the news combine current weather danger with local vulnerability. This is the backbone, for instance, of the European Forest Fire Information System (EFFIS).
At the other end of the spectrum lies a more “strategic” tool called quantitative simulation. Rather than looking at what might happen tomorrow (or is currently happening), this approach uses advanced modelling techniques to guide long-term planning and risk mitigation.
To anticipate possible effects of warmer and drier fire seasons or landscape transformation (e.g., land abandonment), we assess wildfire exposure using a blend of empirical modelling (learning from the history of how fires have actually occurred and behaved) and stochastic modelling (using complex algorithms to play out thousands of “what-if” scenarios).
Essentially, we study past fires to evaluate how a landscape is likely to foster or fight future fires and assess to what extent we are potentially exposed or threatened by them. To quantify this exposure, we first identify the specific drivers that cause ignitions: human or natural. Then, we set thousands of theoretical fires loose across a ‘digital twin’ of the landscape.
We run these simulations under various climate settings to generate realistic patterns of fire exposure. The result is a set of clear, actionable metrics (Fig. 2) that tell us not just where a fire is likely to strike, but how virulent it will be.
This transition from reactive to proactive allows us to implement truly effective strategies. Whether it’s rearranging forest fuels, updating urban building codes, or designing fire-smart neighbourhoods, these decisions are rooted in data.
Courtesy of Marcos Rodrigues Mimbrero, Author provided (no reuse)
How this helps during emergencies
The true test of these technologies occurs during an active emergency. In an operational setting, fire spread modelling shifts from strategic
planning to a race against time. UC San Diego’s exemplary WIFIRE Program, for instance, provides real-time information to wildfire responders.
By integrating near-real-time satellite data with high-resolution weather forecasts, researchers can generate projections that predict a fire’s path over the coming hours.
One of the most effective tools in operational evacuation is the use of “isochrones” – contour lines on a map that represent the fire’s predicted arrival time (e.g. 30, 60, or 90 minutes from the current position).
Overlaying these contour lines with trigger points (specific ridges, roads, or landmarks) lets emergency managers automate the decision-making process.
The Axa science philanthropy is now part of the Axa Foundation for Human Progress, which brings together the commitments of Axa Group and Mutuelles d’Assurances in the fields of Science, Nature, Solidarity, and Culture. Before 2025, the global science philanthropy was held by the Axa Research Fund, which has supported over 750 projects around the world since its inception back in 2007. To learn more, visit Axa Foundation for Human Progress.
A weekly e-mail in English featuring expertise from scholars and researchers. It provides an introduction to the diversity of research coming out of the continent and considers some of the key issues facing European countries. Get the newsletter!
