environment

Ubuntu18.04LTS Python3

Environment

A NetCDF4 module is required to use NetCDF format files in Python3. This module was easy to install with pip3.

NetCDF file I / O in Python

The following is a program that opens a NetCDF file that stores the three-dimensional variables T2 of latitude, longitude, and time, and creates a new file called test_ncout.nc.

#coding: utf-8
# This is a sample program to read and write s netCDF file with Python3
from netCDF4 import Dataset
import numpy as np

#*~*~*~*~*~*~*~*~*~*~*~*~*~*~*
#read netCDF file
#*~*~*~*~*~*~*~*~*~*~*~*~*~*~*

#open a netCDF file to read
ifile = "test.nc"
ncin = Dataset(ifile, 'r')

#check
#print(ncin.file_format)

#get variables
#print(ncin.variables.keys())

#get axis data
tin = ncin.variables['time']
latin = ncin.variables['lat']
lonin = ncin.variables['lon']

#get length of axis data
ntime = len(tin)
nlat = len(latin)
nlon = len(lonin)

#check axis
#print(tim[:])
#print(latin[:])
#print(lonin[:])

#read data
vin = ncin.variables['T2']

#check data
#print(vin[:,:,:,:])

#*~*~*~*~*~*~*~*~*~*~*~*~*~*~*
#write netCDF file
#*~*~*~*~*~*~*~*~*~*~*~*~*~*~*

#open a netCDF file to write
ncout = Dataset('test_ncout.nc', 'w', format="NETCDF4")

#check file format
#print(ncout.file_format)

#define axix size
ncout.createDimension('time',ntime)
ncout.createDimension('lat', nlat)
ncout.createDimension('lon', nlon)

#create time axis
time = ncout.createVariable('time', np.dtype('f4').char, ('time'))
time.long_name = 'time'
time.units = 'months since 1850-1-1'
time.axis = 'T'

#create lat axis
lat = ncout.createVariable('lat', np.dtype('f4').char, ('lat'))
lat.long_name = 'latitude'
lat.units ='degrees_north'
lat.axis = 'Y'

#create lon axis
lon = ncout.createVariable('lon', np.dtype('f8').char, ('lon'))
lon.long_name = 'longitude'
lon.units = 'degrees_east'
lon.axis = 'X'

#create variable arry
vout = ncout.createVariable('T2', np.dtype('f4').char, ('time', 'lat', 'lon'))
vout.long_name = '2m temperature'
vout.units = 'K'

#copy axis from original data
time[:] = tin[:]
lon[:] = lonin[:]
lat[:] = latin[:]
vout[:] = vin[:]

ncin.close()
ncout.close()

reference

http://unidata.github.io/netcdf4-python/ https://github.com/Unidata/netcdf4-python

Handle NetCDF format data in Python

environment

Environment

NetCDF file I / O in Python

reference