Fw:Help: Interpolation failure using linint2_Wrap

From: heaven_bird334 <heaven_bird334_at_nyahnyahspammersnyahnyah>
Date: Wed Jun 13 2012 - 21:06:46 MDT

Hi,
       I met a problem when interpolating SST from fixed grids (89*180) to R42 grids (108*128) using linint2_Wrap. After the interpolation, the new SST data contains too much "_FillValue" of "missing" value, and the new SST can not plot out contours. This is not the right result, however, no error information is given out during the whole process.
      The flowing is my scripts, which is simplified with detailed annotation. Also, the relative datas are all attached behind this letter as a tar bar. Any suggestion or help is appreciable!

;........................................................
load "$NCARG_ROOT/lib/ncarg/nclscripts/csm/contributed.ncl"
begin

  inf = "sst_initial.nc" ; Initial sst(89,180)
  latlonfi = "latlon_Out.nc" ; New lat(108) and lon(128) for interpolation , that is R42 grid
  outfi = "sst_interpolation.nc" ; New interpolated sst(108,128)

;
; read in initial sst
;
  f = addfile(inf,"r") ; open initial sst data for reading
  sst = f->sst
  delete(f)
  printVarSummary(sst) ; sst [lat | 89] x [lon | 180]
                              ; lat: [88..-88]
                              ; lon: [ 0..358]

;
; read in R43 lat and lon info
;
  f = addfile(latlonfi,"r") ; open initial sst data for reading
  LAT = f->lat
  LON = f->lon
  delete(f)
  printVarSummary(LAT) ; [lat | 108]; lat: [88.73005..-88.73004]
  printVarSummary(LON) ; [lon | 128]; lon: [ 0..357.1875]

;
; interpolation
;
  sstNew = linint2_Wrap (sst&lon,sst&lat,sst, False, LON,LAT, 0)
  printVarSummary(sstNew) ; [lat | 108] x [lon | 128]
                             ; lat: [88.73005..-88.73004]
                             ; lon: [ 0..357.1875]
 

;
; output nc
;
  system ("rm "+outfi) ; remove any pre-existing file
  ncdf = addfile(outfi,"c") ; create new netCDF file
  ncdf->sstNew = sstNew

  print(sstNew({0},{190:230})) ; All missing data!
                                   ;(0) 32767
                                   ;(1) 32767
                                   ;... ...
                                   ;(13) 32767

                                   ; However, SSTs over Topical Easten Pacific shouldn't be missing value!
end

       

Best Regards,
Li

_______________________________________________
ncl-talk mailing list
List instructions, subscriber options, unsubscribe:
http://mailman.ucar.edu/mailman/listinfo/ncl-talk

Received on Wed Jun 13 21:06:58 2012

This archive was generated by hypermail 2.1.8 : Fri Jun 15 2012 - 14:51:31 MDT