This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
WMO CBS operational collection/display of seasonal (and sub-seasonal) forecastsRichard Graham, Met Office Hadley Centre. With acknowledgements to all GPCs, LC-LRFMME, LC-SVSLRF and members of the WMO Expert Team on Extended and Long-range Forecasting (ET-ELRF)
WMO/THORPEX/WCRP sub-seasonal to seasonal implementation planning meeting, WMO HQ, Geneva,
Seasonal AND monthly products displayed on LC-LRFMME website
Dec Jan Feb
DJFGPC WashingtonEnsemble mean anomalies: pmsl(2011)
GPC products from LC-LRFMME website:DJF 2010/11: ensemble mean pmsl anomaly
Exeter ECMWF Toulouse
Beijing
Washington
TokyoMelbourne
Seoul Montreal
Coupled systems
Un-coupled systems
Pretoria
‘Dynamic’ user-selectable domain
Same available for individual months
Multi-model GPC products from LC-LRFMME website: DJF 2010/11, ensemble mean pmsl anomaly
Ensemble mean of 10 GPCs
Ensemble mean of 7 GPCs (coupled systems)
Ensemble mean of 3 GPCs (un-coupled systems)
multi-model subsets produced interactively from LC-LRFMME website
Presented at South East Europe Regional Climate Outlook Forum (SEECOF), 22-26 November 2010
2mT, precipitation, T850, Z500, SST and for individual calendar month periods
Nino3.4 plumes
GPC model consistency plots:Individual months:number of models with +ve /-ve anomalies
Z500 T850 PMSL
Precip 2mT SST
Dec 2011
GPC model consistency plots:Individual months:number of models with +ve /-ve anomalies
Z500 T850 PMSL
Precip 2mT SST
DJF 2011
Probabilistic products with subset of GPC providing hindcasts
Dec 2011
Probabilistic products with subset of GPC providing hindcasts
DJF 2011
Use at RCOFs: WMO Lead Centre information and final GHACOF consensus, SOND 2010
LC-LRFMME Model consistency
Statistical models + Forecaster judgement
Final consensus
Verification
Observed SON anomalies
25
40
35
35
25
40
Summary – global LRF
• 12 WMO designated GPCs supply seasonal and ‘subseasonal‘ data
• 2 WMO Lead centres: dealing with standardised display/multi-modelling and forecast validation
• Deterministic products and probabilistic products
• In active use by RCCs, RCOFs, NMHSs (~123 registered users, 49 countries – as of early 2011)
• Probability products will form basis of prediction component of the WMO Global Seasonal Climate Update (GSCU).
• Key plans (guided by the Expert Team for Extended and Long-range Forecasts) – in line with GFCS vision - include:
• verification of multi-model products
• Investigation/development of extended (monthly) range capability
• investigation/development of multiannual–to-decadal capability (decadal GPCs?)
CSRP: consultation in Africa
0123456789
No
. o
f vo
tes Rank 1
Rank 2
Rank 3
Rank 4
Type and number of organisations interviewed Ranking of priorities
Questionnaire fielded to 9 climate service providers
Total of 52 interviews across 8 African countries
Science component (3): Predicting onset timingbased on local time of arrival of 20% of long-term seasonal average
Average southward progression of rains with ITZC
observed
modelled
Skill of Met Office seasonal forecasts of ‘onset’ timing
early onset
late onset
orange/red =
‘good’ skill
Encouraging first results: trial onset forecasts have been provided to Regional Climate Outlook Forums ICPAC, ACMAD and SADC-DMC
Example:East Africa short-rains (OND)
Michael Vellinga
Predicting onset timing: example Greater Horn of Africa short-rains season 2011 (OND) – prediction from Augustbased on local time of arrival of 20% of long-term seasonal average
Probability of early arrival
Probability of late arrival
Observed time of arrival
Early thoughts on LC-LRFMME operational exchange
Strawman proposal for operational exchange:
• Coordinate research and operational exchanges with operational exchange a subset of research (delayed release) exchange?
• 2-weekly issues: first, with the release of LC-LRFMME seasonal forecast; second, 2 weeks later;
• Daily data to ~32 days; small number of essential variables;
• Products to be period means;
7 GPCs responded so far:
• 5 basically ‘yes’; 1 ‘too early’: 1 system not operational;
• Key issue: coordinating release date to get usefully short lead time, different GPCs have different release dates; should release date be fixed day of month or day of week?
• Some with data policy issues and other reservations (e.g. Timing of products)