Hello
I am having a high-resolution daily data (~20 gb) with 0.25 deg horizontal resolution from 1998-2020. I am trying to compute the daily anomalies and I tried python, Ferret, NCL and CDO.
in Python using xarray it produced the output but it does not have anomalies for 29-Feb. Then I tried Ferret but it does not correctly account for the daily climatology because it subtracted 1-March from 29-Feb from a non-leap year same problem persists for the CDO.
At last I tried NCL but because of heavy data the NCL operation got killed due to machine limitation.
Is anyone having a code or process to do the daily anomalies correctly for heavy multi-year data containing leap years?
Cheers, Saurabh
If you use CDO, it will be simple and efficient. You just need to take two steps:
# calculate multi-year daily mean
a. cdo ydaymean infile outfile1
# calculate multi-year daily anomaly
b. cdo sub infile outfile1 outfile2
infile in the two steps is the same and it is your origin file. outfile2 is what you want.
Actually, if you do not want the outfile1, you can do it in one step by combine step a and step b.
cdo -L -sub infile -ydaymean infile outfile2.
You will get the same outfile2 as you obtain from the first method.
In cdo, you can calculate
cdo -ydaysub infile -ydaymean infile outfile
Hope this helps.
- Badges
- Science topic