For eg. I have a netcdf file which is in the form of a time-series and the other data is a gridded .nc file for geopotential height. I want to correlate both of these datasets.
Please guide me through this.
Thank you!
No, it can't be done like that.
you have to do nanmean of multidimensional data to a 1D array and then only correlate with other 1D data
Have you tried using the UK Met office Iris package on python, this may help solve your problem, the very basic concept it uses is loading each NetCDF files and cubes and correlating the two cubes using the code :
iris_corr=iris.analysis.stats.pearsonr(netcdf1, netcdf2)
print(iris_corr.data) .
Why don't you check their documentation
https://scitools.org.uk/iris/docs/latest/userguide/cube_maths.html
A non-parametric form of correlation should do it, in principle, assuming correct "levels of measurement" are identified.
- Badges
- Science topic
- Similar topics
- Filing