[ncl-talk] coordinates in raster file

Vanúcia Schumacher vanucia-schumacher at hotmail.com
Mon Jun 3 17:23:36 MDT 2019


Hi users,

Could anyone check if I am reading and associating correctly the lat and lon coordinates of this raster file (attachment) with NCL?

Thanks

Part of the script:

      files  = systemfunc("ls test.asc")
       hdr = readAsciiHead(files, 6)
     ncols = stringtoint  ( str_get_field(hdr(0), 2, " ") )
     nrows = stringtoint  ( str_get_field(hdr(1), 2, " ") )
     lonLL = stringtofloat( str_get_field(hdr(2), 2, " ") )
     latLL = stringtofloat( str_get_field(hdr(3), 2, " ") )
  deltaLon = stringtofloat( str_get_field(hdr(4), 2, " ") )
  deltaLat = stringtofloat( str_get_field(hdr(4), 2, " ") )
missingVal = stringtofloat( str_get_field(hdr(5), 2, " ") )

data     = readAsciiTable(files,ncols,"float",6)
                   data at _FillValue = missingVal

;----- construct 1D coordinates....
lons             = ispan(0,ncols-1,1) * deltaLon + lonLL
lats             = ispan(0,nrows-1,1) * deltaLat + latLL
lats             = lats(::-1)

;---- gsn_coordinates wants 2D coordinate arrays
data at lon2d       = conform_dims((/ nrows, ncols/), lons, 1)
data at lat2d       = conform_dims((/ nrows, ncols/), lats, 0)

  data!0                 = "lat"
  data&lat              = lats
  data&lat at units  = "degrees-north"
  data!1                 = "lon"
  data&lon            = lons
  data&lon at units = "degrees-east"

;---- Open shapefile and read lat/lon values.
  ...
   data_mask = shapefile_mask_data(data,shp_filename,True)
                        copy_VarMeta(data, data_mask)
  data_point := (/data_mask({-34.64},{-70.32})/)
...
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20190603/b5324df0/attachment.html>
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: test.asc
URL: <http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20190603/b5324df0/attachment.asc>


More information about the ncl-talk mailing list