Open jiarong-wu opened 11 months ago
Thanks for adding this issue @jiarong-wu! I am working on the recipe in #57, could you clarify the following points?
License
This points to several licenses. Do you know exactly which one is used for the WW3 product?
https://www.umr-lops.fr/en/Donnees/Vagues/sextant#/metadata/857a3337-f59a-481a-bf98-5561e8b61e7b
This link does not seem to work for me, could you double check on those?
After double checking, the link to the documentation should work (https://www.umr-lops.fr/en/Donnees/Vagues/sextant#/metadata/857a3337-f59a-481a-bf98-5561e8b61e7b) and the license is Attribution-ShareAlike CC BY-SA.
Hi @jiarong-wu I have restarted these efforts over https://github.com/leap-stc/wavewatch3_feedstock. This dataset is still very much posing a problem, but I am chipping away at it. Sorry for the long wait.
Dataset Name
Ifremer WW3 global wave hindcasts
Dataset URL
https://data-dataref.ifremer.fr/ww3/GLOBMULTI_ERA5_GLOBCUR_01/GLOB-30M/
Description
This data set is a 20 year global ocean surface wave hindcast provided by Ifremer (French Research Institute for Exploitation of the Sea) and the documentation is here: https://www.umr-lops.fr/en/Donnees/Vagues/sextant#/metadata/857a3337-f59a-481a-bf98-5561e8b61e7b
Size
About 2.5GB per file 12 months 20 years.
License
https://creativecommons.org/licenses/?lang=en
Data Format
NetCDF
Data Format (other)
No response
Access protocol
HTTP(S)
Source File Organization
There is one subfolder for each year under
/ww3/GLOBMULTI_ERA5_GLOBCUR_01/GLOB-30M/
. Within the subfolder, the files of interest are located inFIELD_NC/
, with each file containing data of one month. For example,/ww3/GLOBMULTI_ERA5_GLOBCUR_01/GLOB-30M/2022/FIELD_NC/LOPS_WW3-GLOB-30M_202201.nc
is the 2022 January data.Example URLs
Authorization
No; data are fully public
Transformation / Processing
I'm not sure... maybe lumping them all together (concatenated along the time axis)? If that is not too big to manipulate.
Target Format
Zarr
Comments
I was provided the url to the data set by the authors directly. The data set seems fully public without need for authentication but I'm not sure if downloading in large quantity is fast enough or if it will trigger some warning. (I'm not sure about the Access protocol option.)