scipy io netcdf issue with NaN written to integer variable

scipy-netcdf-nan-integers.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from __future__ import print_function\n",
    "from scipy.io import netcdf\n",
    "import os.path\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'0.17.0'"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import scipy\n",
    "scipy.__version__"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "rootdir = os.path.join(os.path.expanduser('~'), 'Downloads')\n",
    "fname = os.path.join(rootdir, 'scipy_nan_test.nc')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Make some sample data in a masked array"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "masked_array(data = [0.0 1.0 2.0 3.0 4.0 5.0 -- -- -- --],\n",
       "             mask = [False False False False False False  True  True  True  True],\n",
       "       fill_value = 1e+20)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = np.ma.array(np.arange(10), float)\n",
    "data[data > 5] = np.ma.masked\n",
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([  0.,   1.,   2.,   3.,   4.,   5.,  nan,  nan,  nan,  nan])"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.filled(np.nan)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Write sample data to netCDF file as floats and integers, filled with `np.nan` or `_FillValue`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "float filled with np.nan: \n",
      " [  0.   1.   2.   3.   4.   5.  nan  nan  nan  nan]\n",
      "\n",
      "int filled with np.nan: \n",
      " [          0           1           2           3           4           5\n",
      " -2147483648 -2147483648 -2147483648 -2147483648]\n",
      "\n",
      "int written as masked array: \n",
      " [0 1 2 3 4 5 6 7 8 9]\n",
      "\n",
      "int filled with _FillValue: \n",
      " [          0           1           2           3           4           5\n",
      " -2147483648 -2147483648 -2147483648 -2147483648]\n"
     ]
    }
   ],
   "source": [
    "with netcdf.netcdf_file(fname, 'w') as ds:\n",
    "    ds.createDimension('time', 10)\n",
    "    time = ds.createVariable('time', 'i', ('time',))\n",
    "    time[:] = np.arange(10)\n",
    "    time.units = 'days since 2008-01-01'\n",
    "    \n",
    "    floatvar = ds.createVariable('data_float', 'f', ('time',))\n",
    "    floatvar[:] = data.filled(np.nan)\n",
    "    print('\\nfloat filled with np.nan: \\n', floatvar[:])\n",
    "    \n",
    "    intvar_masked = ds.createVariable('data_int_masked', 'i', ('time',))\n",
    "    intvar_masked[:] = data\n",
    "    print('\\nint written as masked array: \\n', intvar_masked[:])\n",
    "    \n",
    "    intvar = ds.createVariable('data_int', 'i', ('time',))\n",
    "    intvar[:] = data.filled(np.nan)\n",
    "    print('\\nint filled with np.nan: \\n', intvar[:])\n",
    "    \n",
    "    intvar_fill = ds.createVariable('data_int_fill', 'i', ('time',))\n",
    "    intvar_fill._FillValue = 1e30\n",
    "    intvar_fill[:] = data.filled(intvar_fill._FillValue)\n",
    "    print('\\nint filled with _FillValue: \\n', intvar_fill[:])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Note: Both when filled with `np.nan` and `_FillValue`, missing values are written to the netCDF file as the integer value of `np.nan`."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}