Seasonal Forecasts Using the RSM Current Status and Outstanding Issues

1. Seasonal Forecasts Using the RSM Current Status and Outstanding Issues Liqiang Sun International Research Institute for Climate and Society (IRI)

2. Regional Climate Modeling LAMs for climate studies • Giorgi and Bates (1989) – one-month simulation • Jones et al. (1995) – continuous multiyear simulations • More than 800 papers published in SCI journals (model development, climate variability, climate prediction and projection, …) RCM Reviews • McGregor (1997) • Giorgi and Mearns (1999) • Leung et al. (2003) • Wang et al. (2004) • Laprise (2006) • Sun and Ward (2007)

3. History of Climate Prediction Predictions based on scientific schemes: • Prediction for Indian Monsoon Rainfall (Blanford 1884) Predictions based, at least in part, on Dynamical Climate Models: • U.K. Met Office since 1988 • Climate Prediction Center since 1994 • Canadian Met Centre since 1995 • CPTEC since 1995 • Australia’s Bureau of Met since 1997 • IRI since 1997

4. History of Climate Prediction Predictions based, at least in part, on Regional Climate Models: • IRI since 1997 • ECPC since 1997 • NR&M (Queensland)/IRI 1998 • FUNCEME/IRI since 2001 • NCEPsince 2002 • CWB/IRI since 2003 • ICPAC/IRI since 2005 • SAWS/IRI since 2006 • ECPC/NTU,HKO, BIU since 2003 • Downscaling DEMETER Hindcasts • Challenges • Scientific issues related to predictability at smaller scales • Technical issues for regional climate modeling • Computational constrains

5. Regional Climate Modeling Smaller spatial scale features developed in regional models are attributed to four types of sources: • The small scale forcing, • the nonlinearities presented in the atmospheric dynamical equations, • hydrodynamic instabilities, and • the noise generated at the lateral boundaries and model errors

6. There is still skepticism toward RCMs in a segment of the modeling community • RCMs are not sufficiently evaluated. Most RCM efforts have been mostly isolated and unorganized. • RCMs are “spreading” too fast through an inexperienced user community • RCM output is taken too un-critically and it is not sufficiently “checked”

7. Introduction to Climate Dynamical Downscaling

8. Outline • Motivation • Values added by RCMs • Managing climate variability • Challenges and Opportunities in regional climate modeling • Summary

9. 1. Motivation RCMs for Seasonal Prediction (Dynamical Downscaling Prediction) • Enhance the scale and relevance of seasonal climate forecasts • and creating information to better support decisions • Advance our understanding of physical processes that contain predictability at smaller spatial and temporal scales • Advance our understanding of interactions between large and small scales and provide the physical evidence for statistical downscaling • Parameterization testbed for GCMs

10. 2. Values Added by RCMs On Seasonal Time Scale • Improvement of Spatial Patterns and Temporal Distribution • Predictability at Smaller Spatial and Temporal Scales

11. It is widely accepted that dynamical downscaling improves spatial patterns and climatologies as compared to the coarse resolution GCMs. Typical spectral distributions of the global model and the regional model Chen et al. (1999)

12. RSM (50km) ECHAM4.5 AGCM(T42) Vorticity Winds Vorticity Winds Precipitation Humidity Precipitation Humidity A Typical Tropical Cyclone Simulated by Climate Models Carmago, Li and Sun (2007)

13. Sun et al. (2005)

14. Climate Variability at Smaller Spatial and Temporal Scales – Is It POTENTIALLY Predictable? • Is the local information skillful? Or is it just noise on top of the large-scale signal? • Is the temporal character improved?

15. Sub-GCM spatial scale climate is POTENTIALLY predictable over many regions Observed DJF rainfall: ENSO composite

16. Qian et al. (2006)

17. Sun et al. (2005)

18. Seasonal Climate Prediction – Is It ACTUALLY Predictable?

20. CLIMATE DYNAMICAL DOWNSCALING FORECAST SYSTEM FOR NORDESTE • HISTORICAL DATA • Extended Simulations • Observations PERSISTED GLOBAL SST ANOMALIES ECHAM4.5 AGCM (T42) NCAR CAMS Persisted SSTA ensembles 1 Mo. lead 10 Post Processing PREDICTED SST ANOMALIES Tropical Pacific Ocean (LDEO Dynamical Model) (NCEP Dynamical Model)(NCEP Statistical CA Model) Tropical Altantic Ocean (CPTEC Statistical CCA Model) Tropical Indian Ocean (IRI Statistical CCA Model) Extratropical Oceans (Damped Persistence) Predicted SSTA ensembles 1-4 Mo. lead 24 RSM97 (60km) RAMS (40km) CPT AGCM INITIAL CONDITIONS UPDATED ENSEMBLES (10+) WITH OBSERVED SSTs IRI FUNCEME Sun et al. (2006)

21. Real-Time Forecast Validation

22. Skill comparison between the driving ECHAM forecasts and the nested RSM forecasts. The RPSS (%) was aggregated for the whole Nordeste region.

23. Managing Climate Variability

24. bridging Climate into Risk Management .. crop models need daily time sequences .. as do malaria models and hydrologic models

25. RSM Hindcast Validation FMAM Drought Index FMAM Rainfal Anomalies 200 5 OBS 4 OBS r=0.74 150 r=0.84 RSM RSM 3 100 2 50 1 0 0 -1 -50 -2 -100 -3 -150 -4 -200 -5 1970 1980 1990 2000 1970 1980 1990 2000 FMAM Flooding Index FMAM Weather Index 20 3 OBS OBS r=0.84 15 r=0.69 RSM RSM 2 10 1 5 0 0 -5 -1 -10 -2 -15 -20 -3 1970 1980 1990 2000 1970 1980 1990 2000

26. Linking prediction and application Corn Yield Prediction Corn Yield Anomaly (Kg/ha) Sun et al. (2007)

27. Proposed Framework for Streamflow Forecasting Block and Sun (2009)

28. Climatological (solid) and pooled ensemble hindcast (dashed) PDFs for total Jan-June 1991 streamflow. Observed streamflow show as dotted vertical line. Block and Sun (2009)

29. Challenges and Opportunities in RCM Seasonal Prediction • Observation • Model Physics • Air-Sea Coupling • Land Process • Nesting Strategy • Signal & Uncertainty • RCM Biases • Changing climate

30. Network of Rainfall Stations

31. Rainfall Anomalies (mm/day)

32. Case Study: Rainfall Diff:1983-1971 ~280km 250km 10km 50km Sensitivity Tests: Model resolution & Model physics Hu and Sun (2002)

33. Air-sea Interaction Correlations between SST and precipitation for the period MJJA 1979-2006. The areas exceeding the 95% significance level are shaded. Cha and Lee (2009)

34. Land initialization

35. Land Process Treatment of Groundwater Reservoir in climate models Ice Caps Present (Regional) climate Models 1. Atmosphere 2. Soil-Plants 3. Surface Water Land Surface Base of soil model ~ 1-3 m Continents Oceans 4. Groundwater ? • Soil water reaching the soil-model base through gravitational flow freely drains out • That water is no longer available for evapotranspiration even during times of water stress Miguez-Macho et al. (2007)

36. GCM Biases – new nesting approach needed

37. Signal & Uncertainty Downscaling of ECHAM4.5 AGCM Forecast for FMA 2009 RSM RAMS

38. Reliably Estimate the Signal & Uncertainty - Multi-Model Ensembling JAS Temperature Robertson et al. (2004) Benefit of Increasing Number of AGCMs in Multi-Model Combination

39. RCM Bias Sun et al. (2005)

40. RCM Bias Correction • Reduce systematic errors (MOS correction, calibration) • Minimize random errors (multi-model approach) • Suppress conditional errors (conditional exceedance probabilities recalibration)

41. Changing Climate Moncunill and Sun (2008) Os parâmetros calculados neste modelo tem um nível de significancia de 0,1% no teste t de student.

42. GHACOF MAM forecasts observations The average forecasts fail to forecast the shift towards dry conditions.

43. SUMMARY • The Regional Climate Model is a useful tool to understand climate process at sub-GCM scales. • The possibility exists to enhance information to higher spatial and temporal scales using RCMs • requires research! Results are often region and season specific. • Successful application of RCMS in climate risk management may require creativity to address users’ needs • Focusing on climate variables that are both relevant and predictable/projectable (e.g., dry spells, rainfall frequency, monsoon onset)