[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

19990929: ECMWF data

>From: Brett Newkirk <address@hidden>
>Organization: .
>Keywords: 199909300018.SAA01958

>Hi Steve:
>       I was looking at bd_gem_grid.f and was going to make my own
>executable of dcgrib with the grid #255 hard-coded in to see if it works.
>But would I not need one grid # for my tile C (continental US) and another
>grid # for my tile D (Pacific) as in the different sectors for ECMWF
>(INFO array numbers 16-27)? In this program, what are scol and srow
>for-are they pertinent to if I were defining a tile of a global grid (i.e.
>because there are only 55 entries in the DATA statements) or do they
>define the global grid? 
>Thank you for answering my questions on this-
>Brett
>
>Brett Newkirk    E-MAIL: address@hidden
>Office: ATG 424  Atmospheric Science/Geophysics Building
>Mailing Address: Department of Atmospheric Sciences, University of Washington
>                 Box 351640
>                 Seattle, WA 98195-1640
>Office Phone: (206) 685-2183
>
>

Brett,

You hit the nail on the head.
The srow/scol are the starting row/col for the grid tile within the
larger grid- eg, the array offsets when writing the data block.

The problem being that the design right now assumes that gridid #255
uses the gds to define a single grid and not a tile of a larger
grid area. Whereas, when you have 8 different grid id's, it
is easy to figure out that they can be stitched together into a larger area.

I need to see how I would determine a) that the grid #255 is a tile
and not the complete grid b) how to compute the srow/scol from
the lat/lon information in the grib header.

We could cheat for this one data set and hard code in something that
looks at grid 255 for ecmwf, and renumbers the grid to something
else dependent on the lat/lon sector contained within the grid.
That would allow us to stitch these grids together for you.
As I mentioned, this will not allow a general solution.

We have been in our User Committee meetings the past 2 days, so I wasn't
able to look at this message yesterday.

If you could ftp to the ~gbuddy/incoming directory a grib file from
both of the tiles you are looking at, I can see what I can do to seam things 
together.

Steve Chiswell