Assigning to part of a NetCDF4 Variable

[Matthew Norman]

I want to assign data to part of a netCDF4 Variable, but what I'm doing below does not seem to be working. What ends up in the file is not correct based on the binary data I'm reading in. Do you see any obvious errors, particularly in the assigning of data to the netcdf Varaible at the very en?

Thanks,

-Mtt

from Numeric import *

from struct import unpack

from netCDF4 import Dataset

x0 = -645*1000

y0 = -3370*1000

dx = 500

dy = 500

nx = 3010

ny = 5460

nt = 2

filebase = [ './2007/greenland_vel_mosaic500_2007_2008', \

             './2008/greenland_vel_mosaic500_2008_2009' ]

print('Creating netCDF file')

nc = Dataset('greenland_vel_mosaic500.nc', 'w', format='NETCDF4')

x_d  = nc.createDimension('x', nx)

y_d  = nc.createDimension('y', ny)

t_d  = nc.createDimension('t', nt)

x_v  = nc.createVariable('x' ,'f4',('x',))

y_v  = nc.createVariable('y' ,'f4',('y',))

t_v  = nc.createVariable('t' ,'f4',('t',))

vx_v = nc.createVariable('vx','f4',('t','y','x',))

vy_v = nc.createVariable('vy','f4',('t','y','x',))

vm_v = nc.createVariable('vm','f4',('t','y','x',))

vx_v.missing_value = -2.e9

vy_v.missing_value = -2.e9

vm_v.missing_value = -2.e9

x_v[:] = [i for i in range(x0,x0+nx*dx,dx)]

y_v[:] = [i for i in range(y0,y0+ny*dy,dy)]

t_v[:] = [2007.,2008.]

for n in range(nt) :

  print('Reading vx and vy for file: ',n)

  f = open(filebase[n]+'.vx','rb')

  vx_bin = f.read()

  f.close()

  f = open(filebase[n]+'.vy','rb')

  vy_bin = f.read()

  f.close()

  vx = [[0. for i in range(nx)] for j in range(ny)]

  vy = [[0. for i in range(nx)] for j in range(ny)]

  vm = [[0. for i in range(nx)] for j in range(ny)]

  for j in range(ny) :

    if (j%200 == 0) :

      print("Percentage complete: ",float(j)/ny*100.)

    for i in range(nx) :

      ind = j*nx+i

      beg =  ind   *4

      end = (ind+1)*4

      vx[j][i] = float( unpack('>f',vx_bin[beg:end])[0] )

      vy[j][i] = float( unpack('>f',vy_bin[beg:end])[0] )

      if (vx[j][i] != -2.e9) :

        vm[j][i] = sqrt( vx[j][i]**2 + vy[j][i]**2 )

      else :

        vm[j][i] = -2.e9

  print('writing vx, vy, and vm to file')

  vx_v[n,:,:] = vx

  vy_v[n,:,:] = vy

  vm_v[n,:,:] = vm

nc.close()

[Гузий Саша]

On Wednesday, 5 June 2013 17:18:23 UTC-4, Matthew Norman wrote:

I want to assign data to part of a netCDF4 Variable, but what I'm doing below does not seem to be working. What ends up in the file is not correct based on the binary data I'm reading in. Do you see any obvious errors, particularly in the assigning of data to the netcdf Varaible at the very en?

Thanks,

-Mtt

Hi Norman:

it would be easier to look at your problem if we had a script converting 1 binary file to a netcdf, and the input binary file.

Cheers

--

Sasha