[Thread Prev][Thread Next][Index]

problems when obtaining anomalies: Summary



Hi All,

This is a message that couldn't reach previously the list, so I insert
the solution given by Ansley which I found very useful. The original
message comes after Ansley's solution

Ansley wrote:

Here is an answer to your question - it is a matter of getting the
region in time consistent between the variables you are comparing.
I find that using the STAT command as I go along helps to see
what the variables' grids and regions are like.

Here is a similar example, using the monthly COADS data set

USE climatological_axes
CANCEL DATA climatological_axes
! USE sst_rey.nc
USE coads
SET REGION/X=10W:5W/Y=34N:38N/L=1:230
STAT sst
LET sst_climatology = sst[GT=month_reg@MOD]
STAT sst_climatology   ! Note here that it might have been

! Here you need to put the climatology data on the same time
! axis as sst, so you can compare them.

LET sst_anomaly=sst-sst_climatology[gt=sst]
STAT sst_anomaly

LET sst_std = sst_anomaly /
sst[x=10w:5w@var,y=34n:38n@var,l=1:230@var]^.5
STAT sst_std

Now, sst_std is a variable in x, y, and time and you should be able
to use the EOF functions with it.


Ansley

>Hi All,
>I'm trying to perform some EOF analysis with my dataset with this
>script:

>USE climatological_axes
>CANCEL DATA climatological_axes
>USE sst_rey.nc
>SET REGION/X=10W:5W/Y=34N:38N/L=1:230
>LET sst_climatology = sst[GT=month_reg@MOD]
>LET sst_anomaly=sst-sst_climatology
>LET sst_std = sst_anomaly /
>sst[x=10w:5w@var,y=34n:38n@var,l=1:230@var]^.5
>LET eofspace= eof_space(sst_std[x=10w:5w,y=34n:38n],0.8)
>LET eofstat = eof_stat(sst_std[X=10w:5w,Y=34n:38N],0.8)

>But when I try, i.e. to obtain a LIST (or plot):

>LIST eofspace

>**ERROR: inconsistent sizes of data regions: T axis
>        SST_CLIMATOLOGY has 10 points (L=0991:1000)
>          expression has 230 points (L=001:230)
>I'm stuck at this point because some times the script runs OK. Maybe,
as
>almost always happens, the answer is just in front of my eyes, but
today
>I'm specially blind.
>Thanks in advance

>Antonio Rodriguez
>CICEM Agua del Pino
>Huelva, Spain





---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.443 / Virus Database: 248 - Release Date: 10/01/03



[Thread Prev][Thread Next][Index]

Dept of Commerce / NOAA / OAR / PMEL / TMAP

Contact Us | Privacy Policy | Disclaimer | Accessibility Statement