Re-Gridding/Variance

[Steve Cocks <scocks@xxxxxxxxxxxxxxxxxx>]

Hello,

I am having trouble computing correlations for data-sets with different resolutions, different time spans and different gridding points.  In particular,
I have

a)  SST Data      -  x=0E:0E,  y=40S:40N,   # x points:1,1440;  #y points:1,320
                               time steps: 1 July 99 to 31 Jul 01

b)Scat. Data- x=0.5W:0.5W, y=85.5s:85.5n,# x points:1,171

        #y points:1,171, time steps: 20 Jul 99 to 04 Nov 01

c)Precip dat- x=0E:0E, y=90s:90n, # x points: 1:360,

        # y points: 1, 180, time steps: 1 Jan 98 to 31 Aug 01

For correlation of data in b) with c) I defined a new grid for the precip variable:
define axis/x=0.5E:0.5W:1 xax
define axis/y=86n:86s:1 yax
define axis/t=01-jan-1997:31-aug-2001:24 tax
define grid/x=xax/y=yax/t=tax pw
let precip_w=precip[g=pw]

Then I define new variables for the region/matching time I wanted to perform the correlation over :

LET p = Precip_w[d=2, x=120e:70w, y=50s:35n,l=943:1704]  NOTE: (both variables begin on 20 Jul 99
LET q = U[d=1, x=120e:70w, y=50s:35n,l=13:774]                                and end on 31 Aug. )
list/x=170e:180e/y=10s:15s/l=1 p, q

However, I get the following errors:

*** NOTE: Invalid L axis limits ignored: P
 *** NOTE: Invalid L axis limits ignored: Q
 **ERROR: inconsistent sizes of data regions: X axis
          Variable 2 is I=171:181
          expression is I=171:180

list/x=170e:180e/y=10s:15s/l=1 p, q
Command file, command group, or REPEAT execution aborted


Is there a different direction I should take for this type of problem?  Do I have to re-write the data into matching time steps?  Would greatly appreciate your help on this question.

SCOCKS
TAMU