Final Report Summary - SPECS (Seasonal-to-decadal climate Prediction for the improvement of EuropeanClimate Services)
Executive Summary:
SPECS was initially motivated by the need to develop 1) a new generation of European climate forecast systems that makes use of the latest scientific progress in climate modelling and in operational forecasting, 2) efficient local and regional forecast methods that produce skilful and reliable predictions over land areas for both the local and large scales, 3) clear examples of how actionable this climate information is for a range of stakeholders and 4) a strategy to disseminate and illustrate the usefulness of improved, high-quality climate prediction information and to integrate it with other climate services initiatives focusing mainly on the long-term climate change problem. SPECS aimed to identify the main challenges in s2d prediction and illustrate a range of solutions from a seamless perspective, both in terms of time scale, and between information producers and users. SPECS proposed a number of specific, innovative global forecast system experiments to test hypotheses for the improvement of s2d predictions. These experiments have delivered not only a better understanding of the role of the natural modes of variability, the initial state and the description of the crucial processes for climate prediction, but have also tested radical changes to forecast systems in terms of variable radiative forcing (natural and anthropogenic) and an increase in spatial resolution in global forecast systems. The improved systems have grown from existing state-of-the-art climate prediction models, data assimilation systems, objective combination and downscaling methods that have been developed in a disparate set of projects, informal cooperative initiatives and operational prediction activities that have emerged in the context of different time and spatial scales. SPECS also integrated multidimensional observational data sets of the coupled atmosphere-ocean-cryosphere-land surface system into the production of climate information, both for the preparation of better initial conditions and for the post-processing of the predictions. The predictions have been regionalised and improved through statistical combination from different sources. The increased understanding offers better estimates of the risk of high-impact extreme climatic events, a series of both recent and historical ones having been selected for detailed study. The project includes illustrating to what degree the proposed developments in current climate forecast systems contribute to improving the reliability, accuracy and value of regional and local operational climate forecasts and to disseminate actionable climate information. New services to convey both local and regional climate information based on predictions and its quality have been used in operational climate prediction and a wide list of stakeholders that have expressed their interest in interacting with SPECS.
In other words, SPECS delivers a new generation of European climate forecast systems, including initialised Earth System Models (ESMs) and efficient regionalisation tools to produce quasi-operational and actionable local climate information over land at s2d time scales with improved forecast quality and a focus on extreme climate events, and provide an enhanced communication protocol and services to satisfy the climate information needs of a wide range of public and private stakeholders. This grand challenge was at the centre of SPECS’ overarching objectives, pushing the boundaries of the current capability in climate forecasting and dissemination of information.
The SPECS philosophy is unique in that it has addressed its objectives by integrating and testing the consolidated knowledge on climate modelling and impact assessment generated by other EU-funded, nationally- and internationally-sponsored projects and operational activities to optimise the project outcome and achieve a maximum impact.
SPECS has been, among other things, the glue to bind together the outcome of many different research efforts that hardly took the climate prediction problem into account, climate services operators (both public and private) with few resources to individually make real developments of the products currently available and a unique set of stakeholders and international institutions willing to exploit the ambitious outcome from SPECS. As a result, government planning, society and European businesses now stand to benefit from having access to improved, reliable and better communicated s2d climate information.
Project Context and Objectives:
Despite a recent call by the World Meteorological Organisation (WMO) for more robust climate information to be used in economic, industrial and political planning, Europe lags behind in terms seasonal-to-decadal (s2d) climate forecasting. The lack of s2d climate prediction capability found in many of today’s research projects, along with the fact that many climate services focus on long-term climate change problems, makes the business of climate prediction an unknown player within the European context, and at the same time an interesting opportunity to provide new sources of climate information to a wide range of users. Traditionally, seasonal-to-decadal climate prediction had limited forecast quality, especially in Europe for shorter time scales. Furthermore, progress, measured by the forecast quality of operational forecast systems had been slow, mainly because new tools and model components for addressing the role of sea ice, land surface, stratosphere and ocean did not make their way into the systems providing real-time information. It was obvious that these systems could greatly benefit from untapped climate predictability.
SPECS was initially motivated by the need to develop 1) a new generation of European climate forecast systems that makes use of the latest scientific progress in climate modelling and in operational forecasting, 2) efficient local and regional forecast methods that produce skilful and reliable predictions over land areas for both the local and large scales, 3) clear examples of how actionable this climate information is for a range of stakeholders and 4) a strategy to disseminate and illustrate the usefulness of improved, high-quality climate prediction information and to integrate it with other climate services initiatives focusing mainly on the long-term climate change problem. SPECS aimed to identify the main challenges in s2d prediction and illustrate a range of solutions from a seamless perspective, both in terms of time scale, and between information producers and users. SPECS proposed a number of specific, innovative global forecast system experiments to test hypotheses for the improvement of s2d predictions. These experiments have delivered not only a better understanding of the role of the natural modes of variability, the initial state and the description of the crucial processes for climate prediction, but have also tested radical changes to forecast systems in terms of variable radiative forcing (natural and anthropogenic) and an increase in spatial resolution in global forecast systems. The improved systems have grown from existing state-of-the-art climate prediction models, data assimilation systems, objective combination and downscaling methods that have been developed in a disparate set of projects, informal cooperative initiatives and operational prediction activities that have emerged in the context of different time and spatial scales. SPECS also integrated multidimensional observational data sets of the coupled atmosphere-ocean-cryosphere-land surface system into the production of climate information, both for the preparation of better initial conditions and for the post-processing of the predictions. The predictions have been regionalised and improved through statistical combination from different sources. The increased understanding offers better estimates of the risk of high-impact extreme climatic events, a series of both recent and historical ones having been selected for detailed study. The project includes illustrating to what degree the proposed developments in current climate forecast systems contribute to improving the reliability, accuracy and value of regional and local operational climate forecasts and to disseminate actionable climate information. New services to convey both local and regional climate information based on predictions and its quality have been used in operational climate prediction and a wide list of stakeholders that have expressed their interest in interacting with SPECS.
In other words, SPECS delivers a new generation of European climate forecast systems, including initialised Earth System Models (ESMs) and efficient regionalisation tools to produce quasi-operational and actionable local climate information over land at s2d time scales with improved forecast quality and a focus on extreme climate events, and provide an enhanced communication protocol and services to satisfy the climate information needs of a wide range of public and private stakeholders. This grand challenge was at the centre of SPECS’ overarching objectives, pushing the boundaries of the current capability in climate forecasting and dissemination of information.
The SPECS philosophy is unique in that it has addressed its objectives by integrating and testing the consolidated knowledge on climate modelling and impact assessment generated by other EU-funded, nationally- and internationally-sponsored projects and operational activities to optimise the project outcome and achieve a maximum impact.
The SPECS project has completed 51 months of activity with some minor deviations from the original plan and a long list of achievements. New forecast systems have been developed, some after investing a large amount of technical and scientific effort (for instance, in those associated with the increased resolution experiments), and a large number of sensitivity experiments have been completed and made publicly available. The experimental set ups defined in the project have made their way into pre-operational suites, like the one testing the impact of the land-surface (soil moisture, snow and vegetation) and the sea-ice components. The experiments assessing the specification of the atmospheric composition and solar irradiance in climate forecast systems, which are a novelty of the project, have allowed the definition of much-needed standards for operational climate forecast systems. In all these cases process-based verification and model assessment have been undertaken, which led to an increased understanding of the mechanisms involved and allowed going beyond the model dependence of the results. The analysis of the selected extreme climate events defined in one of the cross-cutting topics, like the European summers or the North Atlantic multiannual shifts, has been used to build stories not generally contained in the comprehensive forecast quality assessments and to better engage with a number of stakeholders, particularly with the European EUPORIAS project partners. A word should be said about the development of common software for forecast quality assessment and bias adjustment, which has been regularly publicly released. It has had an impact, thanks also to its comprehensive documentation and tutorials, both in Europe, through for instance its inclusion in the Copernicus Climate Change Service, and beyond.
Many of the results obtained during this period have illustrated the importance of considering model improvement as not just a problem of the coupled model used, but also of the whole forecast system (including the initialisation). This means that while the initialisation approach should be developed and thoroughly tested as it happens with the climate models used in climate projections, the climate forecast system has to engage with other communities using the same climate models in different contexts (e.g. paleomodelling, long-term climate change). This is achieved, thanks to both the technical investment (in for instance data storage and dissemination) and coordination efforts of many SPECS partners, by liaising with the activities undertaken in the framework of the next Coupled Model Intercomparison Project or in the new climate modelling H2020 projects (like PRIMAVERA or APPLICATE), among others. Some partners have been heavily involved in discussions with other communities to ensure that the SPECS legacy goes beyond an improvement of operational climate forecast systems and, instead, pave the way for a more seamless way of working in climate modelling across many time scales. A good example of this new way of addressing climate forecast system development comes from the recent results on the initial shock and drift, some of the reasons suspected to limit the conversion of the untapped predictability into effective skill. As a result, SPECS has inspired the efforts of the World Climate Research Programme to bring together climate modellers and forecasters to decide how to interpret the model drift in climate forecasts using the information about the climate model systematic error in stationary mode.
Complementary forecasting tools, such as those used for empirical downscaling, statistical forecasting and multi-model combination have been thoroughly explored. Public software packages have been released to allow the community and a range of stakeholders to perform these analyses and produce the specific climate information that they require. This has been complemented with an unprecedented series of information factsheets, available from the SPECS web site, which target a wide audience. These factsheets try to satisfy the requests for basic information on climate forecasting, an aspect that has been widely covered for the climate change problem, but not for climate prediction.
These activities, and many others in the project, are ensuring a long-term legacy for SPECS and a substantial increase of the European contribution to the developing field of climate services. The fruitful collaboration established with the EUPORIAS project has also been very important. Both projects communicated regularly, exchanged information and shared data and results to maximise the return of the European investment in climate prediction and services research.
The main final results, following the specific objectives laid out in the initial proposal, have been:
• An objective exhaustive evaluation of current forecast quality from dynamical, statistical, and consolidated systems to identify the factors limiting s2d predictive capability
• Test of specific hypotheses for the improvement of s2d predictions, including novel mechanisms responsible for high-impact events using a process-based verification approach
• Innovative methods for a comprehensive forecast quality assessment, including the maximum skill currently attainable
• The integration of multidimensional observational data of the atmosphere-ocean-cryosphere-land system as sources of initial conditions, and to validate and calibrate climate predictions
• An improved forecast quality at regional scales by better initialising the different components, an increase in the spatial resolution of the global forecast systems and the introduction of important new process descriptions
• The assessment of the best alternatives to characterise and deal with the uncertainties in climate prediction from both dynamical and statistical perspectives for the increase of forecast reliability
• Reliable and accurate local-to-regional predictions via the combination and calibration of the information from different sources and a range of state-of-the-art regionalisation tools
• The illustration of the usefulness of the improvements for specific applications and the development of methodologies to better communicate actionable climate information to policy-makers, stakeholders and the public through peer-reviewed publications, e-based dissemination tools, multi-media, examples for specific stakeholders (energy and agriculture), stakeholder surveys, conferences and targeted workshops
• The support of the European contributions to WMO research initiatives on s2d prediction such as the GFCS and the enhancement of the European role on the provision of climate services according to WMO protocols by creating examples of improved tailored forecast-based products for the global producing centres and participating in their transfer to worldwide regional climate centres and national hydro-meteorological services.
SPECS has been, among other things, the glue to bind together the outcome of many different research efforts that hardly took the climate prediction problem into account, climate services operators (both public and private) with few resources to individually make real developments of the products currently available and a unique set of stakeholders and international institutions willing to exploit the ambitious outcome from SPECS. As a result, government planning, society and European businesses now stand to benefit from having access to improved, reliable and better communicated s2d climate information.
Project Results:
See document attached.
Potential Impact:
See document attached.
List of Websites:
The public website of the SPECS project is www.specs-fp7.eu.
The SPECS contacts are the following:
- Scientific coordinator and main contact: Prof Francisco J. Doblas-Reyes (francisco.doblas-reyes@bsc.es)
- Project manager: Mar Rodríguez (mar.rodriguez@bsc.es)
- Other issues: info@bsc.es
SPECS was initially motivated by the need to develop 1) a new generation of European climate forecast systems that makes use of the latest scientific progress in climate modelling and in operational forecasting, 2) efficient local and regional forecast methods that produce skilful and reliable predictions over land areas for both the local and large scales, 3) clear examples of how actionable this climate information is for a range of stakeholders and 4) a strategy to disseminate and illustrate the usefulness of improved, high-quality climate prediction information and to integrate it with other climate services initiatives focusing mainly on the long-term climate change problem. SPECS aimed to identify the main challenges in s2d prediction and illustrate a range of solutions from a seamless perspective, both in terms of time scale, and between information producers and users. SPECS proposed a number of specific, innovative global forecast system experiments to test hypotheses for the improvement of s2d predictions. These experiments have delivered not only a better understanding of the role of the natural modes of variability, the initial state and the description of the crucial processes for climate prediction, but have also tested radical changes to forecast systems in terms of variable radiative forcing (natural and anthropogenic) and an increase in spatial resolution in global forecast systems. The improved systems have grown from existing state-of-the-art climate prediction models, data assimilation systems, objective combination and downscaling methods that have been developed in a disparate set of projects, informal cooperative initiatives and operational prediction activities that have emerged in the context of different time and spatial scales. SPECS also integrated multidimensional observational data sets of the coupled atmosphere-ocean-cryosphere-land surface system into the production of climate information, both for the preparation of better initial conditions and for the post-processing of the predictions. The predictions have been regionalised and improved through statistical combination from different sources. The increased understanding offers better estimates of the risk of high-impact extreme climatic events, a series of both recent and historical ones having been selected for detailed study. The project includes illustrating to what degree the proposed developments in current climate forecast systems contribute to improving the reliability, accuracy and value of regional and local operational climate forecasts and to disseminate actionable climate information. New services to convey both local and regional climate information based on predictions and its quality have been used in operational climate prediction and a wide list of stakeholders that have expressed their interest in interacting with SPECS.
In other words, SPECS delivers a new generation of European climate forecast systems, including initialised Earth System Models (ESMs) and efficient regionalisation tools to produce quasi-operational and actionable local climate information over land at s2d time scales with improved forecast quality and a focus on extreme climate events, and provide an enhanced communication protocol and services to satisfy the climate information needs of a wide range of public and private stakeholders. This grand challenge was at the centre of SPECS’ overarching objectives, pushing the boundaries of the current capability in climate forecasting and dissemination of information.
The SPECS philosophy is unique in that it has addressed its objectives by integrating and testing the consolidated knowledge on climate modelling and impact assessment generated by other EU-funded, nationally- and internationally-sponsored projects and operational activities to optimise the project outcome and achieve a maximum impact.
SPECS has been, among other things, the glue to bind together the outcome of many different research efforts that hardly took the climate prediction problem into account, climate services operators (both public and private) with few resources to individually make real developments of the products currently available and a unique set of stakeholders and international institutions willing to exploit the ambitious outcome from SPECS. As a result, government planning, society and European businesses now stand to benefit from having access to improved, reliable and better communicated s2d climate information.
Project Context and Objectives:
Despite a recent call by the World Meteorological Organisation (WMO) for more robust climate information to be used in economic, industrial and political planning, Europe lags behind in terms seasonal-to-decadal (s2d) climate forecasting. The lack of s2d climate prediction capability found in many of today’s research projects, along with the fact that many climate services focus on long-term climate change problems, makes the business of climate prediction an unknown player within the European context, and at the same time an interesting opportunity to provide new sources of climate information to a wide range of users. Traditionally, seasonal-to-decadal climate prediction had limited forecast quality, especially in Europe for shorter time scales. Furthermore, progress, measured by the forecast quality of operational forecast systems had been slow, mainly because new tools and model components for addressing the role of sea ice, land surface, stratosphere and ocean did not make their way into the systems providing real-time information. It was obvious that these systems could greatly benefit from untapped climate predictability.
SPECS was initially motivated by the need to develop 1) a new generation of European climate forecast systems that makes use of the latest scientific progress in climate modelling and in operational forecasting, 2) efficient local and regional forecast methods that produce skilful and reliable predictions over land areas for both the local and large scales, 3) clear examples of how actionable this climate information is for a range of stakeholders and 4) a strategy to disseminate and illustrate the usefulness of improved, high-quality climate prediction information and to integrate it with other climate services initiatives focusing mainly on the long-term climate change problem. SPECS aimed to identify the main challenges in s2d prediction and illustrate a range of solutions from a seamless perspective, both in terms of time scale, and between information producers and users. SPECS proposed a number of specific, innovative global forecast system experiments to test hypotheses for the improvement of s2d predictions. These experiments have delivered not only a better understanding of the role of the natural modes of variability, the initial state and the description of the crucial processes for climate prediction, but have also tested radical changes to forecast systems in terms of variable radiative forcing (natural and anthropogenic) and an increase in spatial resolution in global forecast systems. The improved systems have grown from existing state-of-the-art climate prediction models, data assimilation systems, objective combination and downscaling methods that have been developed in a disparate set of projects, informal cooperative initiatives and operational prediction activities that have emerged in the context of different time and spatial scales. SPECS also integrated multidimensional observational data sets of the coupled atmosphere-ocean-cryosphere-land surface system into the production of climate information, both for the preparation of better initial conditions and for the post-processing of the predictions. The predictions have been regionalised and improved through statistical combination from different sources. The increased understanding offers better estimates of the risk of high-impact extreme climatic events, a series of both recent and historical ones having been selected for detailed study. The project includes illustrating to what degree the proposed developments in current climate forecast systems contribute to improving the reliability, accuracy and value of regional and local operational climate forecasts and to disseminate actionable climate information. New services to convey both local and regional climate information based on predictions and its quality have been used in operational climate prediction and a wide list of stakeholders that have expressed their interest in interacting with SPECS.
In other words, SPECS delivers a new generation of European climate forecast systems, including initialised Earth System Models (ESMs) and efficient regionalisation tools to produce quasi-operational and actionable local climate information over land at s2d time scales with improved forecast quality and a focus on extreme climate events, and provide an enhanced communication protocol and services to satisfy the climate information needs of a wide range of public and private stakeholders. This grand challenge was at the centre of SPECS’ overarching objectives, pushing the boundaries of the current capability in climate forecasting and dissemination of information.
The SPECS philosophy is unique in that it has addressed its objectives by integrating and testing the consolidated knowledge on climate modelling and impact assessment generated by other EU-funded, nationally- and internationally-sponsored projects and operational activities to optimise the project outcome and achieve a maximum impact.
The SPECS project has completed 51 months of activity with some minor deviations from the original plan and a long list of achievements. New forecast systems have been developed, some after investing a large amount of technical and scientific effort (for instance, in those associated with the increased resolution experiments), and a large number of sensitivity experiments have been completed and made publicly available. The experimental set ups defined in the project have made their way into pre-operational suites, like the one testing the impact of the land-surface (soil moisture, snow and vegetation) and the sea-ice components. The experiments assessing the specification of the atmospheric composition and solar irradiance in climate forecast systems, which are a novelty of the project, have allowed the definition of much-needed standards for operational climate forecast systems. In all these cases process-based verification and model assessment have been undertaken, which led to an increased understanding of the mechanisms involved and allowed going beyond the model dependence of the results. The analysis of the selected extreme climate events defined in one of the cross-cutting topics, like the European summers or the North Atlantic multiannual shifts, has been used to build stories not generally contained in the comprehensive forecast quality assessments and to better engage with a number of stakeholders, particularly with the European EUPORIAS project partners. A word should be said about the development of common software for forecast quality assessment and bias adjustment, which has been regularly publicly released. It has had an impact, thanks also to its comprehensive documentation and tutorials, both in Europe, through for instance its inclusion in the Copernicus Climate Change Service, and beyond.
Many of the results obtained during this period have illustrated the importance of considering model improvement as not just a problem of the coupled model used, but also of the whole forecast system (including the initialisation). This means that while the initialisation approach should be developed and thoroughly tested as it happens with the climate models used in climate projections, the climate forecast system has to engage with other communities using the same climate models in different contexts (e.g. paleomodelling, long-term climate change). This is achieved, thanks to both the technical investment (in for instance data storage and dissemination) and coordination efforts of many SPECS partners, by liaising with the activities undertaken in the framework of the next Coupled Model Intercomparison Project or in the new climate modelling H2020 projects (like PRIMAVERA or APPLICATE), among others. Some partners have been heavily involved in discussions with other communities to ensure that the SPECS legacy goes beyond an improvement of operational climate forecast systems and, instead, pave the way for a more seamless way of working in climate modelling across many time scales. A good example of this new way of addressing climate forecast system development comes from the recent results on the initial shock and drift, some of the reasons suspected to limit the conversion of the untapped predictability into effective skill. As a result, SPECS has inspired the efforts of the World Climate Research Programme to bring together climate modellers and forecasters to decide how to interpret the model drift in climate forecasts using the information about the climate model systematic error in stationary mode.
Complementary forecasting tools, such as those used for empirical downscaling, statistical forecasting and multi-model combination have been thoroughly explored. Public software packages have been released to allow the community and a range of stakeholders to perform these analyses and produce the specific climate information that they require. This has been complemented with an unprecedented series of information factsheets, available from the SPECS web site, which target a wide audience. These factsheets try to satisfy the requests for basic information on climate forecasting, an aspect that has been widely covered for the climate change problem, but not for climate prediction.
These activities, and many others in the project, are ensuring a long-term legacy for SPECS and a substantial increase of the European contribution to the developing field of climate services. The fruitful collaboration established with the EUPORIAS project has also been very important. Both projects communicated regularly, exchanged information and shared data and results to maximise the return of the European investment in climate prediction and services research.
The main final results, following the specific objectives laid out in the initial proposal, have been:
• An objective exhaustive evaluation of current forecast quality from dynamical, statistical, and consolidated systems to identify the factors limiting s2d predictive capability
• Test of specific hypotheses for the improvement of s2d predictions, including novel mechanisms responsible for high-impact events using a process-based verification approach
• Innovative methods for a comprehensive forecast quality assessment, including the maximum skill currently attainable
• The integration of multidimensional observational data of the atmosphere-ocean-cryosphere-land system as sources of initial conditions, and to validate and calibrate climate predictions
• An improved forecast quality at regional scales by better initialising the different components, an increase in the spatial resolution of the global forecast systems and the introduction of important new process descriptions
• The assessment of the best alternatives to characterise and deal with the uncertainties in climate prediction from both dynamical and statistical perspectives for the increase of forecast reliability
• Reliable and accurate local-to-regional predictions via the combination and calibration of the information from different sources and a range of state-of-the-art regionalisation tools
• The illustration of the usefulness of the improvements for specific applications and the development of methodologies to better communicate actionable climate information to policy-makers, stakeholders and the public through peer-reviewed publications, e-based dissemination tools, multi-media, examples for specific stakeholders (energy and agriculture), stakeholder surveys, conferences and targeted workshops
• The support of the European contributions to WMO research initiatives on s2d prediction such as the GFCS and the enhancement of the European role on the provision of climate services according to WMO protocols by creating examples of improved tailored forecast-based products for the global producing centres and participating in their transfer to worldwide regional climate centres and national hydro-meteorological services.
SPECS has been, among other things, the glue to bind together the outcome of many different research efforts that hardly took the climate prediction problem into account, climate services operators (both public and private) with few resources to individually make real developments of the products currently available and a unique set of stakeholders and international institutions willing to exploit the ambitious outcome from SPECS. As a result, government planning, society and European businesses now stand to benefit from having access to improved, reliable and better communicated s2d climate information.
Project Results:
See document attached.
Potential Impact:
See document attached.
List of Websites:
The public website of the SPECS project is www.specs-fp7.eu.
The SPECS contacts are the following:
- Scientific coordinator and main contact: Prof Francisco J. Doblas-Reyes (francisco.doblas-reyes@bsc.es)
- Project manager: Mar Rodríguez (mar.rodriguez@bsc.es)
- Other issues: info@bsc.es
