Summary

• The European Commission and Environment and Climate Change Canada signed a cooperation arrangement on 30 October.
• Wildfire modelling, Arctic environmental change, and Caribbean climate resilience are priority areas.
• The agreement builds on Destination Earth's existing digital twins, with further joint development still ahead.

The European Commission and Environment and Climate Change Canada have agreed to expand cooperation on advanced climate modelling, bringing Canadian scientific expertise into Destination Earth, the European programme developing detailed digital representations of the planet’s environmental systems.

The administrative arrangement was signed at the EU–Canada Summit in Montréal on 30 October. It establishes joint priorities covering wildfire behaviour, Arctic climate dynamics, and regional climate risk assessment, with the partners intending to combine data, research capabilities, and AI-assisted simulations.

Destination Earth, also known as DestinE, uses high-performance computing and numerical models to simulate environmental processes at different geographical scales. The programme is developing digital twins that allow researchers to examine extreme weather and longer-term climate change, including scenarios that differ from previously observed conditions.

The agreement extends the EU–Canada Digital Partnership established in December 2025. It creates a framework for further research and technical cooperation rather than confirming that new operational forecasting services have already been delivered to emergency responders or public authorities.

From environmental observations to simulations

Climate models draw on observations and physical equations to estimate how different parts of the Earth system interact. High-resolution simulations can examine local and regional behaviour in greater detail, although doing so requires substantial computing capacity and careful treatment of uncertainties in the input data and underlying models.

Destination Earth is implemented by three European organisations under the Commission’s direction. The European Centre for Medium-Range Weather Forecasts (ECMWF) develops its principal weather and climate digital twins, EUMETSAT provides the associated data lake, and the European Space Agency manages the core service platform.

The infrastructure is intended to make simulation results available through services that researchers and other users can access without directly operating the underlying supercomputers. This requires tools for preparing experiments, retrieving information, and interpreting outputs that can be complex and computationally demanding.

Unlike a conventional forecast prepared for a particular date, a digital twin can be used to investigate alternative conditions. Researchers might examine how a previously observed heatwave would unfold in a warmer climate or how changing background conditions could influence the effects of extreme rainfall.

Such experiments can help public authorities explore potential risks before infrastructure or emergency plans are finalised. They do not remove uncertainty, however, because simulation results depend on the quality of observations, the representation of physical processes, and the assumptions used to construct each scenario.

Canadian participation may broaden the information and expertise available for these investigations. Both regions experience environmental conditions whose effects cross borders, while their research institutions have complementary interests in high-latitude climate systems, extreme events, and adaptation planning.

Wildfire modelling becomes a joint priority

The cooperation identifies wildfire behaviour as an area for further development, with planned work on ignition, fire spread, and interactions between fires and the atmosphere. These processes influence how incidents evolve and the information needed by authorities preparing for possible emergencies.

Wildfire behaviour depends on several interacting conditions, including vegetation, temperature, wind, moisture, and terrain. A simulation that captures more of those relationships may help researchers investigate how a fire could develop under different circumstances.

Atmospheric feedback is particularly challenging because large fires can influence surrounding weather conditions while remaining affected by them. Combining environmental models with specialised fire research could improve understanding of those interactions, although the arrangement does not establish a measured increase in forecast accuracy.

Emergency services need information that can be acted upon within operational timescales. A more detailed scientific simulation may support planning or preparedness, while its suitability for immediate incident response depends on processing speed, reliability, and the way results are communicated.

The partners intend to investigate these capabilities through the existing Destination Earth infrastructure and Canadian research contributions. Practical applications will require validation before they can be relied upon in emergency operations.

Arctic change connects two continents

The Arctic forms another priority because changes in sea ice, frozen ground, and atmospheric conditions affect communities and infrastructure across northern regions. The agreement identifies sea ice decline, permafrost thaw, polar vortices, and stratospheric changes as subjects for strengthened modelling cooperation.

These processes interact over different timescales. Permafrost thaw can affect the stability of buildings and transport infrastructure, while changes in ice cover and atmospheric circulation can influence regional environmental conditions and the interpretation of longer-term climate risks.

Arctic observations can be difficult to collect consistently because of geographical remoteness and challenging operating conditions. Research organisations therefore combine satellite observations, ground measurements, and physical models to investigate areas where direct measurements are limited.

Canadian research expertise and environmental datasets could complement the European programme’s modelling capabilities, particularly where the partners investigate processes affecting both Canada and northern Europe.

Improved simulations may help authorities evaluate infrastructure exposure and adaptation options, although translating scientific information into investment decisions requires additional assessment of local conditions, engineering requirements, and the uncertainty associated with each scenario.

Regional resilience extends beyond Europe

The agreement also includes cooperation on high-resolution climate risk modelling for vulnerable regions, with the Caribbean identified as one potential beneficiary.

Regional modelling can support investigations into severe weather, coastal risks, and other environmental pressures where a country’s infrastructure and planning decisions depend on understanding changing conditions. Applying a global or continental model directly to local decisions may not provide sufficient detail, creating demand for more specialised simulations.

Building those capabilities requires regional data and engagement with the institutions responsible for using the results. Information intended for emergency planning may need different formats and update schedules from data used to assess long-term infrastructure investment.

The new arrangement establishes the intention to collaborate on these applications, while the practical scope of individual projects remains to be determined.

Destination Earth enters its operational phase

Destination Earth began its third phase in June 2026, following work to develop and operationalise its core digital twins. ECMWF and its partners are now concentrating on maintaining the systems, improving their integration, and making the resulting information more useful for climate-sensitive activities.

The current phase runs until June 2028. It includes further development of AI and machine-learning components and work to turn digital twin outputs into datasets that other researchers and software developers can use.

Those datasets could support applications developed through European AI infrastructure programmes, although individual services will need to demonstrate that their outputs are reliable and appropriate for the decisions they inform.

For public authorities and organisations managing physical assets, a useful climate model must connect scientific information with practical questions about location, exposure, costs, and available responses. Modelling can inform those choices, while decisions about investment and emergency management remain dependent on local circumstances.

The EU and Canada also agreed to expand cooperation through EuroHPC at the Montréal summit, providing a separate framework for collaboration on advanced computing infrastructure. Although the two agreements address different programmes, they connect scientific objectives with the computational resources required to investigate them.

Destination Earth’s existing digital twins provide a technical foundation for the new partnership. The next work will involve combining research capabilities and developing specific projects, with the value of the cooperation becoming clearer as validated modelling outputs and practical services emerge.