[Thread Prev][Thread Next][Index]

Re: [ferret_users] EOF_TFUNC for 4-D data



hi Thomas,
The EOF functions do assume the data is in XY and T, and if there is a Z axis it computes the EOF for each level of Z. You should be able to RESHAPE the variable to an XYT grid before sending it to the function and then apply another RESHAPE to the result.

Ansley


Thomas Froelicher wrote:
Dear ferret-users,

I d'like to calculate EOF time amplitudes from a 4 dimensional model output in y-z-time space. The model data looks like: name title I J K L
 AOUSIGMA                            1:1       1:61      1:221     1:470

If I want to plot EOF time amplitude fields like:
let eof_xyfcn = eof_space(AOUSIGMA[y=35N:63N,z=31:32.6], 0.5)
plot eof_xyfcn[I=1]

then an error appears:

 **ERROR: dimensions improperly specified: specified data is not a line
           - its a 2D region: "plot eoftime[I=1]"

But If I change the model ouput axis (eg. I define z axis on the x axis) like:

name     title                      I         J         K         L
 AOUSIGMA                          1:221     1:61      1:1       1:470

then the calculation works!

I used Ferret version ferret-6.0.6,ferret.5.8.1 and ferret6.0, but no of them works. Do you have any ideas to avoid the permutation of the axis?

Cheers,

thomas







[Thread Prev][Thread Next][Index]

Contact Us
Dept of Commerce / NOAA / OAR / PMEL / TMAP

Privacy Policy | Disclaimer | Accessibility Statement