[ncl-talk] bug found when using the EOF to reconstruct the data

Xia, Geng gxia at albany.edu
Tue Nov 11 12:14:38 MST 2014


??

Hi all


Recently, I try to do the EOF analysis using ncl for my research topic.  However, I find that after I did the EOF analysis (using function eofunc and eofunc_ts), I can't reconstruct the result back to the original dataset (using function eof2data). The reconstructed data (including all EOFs and PCs) I got differs quiet a bit from the original data set.


I have tried different function such as eofunc_Wrap but the result does not get better.


I have attached the sample ncl script I use and the data set in the email. Hopefully you can take a look.

I have commend on the steps I did so It should be very straight forward.


my ncl version is 6.1.2

the version of Linux is  x86_64 x86_64 x86_64 GNU/Linux?


Hope you can take a look at my problem

Best


Geng Xia


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20141111/79f869ea/attachment-0001.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: gxia_test.bin
Type: application/octet-stream
Size: 688128 bytes
Desc: gxia_test.bin
Url : http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20141111/79f869ea/attachment-0001.bin 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: eof_test.ncl
Type: application/octet-stream
Size: 2370 bytes
Desc: eof_test.ncl
Url : http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20141111/79f869ea/attachment-0001.obj 


More information about the ncl-talk mailing list