A lot of Earth Science data - particularly that related to climate and climate change - comes in daily resolution but using a day_360 calendar (such as defined in the CF-convention version 1.9). For instance, CORDEX data is typically produced like that.
Is there a way to make Climpact work directly with day_360 data (i.e. allow a date like 2022-02-30, and 2022-05-30 is followed by 2022-06-01)?
If not, is there any interest to make this a feature request for the next version? I presume that this impacts primarily the quality control checks (qualityControlCheck() in services) and the generation of the time series of data. The climdex.pcic.ncdf already seems to allow it (not tested) as it is using PCICt date class (although that funks up lengths of the month into what I would call pseudo-Julian). All indices are computed on a monthly or coarser basis so there should be no problem there.
A lot of Earth Science data - particularly that related to climate and climate change - comes in daily resolution but using a day_360 calendar (such as defined in the CF-convention version 1.9). For instance, CORDEX data is typically produced like that.
Is there a way to make Climpact work directly with day_360 data (i.e. allow a date like 2022-02-30, and 2022-05-30 is followed by 2022-06-01)?
If not, is there any interest to make this a feature request for the next version? I presume that this impacts primarily the quality control checks (
qualityControlCheck()inservices) and the generation of the time series of data. Theclimdex.pcic.ncdfalready seems to allow it (not tested) as it is usingPCICtdate class (although that funks up lengths of the month into what I would call pseudo-Julian). All indices are computed on a monthly or coarser basis so there should be no problem there.