-
Notifications
You must be signed in to change notification settings - Fork 40
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Work on BGC from 2024 LOV visit #356
base: master
Are you sure you want to change the base?
Conversation
Hi folks, @catsch @Sauzede @grgdll @HCBScienceProducts, Could I have some advice with regard to how can I easily identify all variables related to Radiometry in the Erddap list of variables below (the BGC dataset filling the Erddap is from synthetic profiles). I would need this in order to filter out radiometry variables not in R data mode for the "Standard user mode", following recommendation in this issue #280 (comment) List of variables to filterList of erddap BGC dataset from synthetic profile files:
|
All variables that contain "up_radiance", "downwelling_par" or "down_irradiance" are radiometry-related. "up_radiance" and "down_irradiance" are followed by 3 numerical digits that give the associated wavelength in nm, if you want to refine your filter.
|
awesome ! thanks @HCBScienceProducts ! |
- also fix a bug in pyarrow search_params
- fix deprec warning
- sort private _minimal_vlist property
one more question @catsch @Sauzede @grgdll @HCBScienceProducts : what about CDOM ? |
We are presently waiting for new calibration factors for CDOM or a list of QC=4 from Seabird |
thanks @catsch ! |
do you think we can keep Radiometry in the 'research' user mode ? |
Most of the time, Radiometry is good data (in R mode) |
do you think we can keep Radiometry in the 'research' user mode ?
Yes to radiometry both in 'research' mode with selection like all other variables (only D-mode, QC flags '1', '5', or '8').
On radiometry in 'standard' mode ***@***.***):
Should this rather be R-mode or A-mode or D-mode with QC flags '1', '2', '5', or '8' (instead of R-mode only). I know there's been some DMQC on radiometry done, i.e., D-mode exist (and are better data than the R-mode data).
On CDOM:
No separate treatment on the argopy side should be needed for the 'standard' mode but can be treated like all other variables (no R-mode but only A-mode or D-mode, QC flags '1', '2', '5', or '8').
The CDOM issues are to be resolved on the Argo data management side. Once solved, it may eventually transition from today's "only R-mode and better not be used" towards A-mode or D-mode and can then appear just like any other variable. Same for the 'research mode': No separate treatment of CDOM (or radiometry) needed.
|
- fix bug with shortcuts in ArgoIndex file and convention - the ArgoIndex now support the "etc/argo-index/argo_aux-profile_index.txt" index file
- start support for more user mode in BGC
- don't remember why these variables were included, not consistant ...
- based on new utility merge_param_with_param_adjusted
- fix bug in to_index data fetcher method whereby the index could differ from the dataset
This assume that this branch will be merged for the v1.0.0
- fix bug whereby dataset without PSAL would raise errors with filters assuming all core parameters
- and refactored argo_accessor into _argo
- fix bug whereby user mode was not passed to downstream instances when URI was split
- list_standard_variables now return list for phy or bgc dataset - improve consistency of the "variable" vs "parameter" terms usage - fix bug whereby transforming the dataset according to data mode was not allowing delayed mode only and adjusted only variables in a dataset - improve docstrings - limit use of explicit list of parameters, fall back on using the utility list functions
- describe BGC support for expert and research user modes
- fix bug whereby pressure constraint was on pres_adjusted rather than pres for researcu user mode erddap uri
- fix bugs whereby list of parameters was not created when the list of params was in the args
/home/runner/micromamba/envs/argopy-tests/lib/python3.10/site-packages/fsspec/registry.py:279: UserWarning: Your installed version of s3fs is very old and known to cause severe performance issues, see also dask/dask#10276
Place holder for work kicked off during 2024 LOV visit.
Closes #361
Adressed #363
What's in this PR:
transform_data_mode
, that is doing what was doing thefilter_data_mode
method before, but this method was very badly named: merge PARAM with PARAM_ADJUSTED according to DATA_MODE valuesfilter_data_mode
is now really doing what is was supposed to do: filter points according to data mode valuespostprocessing
method in the data fetcher facade, so that it clearly implement user mode requirements for 'phy' and 'bgc' datasetauxiliary
index in the ``etc` GDAC folderPoke @Sauzede and @catsch
Breaking changes in the API:
history
is replaced byProcessing_history
filter_data_mode
filter_data_mode
astransform_data_mode