Until very recently, the only sources of future climate information that were available to interested users were seasonal predictions and climate projections. The former provide a future outlook of the Earth’s climate system for a period ranging from 1 to 18 months into the future while the latter covers a continuous temporal range from the past century to the end of this century (or beyond) but with no relationship with the contemporaneous internal climate variability. Recently pioneered decadal climate prediction systems attempt to fill the gap that exists between these two timescales (i.e. from a year up to a decade). Predicting the variations in climate over this time horizon is considered one of the most challenging problems faced by the climate forecasting community due to the relatively weak constraints that can be applied on the internal variability and the relatively weak anthropogenic external forcings at this timescale. However, it is also of great interest to users because the period over which it provides information coincides with their operational planning.
Who we are
The Department of Earth Sciences of the Barcelona Supercomputing Centre-Centro Nacional de Supercomputación is one of the most active groups in climate prediction and climate services in Europe. The department is currently composed of over 100 people structured in four interacting research groups. The department’s mission is to perform research on and developing methods for environmental forecasting, with a particular focus on the atmosphere-ocean-biosphere system. This includes managing and transferring technology to support the main societal challenges through the use of models and data applications in high-performance computing and Big data infrastructures. It also includes the dissemination of real-time air quality and climate information based on its research expertise in collaboration with both the Spanish authorities and the World Meteorological Organisation (WMO). The decadal climate predictions displayed here result from the collaboration of three groups within the Earth Science department: the Climate Prediction group, the Earth System Services group and Computational Earth Sciences group.
Climate Prediction
The Climate Prediction group aims at developing climate prediction capability for time scales ranging from a few weeks to a few decades and from regional to global scales. This objective relies on a deep analysis of the strengths and weaknesses of state-of-the-art climate forecast systems, via a thorough comparison with the most up-to-date observational datasets, and on exploiting these detailed analyses to refine the representation of processes relevant to climate in our forecast systems and their initialization. The group activities focus both on understanding climate variability and the sources of predictability and improving forecast quality.
Earth System Services
The Earth System Service group facilitates the interpretation and application of research coming from the BSC-ES. The group also carries out applied research to demonstrate the ongoing value of these services to advance sustainable development in key sectors of society and economy such as renewable energy, urban development, insurance, agriculture, water management or health. The Earth System Services group aims at developing tailored services on weather and atmospheric composition model simulations (focusing on short-term time scales) and climate predictions (focusing on the sub-seasonal, seasonal and decadal timescales).
Computational Earth Sciences
The Computational Earth Sciences group is a multidisciplinary group with members of different technical and scientific profiles, that interacts closely with the other groups of the Department. The group provides help and guidance on the technical aspects of the scientists’ work and develops a framework that ensures an efficient use of high-performance computing resources.
Acknowledgements
The decadal climate prediction website has been developed as part of the CLINSA project (CGL2017-85791-R), funded by the Spanish Ministry of Economy and Competitiveness (MINECO).
Coordination
Team Member
Francisco J. Doblas-Reyes
Head of Earth Sciences Department
Markus Donat
Co-Leader of Climate Prediction Group, Climate scientist
Pablo Ortega
Co-Leader of Climate Prediction Group, Climate scientist
Team members
Team Member
Arthur Amaral
Climate scientist
Aude Carreric
Climate scientist
Balakrishnan Solaraju M.
Ph.D. student/Climate scientist
Carlos Delgado
Ph.D. student
Diana Urquiza
User Experience researcher
Étienne Tourigny
Climate scientist
Isadora Christel Jimenez
Science communication specialist
Louis-Philippe Caron
Climate scientist
Marina Conde
Front-end developer
Miguel Castrillo
Computer scientist
Nube Gonzalez-Reviriego
Postdoctoral Researcher
Pierre-Antoine Bretonnière
Computer scientist
Rashed Mahmood
Climate scientist
Roberto Bilbao
Climate scientist
Simon Wild
Climate scientist
Vladimir Lapin
Climate scientist
Yohan Ruprich-Robert
Climate scientist
FAQ
Decadal predictions are typically produced using a technique similar to that used for seasonal forecasts, i.e. by initializing a climate model and integrating it forward in time. These climate models are a mathematical representation of the Earth’s climate and are built using the basic laws of classical physics and thermodynamics. By initializing the model, we align its natural variability with that of the Earth climate system. The most common source of initial conditions are reanalyses data. Reanalyses are a combined form of observational data and climate models, thus representing a best estimate of the climate system at a specific time.
Decadal prediction systems will typically forecast surface variables such as temperature or precipitation over multiple years. Other times, they will forecast slowly varying climate oscillations that are known to impact the climate of certain regions such as the Atlantic Multidecadal Variability (AMV). It’s important to note that they do not predict the exact weather conditions at a precise date and location in the future, but only the average conditions over an extended period of time (in this case, multiple years).
Systems used in the context of decadal predictions typically include an atmosphere, ocean, sea ice and land surface components. The addition of other components (e.g. vegetation and carbon models) could potentially contribute to improving the skill of the forecasts and there is research currently underway to incorporate some of these components in the decadal prediction framework and study their impact. However, current decadal prediction systems are typically limited to these four components. More details on our forecasting system can be found in the Documentation section.
Yes. All of our data are freely available on the ESGF nodes.
Evaluating the quality of the predictions is considered a fundamental step in climate prediction because it assesses whether the prediction system can be trusted to forecast certain events. This verification process is typically based on validating extensive sets of hindcasts or retrospective predictions against observational references.
To evaluate the quality of our forecasts, we use both deterministic and probabilistic metrics. While deterministic measures provide information on the ensemble mean or a deterministic categorical forecast, probabilistic metrics attempt to evaluate the full hindcast distribution in order to provide a more comprehensive picture of system performance. In the Forecast section, we provide examples of such evaluation along with the actual forecast.
While decadal predictions have an enormous potential for helping a wide range of end-users, they are mostly a research activity at this stage and there are relatively few studies assessing the added value of decadal predictions for decision-making. Scientists from the Earth System Services group are investigating various ways this type of climate information can benefit stakeholders in different sectors. If you are interested in using this type of information, please get in touch.
FEEDBACK
Would you like to provide feedback on the website, on our products or ask us a question?
Please use the link below