Even with rapid action to reduce greenhouse gas emissions enough to prevent warming beyond 1.5 degrees, new strategies are needed to manage the impact of climate change. The increasing frequency of extreme weather events across Europe, including floods, wildfires and droughts, demonstrates the urgent need to build resilience to the effects of climate change. This requires managing both the impact of these events and the associated social and environmental risks, including threats to biodiversity, human health, and infrastructure.
Climate adaptation is the process by which technical, social, ecological, or economic systems make changes to increase their resilience to climate change. This process can involve a variety of interventions, including policy, infrastructure, environmental, and behavioural changes. With the aim of enabling climate adaptation across Europe, the goal of the EU’s Climate Adaptation Mission is to “support at least 150 European regions and communities to become climate resilient by 2030”.
To deliver this goal, the Mission aims to:
In delivering these objectives, AI can support climate adaptation efforts by:
Improving climate models and simulations
Complementing existing, physics-based models of the Earth’s climate system, AI can help improve climate predictions by enabling sophisticated simulations of climate sub-systems characterised by high uncertainty. Clouds, for example, affect the Earth’s temperature in different ways: bright clouds block sunlight, helping to cool the Earth, while dark clouds have the opposite effect. Processes such as cloud convection can be too computationally costly for physics-based models, but are amenable to AI-enabled analysis. The resulting hybrid models have been shown to increase the accuracy of forecasts, and are helping researchers develop better understanding of the Earth’s climate system.
Predicting extreme weather events
Extreme weather events in Europe have caused an estimated half a trillion euros in economic losses and between 85,000 and 145,000 human fatalities over the last 40 years. Predicting when and where such events – such as storms, heatwaves, and flooding – are likely to occur is important in enabling rapid policy responses. While climate models analyse long-term trends and meteorological models provide near-term weather information, seasonal forecasting of the type that can help prepare for extreme weather remains challenging. AI can bridge between weather and climate data, helping increase the accuracy of predictions relating to the location and duration of high-impact weather events and increasing the robustness of local forecasts.
Forecasting the impacts of climate change
Effective climate adaptation requires both the ability to forecast extreme weather events, and the ability to predict the impact of extreme weather events in different areas. At a local level, the climate, biosphere, land surface and geology, and human behaviours each influence how an extreme weather event will affect a community and its infrastructure. These factors are difficult to model and are not easily captured by physics-based models. AI can support the analysis of these complex scenarios, using various data sources to help develop early-warning systems that highlight where an extreme weather event is likely to translate into dangerous climate impacts.
Communicating climate risks
The pathway from Earth observation to policy decision involves complex interactions between data, analysis, and decision-maker. Through visualisations or simulations that help convey future scenarios, AI can help communicate the impact of climate change to decision-makers and inform decisions about which interventions to take.
Monitoring and analysis to help design adaptation measures
The Climate Adaptation Mission suggests a collection of interventions to support adaptation, including risk assessments, early warning systems, infrastructure management, nature-based solutions, human health protections, and food security measures. AI can play a role in supporting the design of these interventions. Many nature-based solutions, for example, rely on protecting biodiversity to ensure healthy ecosystems; AI-enabled analysis can help monitor biodiversity from satellite imagery.
