August 20, 2025 · Geosapiens
Traditionally recognized for our industry-leading flood models, Geosapiens is proud to announce a major milestone in its evolution into a comprehensive multi-risk climate intelligence firm. Today, we launch our new wildfire model, expanding our capabilities to address a broader spectrum of climate-related perils. This advancement empowers insurers and stakeholders with deeper, more integrated insights—strengthening resilience and decision-making in an increasingly complex risk landscape.
A science-driven, cutting-edge tool to help predict and manage the growing threat of wildfires
The 2023 wildfire season was the most destructive in Canada's history, scorching a record-breaking 18.5 million hectares of land. This catastrophic event, which forced the evacuation of thousands of people and sent plumes of smoke across North America and Europe, served as a stark reminder of the escalating threat of wildfires. As our climate changes, seasons of intense fire activity are becoming more common, leading to longer and more severe fire seasons and underscoring the urgent need for advanced tools to anticipate and manage this growing risk.
In response to this challenge, we are proud to announce the launch of a new, national-scale wildfire risk model for Canada, powered by a sophisticated deep learning framework. This innovative tool represents a significant leap forward in our ability to understand and predict where wildfires are most likely to occur.
The Science Behind the Model
At its core, our wildfire model leverages artificial intelligence—specifically, a deep learning framework—trained on a comprehensive dataset of 77 wildfire risk factors across Canada. These include:
Climate: Several climatic factors are used, including long-term temperature and precipitation patterns that create the underlying conditions for fire.
Vegetation: The types of forests and other plant life that can act as fuel.
Topography: The shape of the land, which can affect how a fire spreads.
Human factors: The presence of roads and communities, which can influence both ignition and suppression.
By learning from historical wildfire events and analyzing how these factors interact, the model generates continuous property-level risk scores, offering a far more precise picture of wildfire susceptibility than traditional broad-area estimates.
The result is a comprehensive, high-resolution cartographic tool of wildfire susceptibility across all of Canada. This isn't just a simple "high" or "low" risk assessment. Our model produces a continuous risk score, providing a more nuanced and detailed understanding of the fire potential in any given area.
Why This Matters for the Insurance and Financial Sectors
For insurers and financial institutions, accurate wildfire risk data is not only a business necessity — it is increasingly a regulatory requirement under OSFI B-15. The guideline emphasizes the need for institutions to integrate climate risk into governance, strategy, risk management, and disclosure practices. This means relying on precise, science-driven data is critical to remain compliant and competitive.
Our wildfire model supports these sectors in their regulatory compliance efforts, while also providing them with a strategic advantage. With its high-resolution insights, it enables them to:
Improved underwriting precision: Move beyond coarse regional hazard maps to assess risk at the individual property level, ensuring alignment with OSFI’s expectations on climate-related risk quantification.
Enhanced catastrophe modeling: Use detailed, data-driven risk insights to better quantify portfolio exposure and support regulatory reporting.
Smarter risk pricing: Reflect true hazard variability in premium calculations, improving both competitiveness and compliance with evolving governance standards.
These benefits translate directly into reduced uncertainty, more resilient portfolios, and better-informed risk transfer strategies.
Flexible Data Integration
At Geosapiens, we understand that seamless integration is key. That’s why our wildfire model outputs are designed to fit easily into existing workflows and systems, ensuring that users can maximize value without added complexity. Our wildfire modeling data can be delivered through:
API (DaaS): Direct integration into your underwriting, pricing, and portfolio management systems for real-time access to wildfire risk insights.
Web Portal (SaaS): A user-friendly platform that allows users to visualize, query, and analyze wildfire risk data without technical overhead.
Bulk Data Delivery: High-resolution datasets provided in standardized formats, enabling advanced internal modeling and analytics.
A Step Toward a More Resilient Future
While lightning and human carelessness are often the immediate triggers, the conditions that allow wildfires to ignite and spread are complex and driven by a combination of factors. Our new AI-powered model provides a crucial, evidence-based tool to help navigate this complexity. By offering a clearer, more detailed picture of wildfire risk, we can make more informed decisions to protect our communities, our forests, and our climate. This is just the beginning of a new era in wildfire management, one where cutting-edge technology helps us to stay one step ahead of the flames.