Attribution of the 2001-2010 global temperature plateau Virginie Guemas, Francisco J. Doblas-Reyes, Isabel Andreu-Burillo and Muhammad Asif International Workshop on Seasonal to Decadal Prediction, Toulouse, 13 May 2013
Experimental setup Methodology 1960 1960 Observations 1990 2012
Experimental setup Methodology 1961 1960 1961 1960 Observations 1990 2012
Experimental setup Methodology 1961 1960 1962 1961 1960 1962 Observations 1990 2012
Methodology Experimental setup 1961 1960 1962. every year 1990 until 2012 1961 1960 1962 Observations 1990 2012
Methodology Experimental setup 1961 1960 1962. every year 1990 until 2012 Greenhouse Gases + Aerosols + Solar Cycle 1961 1960 1962 Observations CMIP5 historical / 2005 / RCP4.5 1990 2012
Methodology Model
Methodology Full field Initialisation ERA40/ERAinterim ORAS4 GLORYS2V1 + NEMO-LIM run forced by DFS4.3
Methodology Analyses: Example : Focus on 3 rd forecast year 1961 Observations 1960 1962 1990 2012
Methodology Analyses: Example : Focus on 3 rd forecast year 1961 Observations 1960 1962 1990 2012
Methodology Analyses: Example : Focus on 3 rd forecast year 1961 Observations 1960 1962 1990 2012
Methodology Analyses: Example : Focus on 3 rd forecast year 51 forecasts : anomalies relative to 1971-2000 1961 Observations 1960 1962 1990 2012
Successful climate of the 2000-2010 global temperature plateau Global mean Sea Surface Temperature (60 S-60 N) Forecast years 1 to 3 ERSST The climate s capture the warming slowdown
Successful climate of the 2000-2010 global temperature plateau Global mean Sea Surface Temperature (60 S-60 N) Init EC-Earth historical simulations starting from 1850 preindustrial control simulations NoInit Initializing from observations is crucial to capture the plateau
Successful climate of the 2000-2010 global temperature plateau Global mean Sea Surface Temperature (60 S-60 N) Smoothing with 1-year running mean Init NoInit ERSST Initializing allows to the SST evolution along the s
Successful climate of the 2000-2010 global temperature plateau Global mean Sea Surface Temperature (60 S-60 N) Smoothing with 1-year running mean Init NoInit Initializing allows to the SST evolution along the s
Methodology Analyses: 3-year mean changes along the forecast 1960 Years 1 to 3 Years 3 to 5 Observations 1990 2012
Methodology Analyses: 3-year mean changes along the forecast Observations 1990 2012
Successful climate of the 2000-2010 global temperature plateau 3-year mean change in global SST (60 S-60 N) Init ERSST NoInit Initialization improves the SST trend along the forecast
Analysis of these s to attribute the 2000-2010 global temperature plateau Global Top-of Atmosphere Excess Energy Forecast years 2 to 4 Init CERES NoInit TOA input energy around 0.4 x 10 23 Joules captured
Analysis of these s to attribute the 2000-2010 global temperature plateau Global TOA Excess Energy Global Ocean Heat Uptake Init ORAS4 NoInit Increased Ocean Heat Uptake compensates for TOA inflow
Analysis of these s to attribute the 2000-2010 global temperature plateau Global Ocean Heat Uptake Init ORAS4 NoInit Largest ever recorded peak in ocean heat uptake
Analysis of these s to attribute the 2000-2010 global temperature plateau ORAS4 Ocean heat uptake (0-800m excluding the mixed layer) at the onset of the plateau The plateau seems due to increased ocean heat absorption
Analysis of these s to attribute the 2000-2010 global temperature plateau ORAS4 Init Increased ocean heat uptake in the Pacific captured by Init
Analysis of these s to attribute the 2000-2010 global temperature plateau ORAS4 NoInit Weak signals after ensemble-mean operator on NoInit
Conclusions Ec-Earth climate s capture the temperature plateau until 5 years ahead. The realism of the SST trend along the forecast is improved with initialization The Earth s heat budget shows that the TOA excess energy has been mainly absorbed in the ocean below the mixed layer at the onset of the plateau
Thank you very much for your attention virginie.guemas@ic3.cat This work was supported by the EU- funded SPECS project, under grant agreement 308378