Re: Re: Problem in creating new 3-dimensional variables.

[andreas macrander <amacrander@xxxxxxxxxxxxxxx>]

Hi Steve!
Thank you very much for your suggestion. I have already tried it, but the
result is not what I wanted to get. 
Now I want to explain the problem better: The layer thicknesses look like 
these:
yes? list/i=1/j=1/l=1  DZ
             Schichtdicke (m)
             Y (km): 0
             DATA SET: ./rho.cdf
                   0
                   1
 -1289.7  /  1:  852.4      ! <-- bottom layer
 -1155.6  /  2:   82.1
 -1025    /  3:   81.8
 -897.5   /  4:   81.5
 -772.7   /  5:   81.3
 -650.1   /  6:   81.1
 -529.4   /  7:   80.9
 -410.2   /  8:   80.8
 -292.2   /  9:   80.7
 -175     / 10:   80.6
 -58.3    / 11:   80.5      ! <-- surface layer

If I follow your suggestion, I get this:        What I want to get:

yes? LET depth = dz[k=1:11@sum] - dz[k=@rsum] ! let depth = ???????????
Replacing definition of variable DEPTH
yes? list/i=1/j=1/l=1 depth                    list/i=1/j=1/l=1 depth
             DZ[K=1:11@SUM] - DZ[K=@RSUM]      ?????????????????????
             Y (km): 0                         Y (km): 0
             DATA SET: ./rho.cdf               DATA SET: ./rho.cdf
                   0                                 0   
                   1                                 1
 -1289.7  /  1:  811.2                         1:  1664.0  !=DZ[K=1:11@SUM]  
 -1155.6  /  2:  729.1                         2:   811.2  !=DZ[K=2:11@SUM]
 -1025    /  3:  647.3                         3:   729.1  
 -897.5   /  4:  565.8                         4:   647.3
 -772.7   /  5:  484.5                         5:   565.8
 -650.1   /  6:  403.4                         6:   484.5
 -529.4   /  7:  322.5                         7:   403.4
 -410.2   /  8:  241.8                         8:   322.5
 -292.2   /  9:  161.1                         9:   241.8
 -175     / 10:   80.5                        10:   161.1  !=DZ[K=10:11@SUM]
 -58.3    / 11:    0.0                        11:    80.5  !=DZ[K=11@SUM]

I have failed to say ferret to do something like "let depth=DZ[K=k:11@SUM]".
Do you know how to force ferret to create a variable like that on the right
side? 

Thank you very much,
Andreas.