[Thread Prev][Thread Next][Index]

derivatives on an uneven grid



Hi Ferret users,

I'm trying to calculate vertical derivatives on an uneven grid (the ncep 
reanalysis pressure grid) that has a vertical axis:

yes? list/i=1/j=1/l=1 z[g=uwnd] 
             VARIABLE : Z
                        axis LEVEL
             DATA SET : mean daily NMC reanalysis (2000)
             FILENAME : uwnd.2000.nc
             SUBSET   : 17 points (HEIGHT (millibar))
 10      /  1:    10.
 20      /  2:    20.
 30      /  3:    30.
 50      /  4:    50.
 70      /  5:    70.
 100     /  6:   100.
 150     /  7:   150.
 200     /  8:   200.
 250     /  9:   250.
 300     / 10:   300.
 400     / 11:   400.
 500     / 12:   500.
 600     / 13:   600.
 700     / 14:   700.
 850     / 15:   850.
 925     / 16:   925.
 1000    / 17:  1000.


If I use a standard centered-finite difference scheme I get the wrong
answer as the grid points aren't equally spaced.  Has anyone 
encountered/solved this problem before?  

One way to deal with this is to assume a variable theta has the form:

theta=Ap**2 + Bp + C where p is pressure and A, B, and C are coefficients.  
THen the vert. deriv dtheta/dp=2Ap + B.  You can invert the 3x3 matrix to 
get the coefficients and determine the deriv.  I'm not sure how to do this 
in ferrret w/out writing an external function which I'm trying to avoid.

Does anyone have a similar/better way to do this?

Thanks,

Brent

-- 
Brent A. McDaniel

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



[Thread Prev][Thread Next][Index]

Dept of Commerce / NOAA / OAR / PMEL / TMAP

Contact Us | Privacy Policy | Disclaimer | Accessibility Statement