pepijn-devries / CopernicusMarine

Subset and download marine data from EU Copernicus Marine Service Information. Import data on the oceans physical and biogeochemical state from Copernicus into R without the need of external software.
https://pepijn-devries.github.io/CopernicusMarine/
GNU General Public License v3.0
25 stars 3 forks source link

Some products are not visible using CopernicusMarine #39

Open DevrekerD opened 9 months ago

DevrekerD commented 9 months ago

Hello, i'm really sorry to bother you again, but i'm still having problems downloading some Copernicus Marins products using CopernicusMarine. Your last exemple still works fine but when i want other products it fails :

details <- cms_product_details("OCEANCOLOUR_ATL_BGC_L4_MY_009_118", variant = "detailed-v3") names(details$layers)

give some results, like the one I'm interested in: "cmems_obs-oc_atl_bgc-plankton_my_l4-multi-1km_P1M_202311/CHL" (which is different from the layer name on the Copernicus website !!!)

However the command : cms_product_details("OCEANCOLOUR_ATL_BGC_L4_MY_009_118", "cmems_obs-oc_atl_bgc-plankton_my_l4-multi-1km_P1M_202311","CHL") give NULL results !

Is that a problem that only concern Copernicus or does your script need updating ?

Thank you.

pepijn-devries commented 9 months ago

Hi @DevrekerD ,

The reason that your call results in NULL is because you have not specified the variant parameter and the empty default of "" is used (instead of the newer "detailed-v3", if you use the latter you will get info, see below). It seems that Copernicus Marine has organised this differently over time. I'm currently in touch with the people at Copernicus Marine, to see how I can best address the implementation of the new services. I do agree that the R package should be more clear about this (either update the interface or expand the documentation). I want to await the input from Copernicus Marine before I proceed with this. Until then, I will leave this issue open as a reminder to myself.

cms_product_details("OCEANCOLOUR_ATL_BGC_L4_MY_009_118",
                    "cmems_obs-oc_atl_bgc-plankton_my_l4-multi-1km_P1M_202311",
                    "CHL",
                    variant = "detailed-v3")
DevrekerD commented 8 months ago

Hello,

it seems that the problem persists. However, when you look at the CoperncusMarine site and prepare a data extraction subset, you can see an "Automate" created automatically by the site (in Phtyon I imagine). For example:

import copernicusmarine as cm

cm.subset( dataset_id="cmems_obs-oc_atl_bgc-plankton_my_l4-gapfree-multi-1km_P1D", dataset_version="202311", variables=["CHL"], minimum_longitude=0.8736138570126676, maximum_longitude=3.2025735314012214, minimum_latitude=50.05281724428351, maximum_latitude=51.96904988776776, start_datetime="1997-09-04T00:00:00", end_datetime="2024-03-14T00:00:00", )

I think the commande line "dataset_version" is missing in your CopernicusMarine R package ? Do you think you can update the code ?

David

pepijn-devries commented 8 months ago

Hello,

it seems that the problem persists. However, when you look at the CoperncusMarine site and prepare a data extraction subset, you can see an "Automate" created automatically by the site (in Phtyon I imagine). For example:

import copernicusmarine as cm

cm.subset( dataset_id="cmems_obs-oc_atl_bgc-plankton_my_l4-gapfree-multi-1km_P1D", dataset_version="202311", variables=["CHL"], minimum_longitude=0.8736138570126676, maximum_longitude=3.2025735314012214, minimum_latitude=50.05281724428351, maximum_latitude=51.96904988776776, start_datetime="1997-09-04T00:00:00", end_datetime="2024-03-14T00:00:00", )

I think the commande line "dataset_version" is missing in your CopernicusMarine R package ? Do you think you can update the code ?

David

Thank you for this report. Note that the R package is developed apart from the Python package and I have tried to avoid any dependency in Python, in order to simplify the installation procedure. In order to achieve this, I had to use a different strategy for approaching the data at CopernicusMarine. Unfortunately, this also means I can't just 'copy' features from the Python package. This might be something I am able to attend to this in the future, but I don't expect to have a quick fix anytime soon... Thank you for your patience...