1 / 25

CHFP2: A coupled multi-seasonal forecast system for Canada

CHFP2: A coupled multi-seasonal forecast system for Canada. Bill Merryfield. Canadian Centre for Climate Modelling and Analysis Victoria, BC. In collaboration with : Woo-Sung Lee, Slava Kharin, George Boer, John Scinocca, Greg Flato (CCCma)

ira
Télécharger la présentation

CHFP2: A coupled multi-seasonal forecast system for Canada

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. CHFP2: A coupled multi-seasonal forecast system for Canada Bill Merryfield Canadian Centre for Climate Modelling and Analysis Victoria, BC In collaboration with: Woo-Sung Lee, Slava Kharin, George Boer, John Scinocca, Greg Flato (CCCma) Juan-Sebastian Fontecilla, Jacques Hodgson, Bertrand Denis, Benoit Archambault, Louis-Philippe Crevier, Lewis Poulin (CMC)

  2. WMO Global Producing Centresfor Long-Range Forecasts

  3. Environment Canada’s Current Multi-Seasonal Forecasts Statistical CCA Dynamical 2-Tier

  4. CHFP2 development CHFP2 = “Coupled Historical Forecasting Project, phase 2” • 2006 Funding from Canadian Foundation for Climate and Atmospheric Sciences (CFCAS) • 2007-2008 CHFP1 pilot project (existing AR4 model, simple SST nudging initialization) • 2008-2009 Model development leading to CanCM3/4, CHFP2 initialization development • 2009-2010 CHFP2 hindcast production • 2011 CHFP2 operational implementation

  5. CHFP2 Models CanAM4Atmospheric model - T63/L35 (2.8 spectral grid) - Deep conv as in CanCM3 - Shallow conv as per von Salzen & McFarlane (2002) - Improved radiation, aerosols CanAM3Atmospheric model - T63/L31 (2.8 spectral grid) - Deep convection scheme of Zhang & McFarlane (1995) - No shallow conv scheme - Also called AGCM3 CanOM4 Ocean model - 1.41°0.94°L40 - GM stirring, aniso visc - KPP+tidal mixing - Subsurface solar heating climatological chlorophyll SST bias vs OISST 1982-2009 C C

  6. ENSO variability in models Monthly SSTA standard deviation HadISST 1970-99 CHFP1 CGCM3.1 IPCC AR4, CanCM3 CHFP2 CHFP2 CanCM4 *

  7. AGCM assim (ERA) SST nudging (OISST) Sea ice nudging (HadISST) Forecast 1 Forecast 2 12 12 mos mos multiple assimilation runs CHFP2 hindcast initialization 3D ocean T, S assimilation (GODAS) Forecast + Anthropogenic forcing 1 Jun 1 Jul 1 Aug 1 Sep 1 Oct 1 May IC1 IC2 … … IC10 Forecast 10 12 mos

  8. AGCM assim (CMC) SST nudging (CMC) Sea ice nudging (CMC) Forecast 2 Forecast 1 12 12 mos mos multiple assimilation runs CHFP2 operational initialization 3D ocean T, S assimilation GODAS daily Forecast + Anthropogenic forcing 1 Jun 1 Jul 1 Aug 1 Sep 1 Oct 1 May IC1 IC2 … … IC10 Forecast 10 12 mos

  9. Benefits of coupled atmospheric assimilation: Improved land initialization Correlation of assimilation run vs offline analysis SST nudging only SST nudging + atmospheric assim Soil temperature (top layer) Soil moisture (top layer)

  10. zonal wind stress thermocline depth zonal surface current sea surface temp Benefits of coupled atmospheric assimilation: Improved ocean initialization Correlations vs obs in equatorial Pacific (5S5N) SST nudging + atmos assim SST nudging only

  11. OBS CanCM4 CanCM3 ENSO SkillCase Study: 1997-98 El Niño Niño3.4 hindcasts initialized monthly from 1 Jan 1997 to 1 Dec 1997 CHFP2 = CanCM3+4 1997 1998 1997 1998

  12. Nino3.4 Anomaly Correlation Skill -1  AC  1 CanCM3 CanCM4 CanCM3+4 Persistence Ensemble Forecasts Initialization 1 June 1979-2008 ERSST/OISST verification =1 perfect fcst =0 clim fcst } Damped persistence

  13. Nino3.4 Mean Square Skill Score (MSSS) -  MSSS  1 CanCM3 CanCM4 CanCM3+4 Persistence Ensemble Forecasts Initialization 1 June 1979-2008 ERSST/OISST verification =1 perfect fcst =0 clim fcst } Damped persistence

  14. Comparison with EU ENSEMBLES CanCM4 CanCM3 Comparison of individual model forecasts (ENSEMBLES forecasts use ensemble size 9, CCCma ensemble size 10), with CHFP2 skills shown for comparison

  15. Comparison with NCEP CFS CHFP2 vs CFSv1 CHFP2 vs CFSv2

  16. Anomaly correlation 2-tier vs CHFP2 2-tier CHFP2

  17. Anomaly correlation 2-tier vs CHFP2 Current system Current system Current system New system New system New system Global Land Ocean

  18. CHFP2 sea ice predictions Anomaly correlation, Sep mean ice concentration CanCM3 CanCM4 Forecasts initialized End of July Forecasts initialized End of June

  19. Probability forecast interface 3-category Probabilistic Forecast year=2010 JFM 0-month lead

  20. Probability forecast interface 3-category Probabilistic Forecast year=2010 JFM 0-month lead Local Probability Forecast Lat=53.6N Lon=62.8W

  21. Probability forecast verification Observed Temperature Percentile year=2010 JFM 3-category Probabilistic Forecast year=2010 JFM 0-month lead

  22. Competitive ENSO skill achieved with limited resources, low-tech ocean assimilation Conclusions

  23. Competitive ENSO skill achieved with limited resources, low-tech ocean assimilation CHFP2 to replace 2-tier + statistical system December 1 Conclusions

  24. Competitive ENSO skill achieved with limited resources, low-tech ocean assimilation CHFP2 to replace 2-tier + statistical system December 1 Watch out for La Nina! Conclusions

  25. New CMC http://iri.columbia.edu/climate/ENSO/currentinfo/SST_table.html

More Related