THE FILE PDEF OPTION THIS HASNT BEEN THOROUGHLY

 JOINT ESTIMATION OF TECHNOLOGY ADOPTION AND LAND ALLOCATION
STUDY NAME OPTIONS ANALYSIS PREPARED BY NAME TITLE JOB
ORGANIZATIONAL STRATEGIC PLANNING SUMMARY THREE OPTIONS OPTION

(SAMPLE CONVERSION OPTION RESOLUTION) [SCHOOL DISTRICT NAME] RESOLUTION NUMBER
1 CONSECUTIVE ORDER NUMBER 2 ANY (OPTIONAL) INTERNAL DESIGNATION
10 REDUCTION OPTIONS (ORS 291216) ACTIVITY OR PROGRAM DESCRIBE 

The "file" pdef Option
-----------------------

This hasn't been thoroughly tested yet.

Using the file specification on the pdef record, you can supply an arbitrary
number of grids to specify the interpolation you want.

The form of this record is:


    PDEF size 1 FILE  num <STREAM>     <BINARY>        fname
                          <SEQUENTIAL> <BINARY-BIG>
                                       <BINARY-LITTLE> 

where the <> brackets indicate required keyword options, and 
where:

      size      size of the input vector (eg, input x-y 'grid')
      num       number of sets of interpolation grids supplied
      fname     the name of a file which contains the indicated number of
                sets of interpolation grids, plus a wind rotation grid.
      STREAM    indicates the file (fname) is stream, or direct access, 
                format.
      SEQUENTIAL  indicates the file is FORTRAN sequential format.
      BINARY    indicates the file is in the binary format of the 
                machine GrADS is currently running on.
      BINARY-BIG  indicates the file is BIG_ENDIAN
      BINARY-LITTLE  indicates the file is LITTLE_ENDIAN

For example:  

pdef 5000 1  file  4 sequential binary-big /usr/local/lib/grads/mygrid.interp.values

Now this gets a bit complicated.  You indicate the number of sets of interpolation
grids.  These interpolation grids determine how the interpolation is done from the
input "grid" (best to think of it as an input vector, since it doesn't have to 
be a grid).  All of the interpolation grids are values provided for each lat-lon 
point, such lat-lon grids determined by the xdef and ydef records.  Each set of
interpolation grids consists of two grids, the first being an offset to dtermine
the input point of interest, the 2nd being a weight to be used for that point.
The frist grid of the set is int, the 2nd grid is float.

So, if you provide four sets of interpolation grids, you would be actually 
supplying 9 grids.   (four sets of two grids plus the wind rotation grid).

To obtain a value for a particular lat-lon point, GrADS will reference the
input vector along with the interpolation grids.  It will obtain each point
(the number of points determined by the indicated number of sets of 
interpolation grids), multiply by the weight, sum over all the weighted input
points, and divide by the summ of the weights.  Input offset values of -999
indicate not to use an input point for that portion of the calculation
(thus you can apply less than the "num" number of interpolation points for
some of the points).  

So in theory this can be used for such things as thinned grids, staggered
grids, etc.    Hope to have an example of this soon.


10144 CHAPTER 335 (REV 253A) HEALTH INSURANCE PURCHASE OPTION
109312 CONSENT TO ADOPTION (1) EXCEPT AS PROVIDED IN
19 NOVEMBER 2021 RESPONSE TO CONSULTATION AND OPTIONS PAPER


Tags: hasnt, thoroughly, option