derivative/finite difference question

[Brent A McDaniel <gte328r@xxxxxxxxxxxxxxxxxxxxx>]

Hi Ferret Users,

I'm trying to calculate the wave activity which requires calculating
vertical as well as meridional (d/dlatitude) derivatives.  The meridional
coordinate is degrees on a regular grid and the vertical cood is pressure
on an uneven grid.

I understand what the meridional derivatives are doing, they convert the
delta(latitude) (2.5degrees in this case) to a linear distance by
converting 2.5 to .04363radians then multiplying by radius of the earth
(6370000m) to get the arc length dx.  Somewhat of a pain as I want d/dlat
but it can be worked around.


The vertical derivative is confusing me though.  Z-axis is in pressure
coods with units mb.  An excerpt from ncdump:

	double MY_ZAX(MY_ZAX) ;
		MY_ZAX:units = "MB" ;
		MY_ZAX:positive = "down" ;
		MY_ZAX:point_spacing = "uneven" ;
		MY_ZAX:edges = "MY_ZAXedges" ;
	double MY_ZAXedges(MY_ZAXedges) ;
		MY_ZAXedges:edges = " " ;


After reading section 2.4.1 I would expect ferret to transform mb to
meters, in essence calculating d/dz instead of d/dp but a quick test shows
me that fy(k=@ddf) is doing a straight finite difference with no unit
conversion.

Can someone explain why the meridional derivative is being converted to
length (meters) but the vertical derivative is not.  (for completeness,
fy is unitless in the netcdf file)

Also, would it be possible to include basic finite difference
transformations in addition to ddf, ddc, and ddb?  Or do these already
exist and I'm just missing them?

Cheers,

Brent

 --
Brent A. McDaniel

Dept of Earth and Atmospheric Sciences
Georgia Institute of Technology
Atlanta, Ga.  USA