compare 4 timeseries

Eric Altshuler ela at COLA.IGES.ORG
Fri Aug 3 16:56:12 EDT 2007


Bernd,

The 'min' function does not work this way. It finds the minimum value of a variable over a specified dimension, e.g. x, y, z or t. It cannot be used to find the minimum value among a set of different variables. You could find the minimum value by brute force (direct comparison) using the grads scripting language, but you might be better off using a fortran or c program to do this.

Also, your dimension environment looks crazy. In the year 25655, the Gregorian calendar will have lost all accuracy, and human beings will have died out long before, along with most other life on earth.

Best regards,


Eric L. Altshuler
Assistant Research Scientist
Center for Ocean-Land-Atmosphere Studies
4041 Powder Mill Road, Suite 302
Calverton, MD 20705-3106
USA

E-mail: ela at cola.iges.org
Phone: (301) 902-1257
Fax: (301) 595-9793

----- Original Message -----
From: "bernd becker" <bernd.becker at METOFFICE.GOV.UK>
To: GRADSUSR at LIST.CINECA.IT
Sent: Friday, August 3, 2007 11:17:00 AM (GMT-0500) America/New_York
Subject: compare 4 timeseries

hello!


I want to pick the lowest value at each time step
from 4 time series.

 how do I get the following to work?

define t=tloop(min(tmean.62,tmean.63,tmean.64,tmean.66))

Syntax Error:  Invalid dimension expression
  Expecting X/Y/Z/T/LAT/LON/LEV/TIME, found tmean
Error from MIN:  1st dimension expression invalid
Operation Error:  Error from min function
Operation Error:  Error from tloop function
  Error ocurred at column 1
DEFINE error:  Invalid expression.
ga-> q dims
Default file number is: 2
X is fixed     Lon = -97  X = -96
Y is fixed     Lat = 33  Y = 124
Z is fixed     Lev = 1  Z = 1
T is varying   Time = 00Z01JAN25655 to 00Z22DEC25655  T = 1 to 356


cheers,
Bernd.
--
Bernd Becker   The Monthly Outlook
Met Office  FitzRoy Road  Exeter   Devon EX1 3PB  United Kingdom
Tel.: +44 (0) 1392 884511 Fax: +44 (0)870 900 5050
E-mail:bernd.becker at metoffice.gov.uk -  http://www.metoffice.gov.uk



More information about the gradsusr mailing list