When downloading monthly surface data for 2024 (or any other year) in GRIB format, the date format is incorrectly structured. It should start at 2024-01-01-00, but the download says it starts at 2023-12-31-18. This suggests that the dates are not being read correctly at the time of download. Could you please advise if there is a problem with the Copernicus download platform?
Here is what was downloaded:
RefTime = 2023-12-31 18:00:00 Units = hours Calendar = proleptic_gregorian
YYYY-MM-DD hh:mm:ss YYYY-MM-DD hh:mm:ss YYYY-MM-DD hh:mm:ss YYYY-MM-DD hh:mm:ss
2023-12-31 18:00:00 2024-01-31 18:00:00 2024-02-29 18:00:00 2024-03-31 18:00:00
2024-04-30 18:00:00 2024-05-31 18:00:00 2024-06-30 18:00:00
Hi. You guys are great for making these data available. Thank you very much!
I’m having a quick bo-peep at monthly averaged ERA5 data as well. And I’m comparing what I’m downloading here with those that I’ve been downloading from the old CDS.
I understand that netcdf formatting is “experimental”, but I’ve noticed some fundamental differences between CDS(original) and CDS(beta) that have made it difficult to reuse the same scripts. So, for example, looking at monthly averaged atmospheric temperature…
In CDS(original) there is a variable called “time” and this is hours since 1/1/1900.
In CDS(beta) there is no variable called “time” Here the variable is called “date” and is a completely different format to that of “time” within CDS(original).
In CDS(original) the temperature field is in kelvin, with a scale factor and offset.
In CDS(beta) the temperature field is in kelvin, but with no scale factor and no offset.
I hope this makes sense. I’m happy to clarify if not.
Again, thanks greatly for making the data available. It’s invaluable.
All the best from Hobart