Climate change is no longer a distant threat...
Its impacts are accelerating, with 2024 marking the hottest year ever recorded globally.
From devastating floods to record-breaking heatwaves, extreme weather is becoming more frequent and severe, affecting communities across the globe—though not equally.
Despite all the data, our ability to predict regional and local impacts remains limited, leaving policymakers and citizens ill-prepared for an increasingly unpredictable future.
That's why initiatives like the EU-funded EXPECT project are more needed than ever. Combining cutting-edge climate modelling, vast observational datasets and advanced AI tools, EXPECT is working to improve our ability to predict regional climate changes and extreme weather, delivering more accurate, actionable forecasts.
We talked to Markus Donat, EXPECT project Scientific Coordinator and Climate Variability and Change Co-leader at the Barcelona Supercomputing Center, to learn more.
What are the challenges involved in forecasting climate change and extreme weather events at regional level, and how does your approach improve on current models?
Climate models show a wide range of forcing responses, and this makes it difficult to understand what the ‘true’ climate response of the real-world climate system is. In particular, changes in atmospheric circulation are highly uncertain, and we have increasing evidence that climate models seem to underestimate the magnitude of atmospheric circulation changes.
EXPECT is addressing these challenges by exploiting a variety of different datasets, climate model simulations and observations in combination. In particular, large ensembles of single-forcing experiments are used to understand uncertainties related to model differences and internal climate variability in the response to specific forcings like greenhouse gases or aerosols. Based on this in-depth evaluation, we will develop process-based constraints to correct for model errors related to the responses to forcing and internal drivers. This will enable more accurate climate predictions for the next years to decades.
How might your findings help a community facing increased flooding or drought in the future?
By developing a capability for integrated attribution and prediction, we will provide improved predictions of such extreme weather events. Besides developing calibration methods to correct for potential model errors, by understanding the processes driving specific prediction signals, we can also enhance the confidence in predictions in situations that are known to be more predictable.
Extreme weather events often have cascading impacts on communities and ecosystems. How do you ensure your predictions are actionable for policymakers working on disaster preparedness and adaptation?
The main objectives of EXPECT are to generate new climate knowledge and develop accurate predictions based on this knowledge. EXPECT does not focus on climate impacts or decision-making processes. However, making the accurate predictions available and communicating them widely provides high-quality climate information that other activities or projects can pick up and integrate with their specific knowledge to understand their implications on impacts and apply the information for disaster preparedness and adaptation.
We’re becoming ever more aware of the need to translate findings and policy to the local level. How will you ensure your findings and tools are accessible and useful for communities and policymakers at local, national and international levels?
While the primary focus of EXPECT is to generate new climate knowledge and improved predictions, all data and findings will be made available to the public and wider science community to translate or apply the information for their specific needs. The annually updated near-term climate predictions enable regular updates also for derived forecast products, and are disseminated via the WMO Global Annual to Decadal Climate Update. Also, the scientific findings and tools developed by EXPECT will be made openly available. To this end, EXPECT is developing an open science platform, which will ensure the reproducibility of AI models developed within the project.
Climate change is advancing faster than originally predicted. How do you balance the need for long-term vision with providing more immediate, actionable information for decision-making now?
EXPECT will develop a prototype operational capability for integrated attribution and prediction of climate phenomena and extremes. This capability will be applied to the annually updated near-term climate predictions produced by the WMO Global Annual to Decadal Climate Update, which provides the most recent climate prediction information to policy-makers, stakeholders and the wider public as a basis for their decision-making.
The EXPECT project is structured around four key themes. Which holds the most potential for short or medium-term societal benefit, and why?
Our Research Themes range from bridging different data sources to generating new climate knowledge and enhancing predictions, to understanding and predicting climate hazards and building the infrastructure necessary for efficient data analysis.
These themes complement each other and work together towards developing the capability for integrated attribution and prediction. We have an overarching Work Package that integrates the key contributions from the different Research Themes to provide refined predictions and projections of extremes for the coming years, based on process-based evaluations and constraints applied to the prediction models. Accurate and well-explained climate predictions provide crucial information to society to underpin targeted adaptation efforts.
Get the latest climate change insights direct from the experts.
Sign up to our newsletter
Disclaimer
Created with funding from the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or the granting authority. Neither the European Union nor the granting authority can be held responsible for them.
The information and documentation in this section are published with permission from their respective sources. While the MAIA Project strives to ensure the accuracy and completeness of the content, we make no guarantees or warranties of any kind, express or implied, about the validity, reliability, or accuracy of the information. The original sources are solely responsible for the content they provide. We disclaim any liability for any errors, omissions, or inaccuracies in the information and for any actions taken in reliance on the information contained on this website.
