Caio A. S. Coelho

1
EUROBRISA A EURO-BRazilian Initiative for
improving South American seasonal forecasts

• Caio A. S. Coelho
• Centro de Previsão de Tempo e Estudos Climáticos
  (CPTEC)
• Instituto Nacional de Pesquisas Espaciais (INPE)
• caio_at_cptec.inpe.br

• PLAN OF TALK
• History
• Aims
• Planned activities
• Motivating results
• Summary

CPTEC-IRI Workshop , Cachoeira Paulista (Brazil),
8 November 2006
2
History of EUROBRISA

• 2001-2005 PhD work on forecast calibration and
  combination (Coelho 2005)
• Developed conceptual framework for forecasting
• (Bayesian approach named forecast assimilation)
• Nino index (Coelho et al. 2003, 2004)
• Equatorial Pacific SST
• (Stephenson et al. 2005)
• South American rainfall
• (Coelho et al. 2006a)
• Regional rainfall and river flow downscaling
• (Coelho et al. 2006b)
• 2005 Preparation, submission and approval of
  EUROBRISA proposal by ECMWF council
• 2005/2006 Preparation, submission and approval
  of young investigator fellowship by FAPESP and
  start of EUROBRISA

3
The EUROBRISA Projectkey Idea To improve
seasonal forecasts in S. Americaa region where
there is seasonal forecast skill and useful
value.
http//www.cptec.inpe.br/caio/EUROBRISA/index.htm
l
Involved institutions Country Partners
CPTEC Brazil Coelho, Cavalcanti, Costa Silva Dias, Pezzi
ECMWF EU Anderson, Balmaseda, Doblas-Reyes, Stockdale
INMET Brazil Moura, Silveira, Lucio
Met Office UK Graham, Davey, Colman
Météo France France Déqué
UFPR Brazil Guetter
Uni. of Reading UK Stephenson, Challinor
Uni. of Sao Paulo Brazil Ambrizzi, Silva Dias

• Aims
• Strengthen collaboration and promote exchange of
  expertise and information between European and S.
  American seasonal forecasters
• Produce improved well-calibrated real-time
  probabilistic seasonal forecasts for South
  America
• Develop real-time forecast products for
  non-profitable governmental use (e.g. reservoir
  management, hydropower production, and
  agriculture)

Affiliated institutions
CIIFEN Ecuador Camacho
IRI USA Baethgen
UFRGS Brazil Bergamaschi
4
Planned activities
Climate prediction research and development

• Produce probabilistic forecasts of precip. and
  temp. with empirical and dynamical coupled models
  
• Deliver objectively combined (dynamical
  empirical) well-calibrated forecasts
• Compare skill of empirical, dynamical and
  combined forecasts using deterministic and
  probabilistic measures
• Dynamical and statistical downscaling
• Seasonal predictability studies

Impacts (collaborative work with users)

• Hydrology Downscaling of seasonal forecasts for
  river flow predictions and use in hydrological
  models
• Agriculture Research on the use of seasonal
  forecasts in agricultural activities Downscaling
  of seasonal forecasts for use in crop models

5
EUROBRISA multi-model ensemble system

• 4 coupled global circulation models 1 empirical
  model

Coupled Model Country Hindcast period
CPTEC Brazil 1982-2001
ECMWF International 1987-2001
Meteo-France France 1993-2001
UKMO U.K. 1987-2005
Empirical model Predictor Atlantic and Pacific
SST Predictands Precipitation and temperature
6
Correlation maps DJF rainfall anomalies
DEMETER Multi-model ()
Empirical
Integrated
ECMWF, Meteo-France, UKMO (1959-2001), I.C.
November

• Comparable level of determinist skill
• Better skill in tropical and southeastern South
  America

7
Mean Anomaly Correlation Coefficient
Empirical
Multi-model
Integrated
Most skill in ENSO years and forecast
assimilation can improve skill
8
Brier Skill Score for DJF rainfall
ENS
Multi-model
Integrated
Empirical
Forecast assimilation improved Brier Skill Score
(BSS) in the tropics
9
Brier Score decomposition
reliability
uncertainty
resolution
10
Reliability component of the BSS
Multi-model
Integrated
Empirical
Forecast assimilation improved reliability in
many regions
11
Resolution component of the BSS
Integrated
Multi-model
Empirical
Forecast assimilation improved resolution in the
tropics
12
Regional rainfall downscaling
Multi-model ensemble 3 DEMETER coupled
models ECMWF, Meteo-France, UKMO 3-month
lead Start Aug NDJ Period 1959-2001
13
South box NDJ rainfall anomaly
Multi-model
- - - Observation
Forecast
Forecast Correlation Brier Score
Multi-model 0.57 0.22
FA 0.74 0.17
Forecast assimilation
(Coelho et al. 2006b)

• Forecast assimilation improves skill substantially

14
North box NDJ rainfall anomaly
Multi-model
- - - Observation
Forecast
Forecast Correlation Brier Score
Multi-model 0.62 0.21
FA 0.63 0.18
Forecast assimilation
(Coelho et al. 2006b)

• Forecast assimilation improved skill marginally

15
River flow predictions (NDJ)
Annual cycle
Forecast Correlation Brier Score
Parana 0.16 0.25
Tocantins 0.29 0.22
(Coelho et al. 2006b)

• Harder to downscale river flow than rainfall

16
Agricultural application
(Challinor et al. 2004)
17
EUROBRISA summary

• Challenging initiative for improving the quality
  of South American seasonal forecasts
• Facilitate exchange and transfer of technology,
  knowledge and expertise between participating
  institutions
• Valuable opportunity to
• - develop an objectively integrated
• (i.e. dynamical empirical) forecasting
  system for
• South America
• - work closely with end-users to evaluate our
  forecasting system in terms of user variables
  rather than solely on
• traditional climate variables
• Collaborative activities with IRI are of great
  interest

18
References

• Coelho C.A.S., S. Pezzulli, M. Balmaseda, F. J.
  Doblas-Reyes and D. B. Stephenson, 2003
• Skill of Coupled Model Seasonal Forecasts A
  Bayesian Assessment of
• ECMWF ENSO Forecasts. ECMWF Technical Memorandum
  No. 426, 16pp.
• Coelho C.A.S., S. Pezzulli, M. Balmaseda, F. J.
  Doblas-Reyes and D. B. Stephenson, 2004
• Forecast Calibration and Combination A Simple
  Bayesian Approach for ENSO.
• J. Climate, 17, 1504-1516.
• Coelho C.A.S. 2005 Forecast Calibration and
  Combination Bayesian Assimilation of Seasonal
  Climate
• Predictions. PhD Thesis. University of Reading.
  178 pp.
• Coelho C.A.S., D. B. Stephenson, M. Balmaseda,
  F. J. Doblas-Reyes and G. J. van Oldenborgh,
  2006a
• Towards an integrated seasonal forecasting system
  for South America. J. Climate , 19, 3704-3721.
• Coelho C.A.S., D. B. Stephenson, F. J.
  Doblas-Reyes, M. Balmaseda, A. Guetter and G. J.
  van
• Oldenborgh, 2006b A Bayesian approach for
  multi-model downscaling Seasonal forecasting of
  regional
• rainfall and river flows in South America.
  Meteorological Applications, 13, 73-82.
• Stephenson, D. B., Coelho, C. A. S.,
  Doblas-Reyes, F.J. and Balmaseda, M., 2005
• Forecast Assimilation A Unified Framework for
  the Combination of
• Multi-Model Weather and Climate Predictions.
  Tellus A, Vol. 57, 253-264.
• Available from http//www.cptec.inpe.br/caio