From: David Small <david.small2_at_nyahnyahspammersnyahnyah>
Date: Sat, 16 May 2009 12:04:59 -0400

Hello everyone,

Iım trying to calculate the gradient of the geostrophic wind from ECMWF
Reanalysis on a reduced gaussian grid. To do so, Iım using the centered
finite difference routine in NCL to avoid a problem near the equator when
using the spherical harmonic routines. I have no problem when I calculate
the meridional derivative, it compares well to that which I calculated from
the 2.5 degree ECMWF on a fixed grid. The problem is with the longitudinal
derivative which is giving me junk. Does anyone have any clue what the
problem might be?

As always, thank you very much for all your help.

Dave Small

dlon = 1.125*0.0174533 ; convert to radians
dlat = dlon*6378388.0

do nl=0,72

dX = 6378388.*cos(0.0174533*lat(nl))*dlon ; constant at this latitude
v_grad_lon(:,:,nl,:) = center_finite_diff(v(:,:,nl,:), dX , True,0)
u_grad_lon(:,:,nl,:) = center_finite_diff(u(:,:,nl,:), dX , True,0)

end do

```--
David Small