writing_netCDF.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "internals": {
     "slide_type": "subslide"
    },
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Writing netCDF data\n",
    "\n",
    "**Important Note**: when running this notebook interactively in a browser, you probably will not be able to execute individual cells out of order without getting an error.  Instead, choose \"Run All\" from the Cell menu after you modify a cell."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": false,
    "internals": {
     "frag_number": 1,
     "slide_helper": "subslide_end"
    },
    "slide_helper": "slide_end",
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "import netCDF4     # Note: python is case-sensitive!\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 1,
     "slide_type": "subslide"
    },
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Opening a file, creating a new Dataset\n",
    "\n",
    "Let's create a new, empty netCDF file named 'data/new.nc', opened for writing.\n",
    "\n",
    "Be careful, opening a file with 'w' will clobber any existing data (unless `clobber=False` is used, in which case an exception is raised if the file already exists).\n",
    "\n",
    "- `mode='r'` is the default.\n",
    "- `mode='a'` opens an existing file and allows for appending (does not clobber existing data)\n",
    "- `format` can be one of `NETCDF3_CLASSIC`, `NETCDF3_64BIT`, `NETCDF4_CLASSIC` or `NETCDF4` (default). `NETCDF4_CLASSIC` uses HDF5 for the underlying storage layer (as does `NETCDF4`) but enforces the classic netCDF 3 data model so data can be read with older clients.  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": false,
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 3,
     "slide_helper": "subslide_end"
    },
    "slide_helper": "slide_end",
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<type 'netCDF4._netCDF4.Dataset'>\n",
      "root group (NETCDF4_CLASSIC data model, file format HDF5):\n",
      "    dimensions(sizes): \n",
      "    variables(dimensions): \n",
      "    groups: \n",
      "\n"
     ]
    }
   ],
   "source": [
    "try: ncfile.close()  # just to be safe, make sure dataset is not already open.\n",
    "except: pass\n",
    "ncfile = netCDF4.Dataset('data/new.nc',mode='w',format='NETCDF4_CLASSIC') \n",
    "print(ncfile)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 3,
     "slide_type": "subslide"
    },
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Creating dimensions\n",
    "\n",
    "The **ncfile** object we created is a container for _dimensions_, _variables_, and _attributes_.   First, let's create some dimensions using the [`createDimension`](http://unidata.github.io/netcdf4-python/netCDF4.Dataset-class.html#createDimension) method.  \n",
    "\n",
    "- Every dimension has a name and a length.  \n",
    "- The name is a string that is used to specify the dimension to be used when creating a variable, and as a key to access the dimension object in the `ncfile.dimensions` dictionary.\n",
    "\n",
    "Setting the dimension length to `0` or `None` makes it unlimited, so it can grow. \n",
    "\n",
    "- For `NETCDF4` files, any variable's dimension can be unlimited.  \n",
    "- For `NETCDF4_CLASSIC` and `NETCDF3*` files, only one per variable can be unlimited, and it must be the leftmost (fastest varying) dimension."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": false,
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 5,
     "slide_helper": "subslide_end"
    },
    "slide_helper": "slide_end",
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "('lat', <type 'netCDF4._netCDF4.Dimension'>: name = 'lat', size = 73\n",
      ")\n",
      "('lon', <type 'netCDF4._netCDF4.Dimension'>: name = 'lon', size = 144\n",
      ")\n",
      "('time', <type 'netCDF4._netCDF4.Dimension'> (unlimited): name = 'time', size = 0\n",
      ")\n"
     ]
    }
   ],
   "source": [
    "lat_dim = ncfile.createDimension('lat', 73)     # latitude axis\n",
    "lon_dim = ncfile.createDimension('lon', 144)    # longitude axis\n",
    "time_dim = ncfile.createDimension('time', None) # unlimited axis (can be appended to).\n",
    "for dim in ncfile.dimensions.items():\n",
    "    print(dim)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 5,
     "slide_type": "subslide"
    },
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Creating attributes\n",
    "\n",
    "netCDF attributes can be created just like you would for any python object. \n",
    "\n",
    "- Best to adhere to established conventions (like the [CF](http://cfconventions.org/) conventions)\n",
    "- We won't try to adhere to any specific convention here though."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": false,
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 7
    },
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "My model data\n"
     ]
    }
   ],
   "source": [
    "ncfile.title='My model data'\n",
    "print(ncfile.title)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 8,
     "slide_helper": "subslide_end"
    },
    "slide_helper": "slide_end",
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "Try adding some more attributes..."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 8,
     "slide_type": "subslide"
    },
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Creating variables\n",
    "\n",
    "Now let's add some variables and store some data in them.  \n",
    "\n",
    "- A variable has a name, a type, a shape, and some data values.  \n",
    "- The shape of a variable is specified by a tuple of dimension names.  \n",
    "- A variable should also have some named attributes, such as 'units', that describe the data.\n",
    "\n",
    "The [`createVariable`](http://unidata.github.io/netcdf4-python/netCDF4.Dataset-class.html#createVariable) method takes 3 mandatory args.\n",
    "\n",
    "- the 1st argument is the variable name (a string). This is used as the key to access the variable object from the `variables` dictionary.\n",
    "- the 2nd argument is the datatype (most numpy datatypes supported).  \n",
    "- the third argument is a tuple containing the dimension names (the dimensions must be created first).  Unless this is a `NETCDF4` file, any unlimited dimension must be the leftmost one.\n",
    "- there are lots of optional arguments (many of which are only relevant when `format='NETCDF4'`) to control compression, chunking, fill_value, etc.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": false,
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 10,
     "slide_helper": "subslide_end"
    },
    "slide_helper": "slide_end",
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<type 'netCDF4._netCDF4.Variable'>\n",
      "float64 temp(time, lat, lon)\n",
      "    units: K\n",
      "    standard_name: air_temperature\n",
      "unlimited dimensions: time\n",
      "current shape = (0, 73, 144)\n",
      "filling on, default _FillValue of 9.96920996839e+36 used\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Define two variables with the same names as dimensions,\n",
    "# a conventional way to define \"coordinate variables\".\n",
    "lat = ncfile.createVariable('lat', np.float32, ('lat',))\n",
    "lat.units = 'degrees_north'\n",
    "lat.long_name = 'latitude'\n",
    "lon = ncfile.createVariable('lon', np.float32, ('lon',))\n",
    "lon.units = 'degrees_east'\n",
    "lon.long_name = 'longitude'\n",
    "time = ncfile.createVariable('time', np.float64, ('time',))\n",
    "time.units = 'hours since 1800-01-01'\n",
    "time.long_name = 'time'\n",
    "# Define a 3D variable to hold the data\n",
    "temp = ncfile.createVariable('temp',np.float64,('time','lat','lon')) # note: unlimited dimension is leftmost\n",
    "temp.units = 'K' # degrees Kelvin\n",
    "temp.standard_name = 'air_temperature' # this is a CF standard name\n",
    "print(temp)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 10,
     "slide_type": "subslide"
    },
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Pre-defined variable attributes (read only)\n",
    "\n",
    "The netCDF4 module provides some useful pre-defined Python attributes for netCDF variables, such as dimensions, shape, dtype, ndim. \n",
    "\n",
    "Note: since no data has been written yet, the length of the 'time' dimension is 0."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "collapsed": false,
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 12,
     "slide_helper": "subslide_end"
    },
    "slide_helper": "slide_end",
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-- Some pre-defined attributes for variable temp:\n",
      "('temp.dimensions:', (u'time', u'lat', u'lon'))\n",
      "('temp.shape:', (0, 73, 144))\n",
      "('temp.dtype:', dtype('float64'))\n",
      "('temp.ndim:', 3)\n"
     ]
    }
   ],
   "source": [
    "print(\"-- Some pre-defined attributes for variable temp:\")\n",
    "print(\"temp.dimensions:\", temp.dimensions)\n",
    "print(\"temp.shape:\", temp.shape)\n",
    "print(\"temp.dtype:\", temp.dtype)\n",
    "print(\"temp.ndim:\", temp.ndim)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 12,
     "slide_type": "subslide"
    },
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Writing data\n",
    "\n",
    "To write data a netCDF variable object, just treat it like a numpy array and assign values to a slice."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": false,
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 14
    },
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "('-- Wrote data, temp.shape is now ', (3, 73, 144))\n",
      "('-- Min/Max values:', 280.00283562143028, 329.99987991477548)\n"
     ]
    }
   ],
   "source": [
    "nlats = len(lat_dim); nlons = len(lon_dim); ntimes = 3\n",
    "# Write latitudes, longitudes.\n",
    "# Note: the \":\" is necessary in these \"write\" statements\n",
    "lat[:] = -90. + (180./nlats)*np.arange(nlats) # south pole to north pole\n",
    "lon[:] = (180./nlats)*np.arange(nlons) # Greenwich meridian eastward\n",
    "# create a 3D array of random numbers\n",
    "data_arr = np.random.uniform(low=280,high=330,size=(ntimes,nlats,nlons))\n",
    "# Write the data.  This writes the whole 3D netCDF variable all at once.\n",
    "temp[:,:,:] = data_arr  # Appends data along unlimited dimension\n",
    "print(\"-- Wrote data, temp.shape is now \", temp.shape)\n",
    "# read data back from variable (by slicing it), print min and max\n",
    "print(\"-- Min/Max values:\", temp[:,:,:].min(), temp[:,:,:].max())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 15,
     "slide_helper": "subslide_end"
    },
    "slide_helper": "slide_end",
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "- You can just treat a netCDF Variable object like a numpy array and assign values to it.\n",
    "- Variables automatically grow along unlimited dimensions (unlike numpy arrays)\n",
    "- The above writes the whole 3D variable all at once,  but you can write it a slice at a time instead.\n",
    "\n",
    "Let's add another time slice....\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": false,
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 15,
     "slide_type": "subslide"
    },
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "('-- Wrote more data, temp.shape is now ', (4, 73, 144))\n"
     ]
    }
   ],
   "source": [
    "# create a 2D array of random numbers\n",
    "data_slice = np.random.uniform(low=280,high=330,size=(nlats,nlons))\n",
    "temp[3,:,:] = data_slice   # Appends the 4th time slice\n",
    "print(\"-- Wrote more data, temp.shape is now \", temp.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 17
    },
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "Note that we have not yet written any data to the time variable.  It automatically grew as we appended data along the time dimension to the variable `temp`, but the data is missing."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": false,
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 18,
     "slide_helper": "subslide_end"
    },
    "slide_helper": "slide_end",
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<type 'netCDF4._netCDF4.Variable'>\n",
      "float64 time(time)\n",
      "    units: hours since 1800-01-01\n",
      "    long_name: time\n",
      "unlimited dimensions: time\n",
      "current shape = (4,)\n",
      "filling on, default _FillValue of 9.96920996839e+36 used\n",
      "\n",
      "(<class 'numpy.ma.core.MaskedArray'>, masked_array(data = [-- -- -- --],\n",
      "             mask = [ True  True  True  True],\n",
      "       fill_value = 9.96920996839e+36)\n",
      ")\n"
     ]
    }
   ],
   "source": [
    "print(time)\n",
    "times_arr = time[:]\n",
    "print(type(times_arr),times_arr)  # dashes indicate masked values (where data has not yet been written)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 18,
     "slide_type": "subslide"
    },
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "Let's add write some data into the time variable.  \n",
    "\n",
    "- Given a set of datetime instances, use date2num to convert to numeric time values and then write that data to the variable."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "collapsed": false,
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 20,
     "slide_helper": "subslide_end"
    },
    "slide_helper": "slide_end",
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[datetime.datetime(2014, 10, 1, 0, 0), datetime.datetime(2014, 10, 2, 0, 0), datetime.datetime(2014, 10, 3, 0, 0), datetime.datetime(2014, 10, 4, 0, 0)]\n",
      "(array([ 1882440.,  1882464.,  1882488.,  1882512.]), u'hours since 1800-01-01')\n",
      "[datetime.datetime(2014, 10, 1, 0, 0) datetime.datetime(2014, 10, 2, 0, 0)\n",
      " datetime.datetime(2014, 10, 3, 0, 0) datetime.datetime(2014, 10, 4, 0, 0)]\n"
     ]
    }
   ],
   "source": [
    "from datetime import datetime\n",
    "from netCDF4 import date2num,num2date\n",
    "# 1st 4 days of October.\n",
    "dates = [datetime(2014,10,1,0),datetime(2014,10,2,0),datetime(2014,10,3,0),datetime(2014,10,4,0)]\n",
    "print(dates)\n",
    "times = date2num(dates, time.units)\n",
    "print(times, time.units) # numeric values\n",
    "time[:] = times\n",
    "# read time data back, convert to datetime instances, check values.\n",
    "print(num2date(time[:],time.units))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 20,
     "slide_type": "subslide"
    },
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Closing a netCDF file\n",
    "\n",
    "It's **important** to close a netCDF file you opened for writing:\n",
    "\n",
    "- flushes buffers to make sure all data gets written\n",
    "- releases memory resources used by open netCDF files"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": false,
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 22,
     "slide_helper": "subslide_end"
    },
    "slide_helper": "slide_end",
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<type 'netCDF4._netCDF4.Dataset'>\n",
      "root group (NETCDF4_CLASSIC data model, file format HDF5):\n",
      "    title: My model data\n",
      "    dimensions(sizes): lat(73), lon(144), time(4)\n",
      "    variables(dimensions): float32 \u001b[4mlat\u001b[0m(lat), float32 \u001b[4mlon\u001b[0m(lon), float64 \u001b[4mtime\u001b[0m(time), float64 \u001b[4mtemp\u001b[0m(time,lat,lon)\n",
      "    groups: \n",
      "\n",
      "Dataset is closed!\n"
     ]
    }
   ],
   "source": [
    "# first print the Dataset object to see what we've got\n",
    "print(ncfile)\n",
    "# close the Dataset.\n",
    "ncfile.close(); print('Dataset is closed!')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 22,
     "slide_type": "subslide"
    },
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Advanced features\n",
    "\n",
    "So far we've only exercised features associated with the old netCDF version 3 data model.  netCDF version 4 adds a lot of new functionality that comes with the more flexible HDF5 storage layer.  \n",
    "\n",
    "Let's create a new file with `format='NETCDF4'` so we can try out some of these features."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "collapsed": false,
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 25,
     "slide_helper": "subslide_end"
    },
    "slide_helper": "slide_end",
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<type 'netCDF4._netCDF4.Dataset'>\n",
      "root group (NETCDF4 data model, file format HDF5):\n",
      "    dimensions(sizes): \n",
      "    variables(dimensions): \n",
      "    groups: \n",
      "\n"
     ]
    }
   ],
   "source": [
    "ncfile = netCDF4.Dataset('data/new2.nc','w',format='NETCDF4')\n",
    "print(ncfile)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 25,
     "slide_type": "subslide"
    },
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Creating Groups\n",
    "\n",
    "netCDF version 4 added support for organizing data in hierarchical groups.\n",
    "\n",
    "- analagous to directories in a filesystem. \n",
    "- Groups serve as containers for variables, dimensions and attributes, as well as other groups. \n",
    "- A `netCDF4.Dataset` creates a special group, called the 'root group', which is similar to the root directory in a unix filesystem. \n",
    "\n",
    "- groups are created using the [`createGroup`](http://unidata.github.io/netcdf4-python/netCDF4.Dataset-class.html#createGroup) method.\n",
    "- takes a single argument (a string, which is the name of the Group instance).  This string is used as a key to access the group instances in the `groups` dictionary.\n",
    "\n",
    "Here we create two groups to hold data for two different model runs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": false,
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 27,
     "slide_helper": "subslide_end"
    },
    "slide_helper": "slide_end",
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "('model_run1', <type 'netCDF4._netCDF4.Group'>\n",
      "group /model_run1:\n",
      "    dimensions(sizes): \n",
      "    variables(dimensions): \n",
      "    groups: \n",
      ")\n",
      "('model_run2', <type 'netCDF4._netCDF4.Group'>\n",
      "group /model_run2:\n",
      "    dimensions(sizes): \n",
      "    variables(dimensions): \n",
      "    groups: \n",
      ")\n"
     ]
    }
   ],
   "source": [
    "grp1 = ncfile.createGroup('model_run1')\n",
    "grp2 = ncfile.createGroup('model_run2')\n",
    "for grp in ncfile.groups.items():\n",
    "    print(grp)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 27,
     "slide_type": "subslide"
    },
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "Create some dimensions in the root group."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "collapsed": false,
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 29
    },
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "lat_dim = ncfile.createDimension('lat', 73)     # latitude axis\n",
    "lon_dim = ncfile.createDimension('lon', 144)    # longitude axis\n",
    "time_dim = ncfile.createDimension('time', None) # unlimited axis (can be appended to)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 30
    },
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "Now create a variable in grp1 and grp2.  The library will search recursively upwards in the group tree to find the dimensions (which in this case are defined one level up).\n",
    "\n",
    "- These variables are create with **zlib compression**, another nifty feature of netCDF 4. \n",
    "- The data are automatically compressed when data is written to the file, and uncompressed when the data is read.  \n",
    "- This can really save disk space, especially when used in conjunction with the [**least_significant_digit**](http://unidata.github.io/netcdf4-python/netCDF4.Dataset-class.html#createVariable) keyword argument, which causes the data to be quantized (truncated) before compression.  This makes the compression lossy, but more efficient."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": false,
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 31,
     "slide_helper": "subslide_end"
    },
    "slide_helper": "slide_end",
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "('model_run1', <type 'netCDF4._netCDF4.Group'>\n",
      "group /model_run1:\n",
      "    dimensions(sizes): \n",
      "    variables(dimensions): float64 \u001b[4mtemp\u001b[0m(time,lat,lon)\n",
      "    groups: \n",
      ")\n",
      "('model_run2', <type 'netCDF4._netCDF4.Group'>\n",
      "group /model_run2:\n",
      "    dimensions(sizes): \n",
      "    variables(dimensions): float64 \u001b[4mtemp\u001b[0m(time,lat,lon)\n",
      "    groups: \n",
      ")\n"
     ]
    }
   ],
   "source": [
    "temp1 = grp1.createVariable('temp',np.float64,('time','lat','lon'),zlib=True)\n",
    "temp2 = grp2.createVariable('temp',np.float64,('time','lat','lon'),zlib=True)\n",
    "for grp in ncfile.groups.items():  # shows that each group now contains 1 variable\n",
    "    print(grp)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 31,
     "slide_type": "subslide"
    },
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "##Creating a variable with a compound data type\n",
    "\n",
    "- Compound data types map directly to numpy structured (a.k.a 'record' arrays). \n",
    "- Structured arrays are akin to C structs, or derived types in Fortran. \n",
    "- They allow for the construction of table-like structures composed of combinations of other data types, including other compound types. \n",
    "- Might be useful for representing multiple parameter values at each point on a grid, or at each time and space location for scattered (point) data. \n",
    "\n",
    "Here we create a variable with a compound data type to represent complex data (there is no native complex data type in netCDF). \n",
    "\n",
    "- The compound data type is created with the [`createCompoundType`](http://unidata.github.io/netcdf4-python/netCDF4.Dataset-class.html#createCompoundType) method."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "collapsed": false,
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 33,
     "slide_helper": "subslide_end"
    },
    "slide_helper": "slide_end",
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<type 'netCDF4._netCDF4.Variable'>\n",
      "compound cmplx_var(time, lat, lon)\n",
      "compound data type: [('real', '<f8'), ('imag', '<f8')]\n",
      "path = /model_run1\n",
      "unlimited dimensions: time\n",
      "current shape = (1, 73, 144)\n",
      "\n",
      "(dtype([('real', '<f8'), ('imag', '<f8')]), (73, 144), (0.578177705604801, 0.18086070805676357))\n"
     ]
    }
   ],
   "source": [
    "# create complex128 numpy structured data type\n",
    "complex128 = np.dtype([('real',np.float64),('imag',np.float64)])\n",
    "# using this numpy dtype, create a netCDF compound data type object\n",
    "# the string name can be used as a key to access the datatype from the cmptypes dictionary.\n",
    "complex128_t = ncfile.createCompoundType(complex128,'complex128')\n",
    "# create a variable with this data type, write some data to it.\n",
    "cmplxvar = grp1.createVariable('cmplx_var',complex128_t,('time','lat','lon'))\n",
    "# write some data to this variable\n",
    "# first create some complex random data\n",
    "nlats = len(lat_dim); nlons = len(lon_dim)\n",
    "data_arr_cmplx = np.random.uniform(size=(nlats,nlons))+1.j*np.random.uniform(size=(nlats,nlons))\n",
    "# write this complex data to a numpy complex128 structured array\n",
    "data_arr = np.empty((nlats,nlons),complex128)\n",
    "data_arr['real'] = data_arr_cmplx.real; data_arr['imag'] = data_arr_cmplx.imag\n",
    "cmplxvar[0] = data_arr  # write the data to the variable (appending to time dimension)\n",
    "print(cmplxvar)\n",
    "data_out = cmplxvar[0] # read one value of data back from variable\n",
    "print(data_out.dtype, data_out.shape, data_out[0,0])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 33,
     "slide_type": "subslide"
    },
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "##Creating a variable with a variable-length (vlen) data type\n",
    "\n",
    "netCDF 4 has support for variable-length or \"ragged\" arrays. These are arrays of variable length sequences having the same type. \n",
    "\n",
    "- To create a variable-length data type, use the [`createVLType`](http://unidata.github.io/netcdf4-python/netCDF4.Dataset-class.html#createVLType) method.\n",
    "- The numpy datatype of the variable-length sequences and the name of the new datatype must be specified. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "collapsed": false,
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 35
    },
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "vlen_t = ncfile.createVLType(np.int64, 'phony_vlen')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 36
    },
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "A new variable can then be created using this datatype."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "collapsed": false,
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 37,
     "slide_helper": "subslide_end"
    },
    "slide_helper": "slide_end",
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "vlvar = grp2.createVariable('phony_vlen_var', vlen_t, ('time','lat','lon'))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 37,
     "slide_type": "subslide"
    },
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "Since there is no native vlen datatype in numpy, vlen arrays are represented in python as object arrays (arrays of dtype `object`). \n",
    "\n",
    "- These are arrays whose elements are Python object pointers, and can contain any type of python object. \n",
    "- For this application, they must contain 1-D numpy arrays all of the same type but of varying length. \n",
    "- Fill with 1-D random numpy int64 arrays of random length between 1 and 10."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "collapsed": false,
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 39,
     "slide_helper": "subslide_end"
    },
    "slide_helper": "slide_end",
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<type 'netCDF4._netCDF4.Variable'>\n",
      "vlen phony_vlen_var(time, lat, lon)\n",
      "vlen data type: int64\n",
      "path = /model_run2\n",
      "unlimited dimensions: time\n",
      "current shape = (1, 73, 144)\n",
      "\n",
      "('data =\\n', array([[[array([0, 4, 0, 9, 2, 2, 2, 4, 2]), array([7, 5, 4, 4, 9, 8, 0]),\n",
      "         array([3, 6, 6, 8, 2, 7]), ..., array([5, 0, 0, 8, 8, 1, 5, 3]),\n",
      "         array([4, 2, 7]), array([0])],\n",
      "        [array([5, 6, 6, 6, 1, 0, 7]), array([7]),\n",
      "         array([7, 5, 8, 9, 6, 9, 3]), ..., array([0, 6, 5, 4]),\n",
      "         array([7, 1, 9, 7, 7, 2]), array([1, 4, 0])],\n",
      "        [array([4, 3, 1]), array([6, 3, 9, 7, 8]), array([8]), ...,\n",
      "         array([6, 5, 8, 0]), array([0]), array([0, 9, 6, 2, 4])],\n",
      "        ..., \n",
      "        [array([8, 4, 4]), array([4, 1, 6]), array([1, 4, 2, 3, 9]), ...,\n",
      "         array([9, 1]), array([7, 2, 5, 1, 5, 8, 2]),\n",
      "         array([2, 9, 9, 1, 4, 6, 3, 5, 2])],\n",
      "        [array([4, 7, 9, 8, 2, 3, 6, 6]),\n",
      "         array([1, 4, 1, 6, 1, 1, 2, 3, 9]),\n",
      "         array([9, 5, 6, 2, 4, 3, 8, 2, 9]), ..., array([9, 5, 7]),\n",
      "         array([3, 9]), array([4, 2, 6, 9])],\n",
      "        [array([8, 9, 9, 2, 2, 8, 8, 5]), array([3]),\n",
      "         array([8, 8, 0, 2, 9, 2, 3, 0, 9]), ..., array([7]),\n",
      "         array([5, 1, 0, 6, 8, 6]), array([8, 6, 3, 6, 9, 8, 4, 2, 5])]]], dtype=object))\n"
     ]
    }
   ],
   "source": [
    "vlen_data = np.empty((nlats,nlons),object)\n",
    "for i in range(nlons):\n",
    "    for j in range(nlats):\n",
    "        size = np.random.randint(1,10,size=1) # random length of sequence\n",
    "        vlen_data[j,i] = np.random.randint(0,10,size=size)# generate random sequence\n",
    "vlvar[0] = vlen_data # append along unlimited dimension (time)\n",
    "print(vlvar)\n",
    "print('data =\\n',vlvar[:])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 39,
     "slide_type": "subslide"
    },
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "Close the Dataset and examine the contents with ncdump."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "collapsed": false,
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 41,
     "slide_helper": "subslide_end"
    },
    "slide_helper": "slide_end",
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "netcdf new2 {\r\n",
      "types:\r\n",
      "  compound complex128 {\r\n",
      "    double real ;\r\n",
      "    double imag ;\r\n",
      "  }; // complex128\r\n",
      "  int64(*) phony_vlen ;\r\n",
      "dimensions:\r\n",
      "\tlat = 73 ;\r\n",
      "\tlon = 144 ;\r\n",
      "\ttime = UNLIMITED ; // (1 currently)\r\n",
      "\r\n",
      "group: model_run1 {\r\n",
      "  variables:\r\n",
      "  \tdouble temp(time, lat, lon) ;\r\n",
      "  \tcomplex128 cmplx_var(time, lat, lon) ;\r\n",
      "  } // group model_run1\r\n",
      "\r\n",
      "group: model_run2 {\r\n",
      "  variables:\r\n",
      "  \tdouble temp(time, lat, lon) ;\r\n",
      "  \tphony_vlen phony_vlen_var(time, lat, lon) ;\r\n",
      "  } // group model_run2\r\n",
      "}\r\n"
     ]
    }
   ],
   "source": [
    "ncfile.close()\n",
    "!ncdump -h data/new2.nc"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "internals": {
     "frag_helper": "fragment_end",
     "frag_number": 41,
     "slide_helper": "subslide_end",
     "slide_type": "subslide"
    },
    "slide_helper": "slide_end",
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "##Other interesting and useful projects using netcdf4-python\n",
    "\n",
    "- [Xray](http://xray.readthedocs.org/en/stable/): N-dimensional variant of the core [pandas](http://pandas.pydata.org) data structure that can operate on netcdf variables.\n",
    "- [Iris](http://scitools.org.uk/iris/): a data model to create a data abstraction layer which isolates analysis and visualisation code from data format specifics.  Uses netcdf4-python to access netcdf data (can also handle GRIB).\n",
    "- [Biggus](https://github.com/SciTools/biggus): Virtual large arrays (from netcdf variables) with lazy evaluation.\n",
    "- [cf-python](http://cfpython.bitbucket.org/): Implements the [CF](http://cfconventions.org) data model for the reading, writing and processing of data and metadata. "
   ]
  }
 ],
 "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": "2.7.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}