SPECS Seasonal-to-decadal climate Prediction for the improvement of European Climate Services F.J. Doblas-Reyes ICREA and IC3, Barcelona, Spain
Climate prediction time scales Progression from initial-value problems with weather forecasting at one end and multi-decadal to century projections as a forced boundary condition problem at the other, with climate prediction (sub-seasonal, seasonal and decadal) in the middle. Prediction involves initialization and systematic comparison with a simultaneous reference. Meehl et al. (2009) Climate services Climate adaptation 2
Seasonal predictions Correlation of System 3 seasonal forecasts of temperature (top) and precipitation (bottom) wrt GHCN and GPCC over 1981-2005. Only values significant with 80% conf. plotted. JJA T2m DJF T2m JJA Prec DJF Prec 4
Decadal predictions (Top) Near-surface temperature multi-model ensemble-mean correlation from CMIP5 decadal initialised predictions (1960-2005), five-year start date frequency;; (bottom) correlation difference with the uninitialised predictions of 2-5 year (left) and 6-9 year (right) wrt ERSST and GHCN. Init ensemblemean correlation Init minus NoInit ensemble-mean correlation difference Doblas-Reyes et al. (2013) 5
Averaged precipitation over 10ºW-10ºE for 1982-2008 for GPCP (climatology) and ECMWF System 4 (systematic error) with start dates November (6-month lead time), February (3) and May (0). (NOV ) (FEB) Systematic error: WAM (MAY ) GPCP climatology ECMWF S4 - GPCP (NOV) ECMWF S4 - GPCP (FEB) ECMWF S4 - GPCP (MAY) 6
Calibration/combination Seasonal forecasts of November Niño3.4 ERSST (four-month lead) with a persistence-based statistical model, the ECMWF System 3 and CFSV2 forecast systems and the combination of the two. Rodrigues et al. (2013) 7
SPECS motivation What: to produce quasi-operational and actionable local climate information Why: need information with improved forecast quality, a focus on extreme climate events and enhanced communication and services for RCOFs, NHMSs and a wide range of public and private stakeholders How: with a new generation of reliable European climate forecast systems, including initialised ESMs, efficient regionalisation tools and combination methods, and an enhanced dissemination and communication protocol Where: over land, focus on Europe, Africa, South America When: seasonal-to-decadal time scales over the longest possible observational period http://www.specs-fp7.eu 8
SPECS objective SPECS will deliver 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 seasonal-todecadal 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. Links to WCRP, WWRP, ENES, GEOSS for Climate 9
Consortium 20 partners, coordination IC3 No. Participant organisation name Participant legal name Country 1 Institut Català de Ciències del Clima IC3 ES 2 Instituto Nacional de Pesquisas Espaciais INPE BR 3 Max-Planck-Gesellschaft zur Förderung der Wissenschaften e.v. represented by Max-Planck-Institut für Meteorologie MPG DE 4 Het Koninklijk Nederlands Meteorologisch Instituut KNMI NL 5 Atmospheric, Oceanic and Planetary Physics, University of Oxford UOXF UK 6 Météo-France MeteoF FR 7 Centre Européen de Recherche et Formation Avancée en Calcul Scientifique CERFACS FR 8 Norsk Institutt for Luftforskning NILU NO 9 Agenzia Nazionale per le Nuove Tecnologie, l'energia e lo Sviluppo Economico Sostenibile ENEA IT 10 University of Leeds UNIVLeeds UK 11 University of Exeter UNEXE UK 12 Meteorologisk Institutt Met.no NO + Brazil 13 Vortex VORTEX ES 14 Met Office METOFFICE UK 15 Sveriges Meteorologiska Och Hydrologiska Institut SMHI SE Long list of affiliated partners and stakeholders, among which GEO 16 Institut Pierre et Simon Laplace, Centre National de la Recherche Scientifique CNRS 17 University of Reading UREAD UK 18 Agencia Estatal Consejo Superior de Investigaciones Científicas CSIC ES 19 European Centre for Medium-Range Weather Forecasts ECMWF UK 20 Universität Hamburg UniHH DE FR 10
SPECS is part of ECOMS European Climate Observations, Modelling and Services (ECOMS) initiative with these objectives: ensure close coordination between projects and activities in Europe in the area of seasonal to decadal climate predictions towards climate services provide thought leadership to the European Commission on future priorities in the area of seasonal to decadal climate predictions towards climate services. Three EU projects are the core of ECOMS: EUPORIAS, NACLIM and SPECS, with a total funding of 26 Meuros. All EU projects related to climate research and climate services are part of ECOMS. 11
Overall strategy SPECS should test in s2d mode the scientific solutions to all challenges in climate modelling. 12
SPECS impact SPECS and ECOMS bring together several communities: climate modelling, weather and climate forecasting, impact modelling, downscaling. The main project deliverables are a set of public tools and data from the most ambitious coordinate seasonal-to-decadal global prediction experiments to this date. Coordinated experiments Core: impact of soil moisture and sea-ice initialization, increased resolution, improved stratosphere and enhanced sample size Tier 1: impact of snow initialization, interactive vegetation/phenology, sensitivity to aerosol and solar irradiance. Central repository using modified CMIP5/CORDEX standards merged to CHFP. Large number of affiliated partners and stakeholders, including major international programmes. 13
Future GLACE-3 SPECS could be the European contribution to the future GLACE- 3. GLACE-3 will aim at assessing the impact of initializing soil moisture and snow. Four-month long forecasts, 10-member ensembles, 1986-2012, start in November, February, May, June, July, August, ERA- Interim based forcings (WFDEI, WATCH), prescribed and fixed vegetation maps and biophysical properties. Four experiments: CLIM: initialize with soil moisture and snow climatology. INI1: initialize with reanalysis of soil moisture and snow. PERFECTBOUNDARY: prescribe daily soil moisture and snow reanalysis. INI2: initialize with multi-model reanalysis of soil moisture and snow. 14
CMIP5 decadal predictions Decadal predictions (2-5 forecast years) from the CMIP5 multi-model (6 systems, initialized solid, historical and RCP4.5 dashed) over 1960-2005 for global-mean temperature, the Atlantic multi-decadal variability and the interdecadal Pacific oscillation. GISS and ERSST data used as reference. Correlation of the ensemble-mean prediction as function of forecast time. Grey area for the 95% confidence level. Root mean square error, where dots represent the forecast times for which Init and NoInit are significantly different at 95% confidence level. Doblas-Reyes et al. (2013) Forecast time (4-year averages) 15
Predictions of the recent hiatus (Top) First three forecast years of the decadal predictions (five-member ensembles performed with the EC- Earth2.3 forecast system over 1960-2005 for global-mean sea surface temperature. The ensemble mean is plotted with a thicker dot. (Bottom) Three historical simulations (extended with the RCP4.5 scenario) performed with EC-Earth2.3. The ensemble mean is plotted with a thicker line. ERSST data used as reference in black. Positive impact of the initialization. Guemas et al. (2013) 16
Hurricane frequency prediction Average number of hurricanes per year estimated from observations and from the CMIP5 multi-model decadal prediction ensemble (forecast years 1-5). The correlation of the ensemble mean for the initialized, uninitialized and statistical predictions are shown with the 95% confidence intervals. Caron et al. (2013) 17
From climate data to information Seasonal predictions of 10-metre wind speed from ECMWF System 4 from February over 1981-2010. Reference from ERA Interim. Correlation of the ensemble mean Probability forecast for spring 2011 (boxes for areas with both resources and skill) 18
SPECS and GFCS 19
SPECS links to IS-ENES SPECS should accelerate the integration of better climate products and services into adaptation processes, and to learn from the impact sector in a co-design process. The PRACE resources are essential for the success of SPECS, s2d being an example of how the climate community could make a good use of current HPC (but always working towards dedicated facilities). The public dissemination of the SPECS experiments using ESGF is an example of what the decadal prediction community will require for CMIP6 (CIM?), with the opportunity to promote the new standards via WGSIP, and of bringing closer together climate change and climate SPECS will link its data repository to the Climate4impact portal, while documenting what is different between a climate prediction and a projection. SPECS should play a central role in the design of the decadalprediction component of CMIP6, whose design is expected to be lead by WGSIP (e.g. suggest a transpose-cmip). Real-time decadal prediction exchange should continue and be enhanced (with more variables) whenever possible. SPECS can help linking ENES to GFCS and to the climate prediction (s2d) community. 20