19/July/2024
C3S graphical and data products from July 2024 forecast
The issue reported on 10 July 2024 (below), related to errors in the July-start forecast data from ECCC's new operation system, CanSIPSv3.0, has now been resolved. Corrected data has been acquired, processed and archived, and ECCC graphical and data products have been published. Data with July starts from ECCC's systems 4 and 5 (in the CDS notation) can now be downloaded from the CDS. Any data from these systems downloaded from ECMWF's MARS archive before 19 July 2024 may be flawed; data downloaded after this date is expected to be error-free.
Unfortunately, hindcast data with August start from these new systems have not yet been made available (in line with the usual practice of publishing hindcasts at least one month before the corresponding real-time forecasts). This data will be processed and published, at the latest at the same time as the August 2024 forecast. The advance publication of hindcast data will be resumed as soon as possible.
17/July/2024
C3S graphical and data products from July 2024 forecast
The corrected DWD July-initialised real-time forecast has now been published, both in the set of graphical products on the website (https://climate.copernicus.eu/charts/packages/c3s_seasonal/) and in the respective datasets of the CDS. Data downloaded from the CDS after 16 July 2024 13 UTC is expected to be error-free. The original error only affected one member of the real-time DWD forecast ensemble; however, all ensemble members have now be re-archived.
10/July/2024
C3S graphical and data products form from July 2024 forecast
Due to potential data quality issues, the suite of graphical seasonal forecast products with July 2024 initial conditions (available at https://climate.copernicus.eu/charts/packages/c3s_seasonal/) do not include the contribution from DWD, at the time of publication. (They also do not include ECCC forecasts - see announcement below.) The DWD July real-time forecast data has been processed and archived, and it is downloadable from the CDS forms and API, but we do not endorse its use until further investigation is completed. Initial examination points to the 22nd member of the ensemble (in the order archived) having been started from wrong initial conditions; at present, there is no guarantee that the other members are 'correct'.
...