20020523: McIDAS DERIVE keywords

>From: "Jennie L. Moody" <address@hidden>
>Organization: UVa
>Keywords: 200205232127.g4NLRTa11165 McIDAS GRDCOPY DERIVE=

Jennie,

>I am just trying to re-derive (follow) what Owen and I have
>done before in calculating PV from grids.  I am trying to 
>find out exactly how math keywords work, like DDX and DDY, 
>when one uses the DERIVE keyword in a GRDCOPY command.

OK, a lofty goal ;-)

>In particular I wanted to see if these are centered differences.
>I found a comment that said these are defined in the McIDAS
>User's Guide, but I cannot locate them.  Do you know where this
>might be?  

The McIDAS-X 7.80 User's Guide section on GRDCOPY has a table
for DERIVE= parameters:

http://www.unidata.ucar.edu/packages/mcidas/780/users_guide/McHTML-136.HTML

This page does not, however, specify exactly how DDX and DDY are defined
other than saying:

DDX - derivative with respect to x
DDY - derivative with respect to y

>C ?    A variety of common meteorological parameters can be derived using
>C ?    the DERIVE keyword. These include VOR (vorticity), ABV (absolute
>C ?    vorticity), DSH (shear deformation), DST (stretching deformation),
>C ?    DVG (divergence), SPD (speed), DIR (direction), TD (dewpoint),
>C ?    COR (Coriolis parameter), and BETA (Beta parameter).
>C ? 
>C ?    The MATH keyword defines the operation to perform on the specified
>C ?    Gn grids. For example, if G1='PARAM Z;LEV 500;TIME 12;SRC MRF'
>C ?    G2='LEV 1000' MATH='G1-G2' is entered, a 1000 mb height grid will be
>C ?    subtracted from a 500 mb height grid, creating a 1000-500 mb thickness
>C ?    grid for the 12 UTC MRF model run. The math operators valid in the MATH
>C ?    keyword are +,-,/,*,**,SQRT,EXP,LOG,LOG10,SIN,COS,TAN,ASIN,ACOS,ATAN,
>C ?    ABS,MIN,MAX,DDX,DDY,DELSQ,COR,BETA,LAT, and LON.
>C ?
>C ?    See the McIDAS User's Guide for definitions of the operators.

>Thanks.

I could try to find out exactly how the derivatives are done by reading
the code if you need the info.  Please let me know.

>be well,

You too!

Tom