Doubt in ERA5 ocean wave reanalysis horizontal resolution

Hello,

Please, could anyone confirm the real ERA5 ocean wave reanalysis horizontal resolution? According to the old ERA5 data documentation:

"The ERA5 HRES (High Resolution) data has a native resolution of 0.28125 degrees (31km) [...] The wave HRES (High Resolution) data has a native resolution of 0.36 degrees [...]"

This information differs from the specified in the CDS website:

"Horizontal resolution: Reanalysis: 0.25°x0.25° (atmosphere), 0.5°x0.5° (ocean waves)"

So I'm a bit confused. I can understand the interpolation of the ERA5 HRES atmospheric fields to a regular 0.25ºx0.25º grid (0.28º is quite similar to 0.25º), but... what is the reason to interpolate the ocean wave fields to a 0.5x0.5º regular grid when theorically the real horizontal resolution is 0.36º?

Thanks in advance.
Regards,

Oriol Ciurana

Hi Oriol,

Yes, it is confusing. ERA5 as a complete dataset is hosted in ECMWF MARS. A subset of ERA5 is hosted on the CDS - all ERA5 records you can see in the CDS catalogue have data hosted on the CDS. While in most cases, reduced Gaussian grids are used in ECMWF MARS, this is not the case for the CDS. ERA5 data was interpolated to fixed regular lat/lon grids when they were put on the CDS.

Regarding the reason why certain spatial resolutions were selected, I'll check with our scientists.

Cheers,

Xiaobo

OK, thank you Xiaobo, we keep in touch.

Cheers,

Hi Oriol,

Please see comment from our scientist below:

"...we needed to make a choice that fitted well with the ERA5 atmosphere re-gridding. We chose 0.5, i.e., half the resolution of ERA5 atmosphere; since waves are at lower resolution. In hindsight we could have chosen 0.25 as well. That would then be an oversampling, though."

The decision was made by our scientists and technical experts jointly.

Regards,

Xiaobo

Hi Xiaobo,

In my opinion this was not the best decision, I would have preferred a little bit oversampling instead of having resolution loss, or simply re-gridding to the native resolution or a closer one.

Anyway, thank you for the explanation and the quick response.
Regards,

Oriol

No problem, Oriol. Thank you for your feedback.

Hi Xiaobo,

Where can we find the lat/lon coordinates of the ERA5 Wave HRES native grid? 

For the ERA5 HRES atmospheric dataset, the documentation specifies that the native grid is a reduced Gaussian grid N320. However there is no specifics on the oceanic data set besides the 0.36 deg resolution as Oriol pointed out... The closest grid I've found is under FAQ: Frequently Asked Questions#openifsgeneral , Q. How do I know the grid from from the 'T' number? : the reduced Gaussian N256 which translates to lat/lon  0.351 deg resolution (see below):














Can you please clarify?

Thanks!


Hi Guillermo,

I'll ask our scientist and then get back to you.

Kind regards,

Xiaobo

Just realised you are in contact with out scientist at ERA5 Wave HRES native grid exact coordinates.