CODEWITHSUNDEEP
pythonnetcdf
jms1980
jms1980

Reputation: 1075

making 4 dimensional netcdf file from hycom data

I am trying to make a python script to make a 4 dimensional netcdf file from the website I have the following python script:

'http://nomads.ncep.noaa.gov:9090/dods/rtofs/rtofs_global'+now.strftime("%Y%m%d")+'/rtofs_glo_3dz_forecast_daily_uvel

and 

'http://nomads.ncep.noaa.gov:9090/dods/rtofs/rtofs_global'+now.strftime("%Y%m%d")+'/rtofs_glo_3dz_forecast_daily_uvel

I get a netcdf file with the following attributes:

netcdf hycom {
dimensions:
    latitude = 30 ;
    longitude = 30 ;
    level = 4 ;
variables:
    float v(level, latitude, longitude) ;
        v:units = "m/s" ;
    float u(level, latitude, longitude) ;
        u:units = "m/s" ;
    float longitude(longitude) ;
        longitude:units = "degrees_east" ;
    float level(level) ;
    float latitude(latitude) ;
        latitude:units = "degrees_north" ;

data:

 longitude = 107.492, 107.5753, 107.6587, 107.742, 107.8253, 107.9087, 
    107.992, 108.0753, 108.1586, 108.242, 108.3253, 108.4086, 108.492, 
    108.5753, 108.6586, 108.742, 108.8253, 108.9086, 108.9919, 109.0753, 
    109.1586, 109.2419, 109.3253, 109.4086, 109.4919, 109.5752, 109.6586, 
    109.7419, 109.8252, 109.9086 ;

 level = 10, 20, 30, 50 ;

 latitude = 76.66, 76.74333, 76.82666, 76.90999, 76.99332, 77.07665, 
    77.15998, 77.24331, 77.32664, 77.40997, 77.4933, 77.57663, 77.65996, 
    77.74329, 77.82662, 77.90995, 77.99328, 78.07661, 78.15994, 78.24327, 
    78.3266, 78.40993, 78.49326, 78.57659, 78.65992, 78.74325, 78.82658, 
    78.90991, 78.99324, 79.07657 ;
}

However when I try to open the file on grads I get the following message:

ga-> sdfopen hycom.nc
Scanning self-describing file:  hycom.nc
SDF file has no discernable time coordinate -- using default values.
gadsdf: SDF file does not have any non-coordinate variables.

I have been able to successfully make netcdf files from 2d hycom data but not from the 3d hycom data. Do you know what I need to do to create a 3d hycom netcdf file with levels, longitude, latitude, and time.

The shape file is (9,33,2142,4369) or (time, level,latitude,longitude).

 #!/usr/bin/python
 from pydap.client import open_url
 from datetime import *
 import datetime
 now = datetime.datetime.now()
 import pickle
 import pupynere
 import shutil

 datasetu = open_url('http://nomads.ncep.noaa.gov:9090/dods/rtofs        /rtofs_global'+now.strftime("%Y%m%d")+'/rtofs_glo_3dz_forecast_daily_uvel')
 datasetv = open_url('http://nomads.ncep.noaa.gov:9090/dods/rtofs/rtofs_global'+now.strftime("%Y%m%d")+'/rtofs_glo_3dz_forecast_daily_vvel')

u = datasetu['u']
u.shape
print ">>>>> u.shape is",u.shape
print u[1,1:5,400:430,2000:2030]

v = datasetv['v']
v.shape
print v[1,1:5,400:430,2000:2030]
y=u.lat
x=u.lon
z=u.lev

f = pupynere.netcdf_file("hycom.nc",'w')
f.createDimension("latitude",30)
f.createDimension("longitude",30)
f.createDimension("level",5)

lats=f.createVariable("latitude","f",("latitude",))
lats[:]=y[2000:2030]
lons=f.createVariable("longitude","f",("longitude",))
lons[:]=x[400:430]
levs=f.createVariable("level","f",("level",))
levs[:]=z[5]

lats.units = 'degrees_north'
lons.units = 'degrees_east'
levs.units = 'meters below sea level'


#print u[1,0,400:430,2000:2030]
ucur = f.createVariable("u","f",("level","latitude","longitude"))
ucur[:]=u[1,1:5,400:430,2000:2030]
ucur.units = "m/s"
########################################################
#print v[0,0,1000:1400,1000:1400]
vcur = f.createVariable("v","f",("level","latitude","longitude"))
vcur[:]=v[1,1:5,400:430,2000:2030]
vcur.units = "m/s"

Upvotes: 0

Views: 704

Answers (1)

N1B4
N1B4

Reputation: 3453

You need to match the shapes between ucar and u and vcur and v during the write out procedure. Since you are extracting a 3D slice of u and v, you need to make sure that you populate 3 dimensions of ucur and vcar: 

ucur[:,:,:] = u[1,1:5,400:430,2000:2030]
vcur[:,:,:] = v[1,1:5,400:430,2000:2030]

Upvotes: 1

Related Questions