#------------------------------------------------------------------------------------------
MAIN Fortran file:   proc_OSTIA.F90
CSH  script:         make_OSTIA_data.x
*************************************

1 . SUBROUTINE read_ncFile:
    Extract lat, lon, sst, frac of sea ice (fraci) from daily ostia nc file
   *lat: -90 + dy/2 to 90-dy/2; lon: -180+dx/2 to 180-dx/2
    dx = 180/nx. dy = 360/ny. where nx = 7200 & ny = 3600.
  
    Returns lat, lon, sst & fraci. taking into account scale factor & offset.
    Over land, you have undef (fill values). Over ocean, you already have correct values for sst [in Kelvin] & fraci \in [0, 1].

2.  FillUpLand: [get these by ncdump -h]
    sst_FillValue       = -32768
    seaIce_FillValue    = -128
    Model needs values @ every grid pt. Fill up values over land using ocean values.

3.  bin2bin: 
    Bin to output resolution (nx_out, ny_out) and to date-line edge, pole-edge:
    
4.  Dump SST & FRACI to bin files with header [start_date, end_date, resolution].

5.  you `build' fortran code using: GNUmakefile
    gmake make_ostia_bcs.x ESMADIR=/discover/nobackup/sakella/GEOSadas

    then use the csh script: OSTIA_BCS.csh
    this will generate: ostia_sst.bin & ostia_fraci.bin  
    which can be plotted using sst_bin.ctl, fraci_bin.ctl

6.  To clean up build
    gmake distclean ESMADIR=/discover/nobackup/sakella/GEOSadas
#------------------------------------------------------------------------------------------
Current dao_ops wkly Reynolds sst is being processed:

/gpfsm/dhome/dao_ops/operations/geos5/sst_eros/ 
mksst.f90

see matlab processing
/discover/nobackup/sakella/processData/SST_forDAS/process_OSTIA1.m

SA
06/21/2012
04/2012
#------------------------------------------------------------------------------------------
#
