Hi - in the past I've downloaded C3S seasonal forecast data using this type of API script below.
I'd played with the "grid" option before now to download the data to different interpolations. I wanted to double check with those in the know - is the 'F320' setting the native resolution of the C3S model - and if not, what setting do I need to ensure the data I'm downloading isn't interpolated to an alternative grid. Am looking at extremes so am keen to keep the output as close to the resolution of the model as possible whereby any interpolation could potentially remove any sharp peaks/extremes.
Many thanks, Richard richard@catinsight.co.uk
c = cdsapi.Client()
c.retrieve(
'seasonal-original-single-levels',
{
'originating_centre': 'ecmwf',
'system': '5',
'variable': 'total_precipitation',
'year': [
'1991',
'2000',
],
'month': '09',
'day': '01',
'leadtime_hour': [
'24', '48', '72',
'5136', '5160',
],
'area': [
70, -30, 30,
30,
],
'format': 'grib', 'grid' : 'F320',
},
'/home/19912000rain.grib')
1 Comment
Eduardo Penabad
Hi Richard,
the data available from the CDS is not the complete data offering from each one of the data providers. In all cases what's available is a subset of the variables provided on a 1x1-degree grid, meaning that's the finest resolution available for each one of the forecasting systems.
You can still use the interpolation features offered within the CDS API (built in ECMWF MARS archive where the data is kept and from where it gets delivered to the CDS) as you have already tested, but the starting point for those interpolations will always be the 'original' C3S data which corresponds to the 1-degree grid I mentioned above.
I hope that answers your question.
Best regards,
Eduardo Penabad
C3S Climate Predictions and Projections Team