![]() |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Implementation of Seasonal Forecast System 4 |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
ECMWF will upgrade its operational seasonal forecasts from System 3 to System 4. The change to operational status will take place on 1 November 2011, when System 3 ceases to be operational, and System 4 becomes the new operational forecast system. This page gives a brief description of System 4 and provides technical details about access to the data during the pre-operational test phase both from MARS and through disseminationThe page will be updated as required. It was last changed on 07.11.2011. For a record of changes made to this page please refer to Document versions. A printable version Further information and advice regarding the upgrade can be obtained from
News18 October 2011
28 July 2011
8 June 2011
TimetableThe planned timetable for implementation of Seasonal Forecasting System 4 is as follows. The exact date for each step will be announced on this web page when it is confirmed.
System 3 will run up to and including February 2012 according to its present schedule (release date on the 15th of each month), but once it ceases to be operational its availability is not guaranteed. Dissemination of System 3 data will not be available; access will only be available via MARS. Technical description of System 4System 4 is based on a more recent atmospheric model version (IFS model cycle 36r4), higher resolution forecasts with a higher top of the atmosphere, more members, and a larger hindcast data set (see Table below). System 4 initial perturbations are defined (as in System 3) with a combination of atmospheric singular vectors and an ensemble of ocean analyses. Atmosphere model uncertainties are simulated using the 3-time level stochastically perturbed parameterized tendency (SPPT) scheme and the stochastic back-scatter scheme (SPBS) operational in the EPS (System 3 used only a 1-time version of SPPT). System 4 uses NEMO instead of HOPE as its ocean component (with the same resolution), with initial conditions generated by the Near Real Time (NRT) NEMOVAR suite instead of HOPE/OI. In February, May, August and November, 15 of the 51 members are extended to 13 months. Key differences between the operational System 3 and the forthcoming System 4:
A description of the technical aspects of changing from System 3 to System 4 is provided in the Technical description of Seasonal Forecasting System 4. This document should be read before the new User Guide, which gives a reference description of the forecast system and the data archive, but does not include information on the changes from System 3 to System 4. Pre-operational test data setsRe-forecasts are available for downloading from MARS, and are intended at this stage to support re-tuning of calibration procedures for real-time products. Re-forecast data are in the same streams as System 3 data, and can be retrieved with ORIGIN=ECMF and SYSTEM=4:
For the standard (up to month 7) forecasts, request METHOD=1. For the longer range forecasts (up to month 13) , request METHOD=3 (see the Technical description for details). Real-time forecast test data are now available in MARS. A full 51 member ensemble is available from May 2011 to present. An initial set of derived products (anomalies, ensemble means and hindcast climate means) have been implemented and will be made available in MARS by the end of October 2011. It is planned to introduce an additional set of new derived products (including probabilities) in the near future, but after the implementation of System 4. Please report any problems you find with this data to User Support System 4 products for disseminationInformation regarding the availability of System 4 products for dissemintation will soon be made available at: Users can already request System 4 products through the ECMWF dissemination system. The contact point for dissemination querries at ECMWF is Dragan Jokic Please note that in seasonal System 4 a validation month inside dissemination file names has been corrected to reflect the step range of products within the file. To illustrate the change, for a November System 4 run for full 7 months range and dissemination stream XX dissemination files will be:
Further reading
Document versions
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||