You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When computing a seasonal mean from monthly data (for example), xarray and xclim will not take the length of each month into account.
Potential Solution
Something somewhere to "resample" monthly data to a coarser timescale with month-length weights.
Additional context
I'm not sure where this should be.
I noted "climatological_mean" on my sheet, but isn't that function meant for interannual means? Here, the seasonal resampling is a step that comes before the interannual mean.
In #80, we noted that extract_dataset was not meant to perform resamples coarser then daily. However, it wouldn't be hard to add a provision in the resampling function to take the period-length into account with periods coarser that a week.
Contribution
I would be willing/able to open a Pull Request to contribute this feature.
The text was updated successfully, but these errors were encountered:
En effet, climatological_mean veut garder la même fréquence et faire une moyenne sur les années. Dans un worflow SSC, le resampling est fait dans le calcul d'indicateurs (ce qui ne fonctionne pas pour les données mensuelles de base de SAC).
La question que je me pose est: est-ce vraiment encore de l'extraction ? Est-ce que c'est la même chose que la création d'une variable daily à partir de subdaily? (peut-être que oui, je ne suis pas certaine)
Si c'est plus une étape de processing, est-ce qu'on voudrait une fonction de resampling plus modulable dans aggregate.py?
Donc, ça ne va clairement pas dans climatological_mean, mais ce n'est pas super évident ça va où pour moi.
Addressing a Problem?
When computing a seasonal mean from monthly data (for example), xarray and xclim will not take the length of each month into account.
Potential Solution
Something somewhere to "resample" monthly data to a coarser timescale with month-length weights.
Additional context
I'm not sure where this should be.
I noted "climatological_mean" on my sheet, but isn't that function meant for interannual means? Here, the seasonal resampling is a step that comes before the interannual mean.
In #80, we noted that
extract_dataset
was not meant to perform resamples coarser then daily. However, it wouldn't be hard to add a provision in the resampling function to take the period-length into account with periods coarser that a week.Contribution
The text was updated successfully, but these errors were encountered: