How can I calculate the temporal "Standard deviation" with cdo?
Asked
Active
Viewed 450 times
-2
-
Welcome to SO. Have you read through the CDO user guide. It lists multiple ways to calculate standard deviation https://code.mpimet.mpg.de/projects/cdo/embedded/cdo.pdf#page128 – Robert Wilson Jul 27 '22 at 06:27
-
you talk of obs and model, are you actually wanting the root mean square error between the two? Usually standard model evaluation metrics are MAE, RMSE and anomaly correlation - perhaps you could clarify your question? – ClimateUnboxed Jul 27 '22 at 07:49
-
1@AdrianTompkins read the question as possibly about comparing variation between model and observations, but clarification would help. – Robert Wilson Jul 27 '22 at 12:11
-
@Khan can you clarify this question? If Rob's answer really solved your issue then you have a lot of superfluous info in your question that needs deleting and you should also write temporal standard deviation (i.e. not spatial). My gut feeling was you were looking for a way to compare model to obs and thus wanted a spatial metric like RMSE. – ClimateUnboxed Aug 04 '22 at 12:44
1 Answers
1
If you want the standard deviation across all time steps and for each grid cell you would do the following:
cdo timstd infile outfile

Robert Wilson
- 3,192
- 11
- 19
-
Very well explained. Thank you so much for your kindness. I (new user) understand from your question and I got my answer for your question. – Khan Jul 30 '22 at 08:28