[Thread Prev][Thread Next][Index]

Problems with the EOF_SPACE external function



Hi

Some time ago, I wrote a message to this group concerning problems
I had with the external function EOF_SPACE. I found a workaround and
those
how are interested, please read on.

What happened was that in some cases, the result was only unvalid data.
But
this was not systematic. In particular, the problem always occured when
I tried
to calculate EOFs for the Northern Hemisphere, and never the the
Southern Hemisphere.
I now found out that the problems seem to come from the grid the
original data are on.
In fact, the grid is regular, with 96 longs and 73 lats. The northermost
latitude
is the North Pole (and similar for the south Pole). Therefore, the first
line of 
data in the file consists of 96 equal values (because physically it is
only one point,
the North Pole). For some reason ferret's EOF functions do not seem to
like
this. Funny enough, at the South Pole there are also 96 equal points,
but the EOF
functions don't seem to have any problems with this there. 

What I did to solve the problem was to manually (and slightly) modify
one of these 96
equal values at the North Pole in the input data (say, the point -180W,
90N would be
set to 1000.1 hPa instead of 1000.0 hPa, while the other points at 90N
would remain
unchanged). With this trick, the EOF function seems to work (the change
must not be
too weak though). It is enough to make this change for only one of the n
(n>1) time
steps given as input to EOF_SPACE.
 
Of course this is only a quick and dirty workaround but it seems to
work. I hope that
someone finds the time to look what really happens, and if somebody can
suggest a
cleaner way to solve this problem, please don't hesitate to tell me.

Cheers
Gerhard

-- 
_____________________________________________

  Gerhard Krinner
  LGGE / CNRS
  54 Rue Moliere, DU BP 96
  38402 Saint Martin d'Heres Cedex, FRANCE
  telephone: (33) 4 76 82 42 41
  facsimile: (33) 4 76 82 42 01
  courriel: krinner@glaciog.ujf-grenoble.fr  
_____________________________________________


[Thread Prev][Thread Next][Index]

Dept of Commerce / NOAA / OAR / PMEL / TMAP

Contact Us | Privacy Policy | Disclaimer | Accessibility Statement