[RegCNET] Differences in EH5OM SST data from CERA and ICTP

[Mega Octaviani]

Dear RegCNET,

I am trying to use ECHAM5/MPI-OM downloaded from CERA as my ICBC input.
I also found that we can get this GCM output only for A1B (run 3) experiment
from http://www.ictp.it/~pubregcm/RegCM3/globedat3.htm.
My question now is to see if the two sources give the same type of data
although in different format.

To compare them, I reformatted both into GrADS binary. Following is the main
part of my codes:
1. For CERA grib data:
   --> after do $ wgrib -i -text gribfile -o textfile

   do j = niy, 1, -1
      do i = 1, njx
         read (10, *, iostat = readstatus) val(i, j)
         if (readstatus > 0) stop 'error on reading unit 10'

         if (val(i, j) .lt. 273.16) val(i, j) = -9999.
      end do
   end do

2. For ICTP data:

    read(10, rec = rrec) offset, scale, itmp

    do j = 1, niy
      do i = 1, njx
         tmp(i, j) = itmp(i, niy + 1 - j) * scale + offset

         if (tmp(i, j) .lt. 273.16) tmp(i, j) = -9999.
      end do
   end do

FYI, I did testing on year 2030.

The results show that over oceans, both SST data are the same. However,
there are differences for temperature over lands.

For instance, SST data from CERA give undefined values over Australia, South
America, and Africa but these are not found in SST from ICTP (values larger
than 273.16K).

Can someone give me some clues on this problem?


Regards,

-
Mega Octaviani
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.ictp.it/pipermail/regcnet/attachments/20101209/1c7c99d2/attachment.html>