Welcome to the new DiscussCESM forum!
We are still working on the website migration, so you may experience downtime during this process.

Existing users, please reset your password before logging in here: https://xenforo.cgd.ucar.edu/cesm/index.php?lost-password/

About pdo.ncl

Hi,I have question about pdo.ncl.In line 156, why the sst data input to the eofunc_ts function wasn't wighted?I am a beginner of EOF.I look forward to your reply. Thank you!Emily                                                                                
 

asphilli

Adam Phillips
CVCWG Liaison
Staff member
Hi Emily,There has been discussion in our group in the past about whether it is correct to use the area-weighted SST anomaly array for the EOF timeseries creation or simply the SST anomaly array. We concluded based on informal testing that it does not matter that much either way, and thus we do not have a standard that we use. (Note that the area-weighted SST anomaly array should always be used when calculating the EOF patterns though.) You are free to test out the differences in the EOF timeseries calculation yourself. Simply change lines 155-156 in pdo.ncl from this:delete(sst_CW)pcts = eofunc_ts(sst2({20:70},{110:260},:),evecv,False)to this:pcts = eofunc_ts(sst_CW({lat|20:70},{lon|110:260},time|:),evecv,False) delete(sst_CW)Best,Adam
 
Top