Re: calculate grid area a function of latitude and longitude onearth's surface

[Ansley Manke <ansley.b.manke@xxxxxxxx>]

Hi Steve,
Ferret can easily use integration to compute this.  If your X and Y axes
have units of degrees longitude and latitude, Ferret automatically uses a
2 dimensional integration, with the cosine(latitude) factor applied.  The
result is in units of square meters.

You need to define a variable on your grid whose value is one everywhere,
and be careful when you specify the limits of integration that you want. The
header lines on a listing show that you have done an "XY integration" and it
will list the region where the transformation was applied. Look up the topic
"transformations, general information" in the Ferret Users Guide for a discussion. 
Here is an example:

! A dataset whose xy grid is in units of latitude and longitude
  USE coads_climatology
  SAY `sst,RETURN=XUNITS`

! define a variable whose value is 1 everywhere.
! its integral will give the area in different grid cells.

  LET v = 1 + (0*x[gx=sst] + 0*y[gy=sst])

! Compare the area of a 1x1 degree box and a 1x2 degree box.

  LET one_by_one = v[x=1:2@din,y=-0.5:0.5@din]; LIST one_by_one
  LET two_by_one = v[x=1:3@din,y=-0.5:0.5@din]; LIST two_by_one
 
  LIST two_by_one/ one_by_one

! 1 by 1-degree area at 45 degrees N
  LIST   v[x=1:2@din,y=44.5:45.5@din]

! 1x1 degree box near the pole
  LIST v[x=1:2@din,y=88:89@din]


Ansley Manke

Steve Knox wrote:

I have a grid with each grid cell  "cellsize" degrees by "cellsize" degrees, covering the world. Does anyone have an algorithm for calculating the grid area  as a function of latitude and longitude?
Thanks for any help.

-- 
_________________________
Steve Knox
NREL
Colorado State University
(970)491-1972