7 Basic xarray - opening multifile datasets.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Opening multifile datasets"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In many cases large datasets are spead over multiple files, so no individual file becomes too large, and operating on a subset of the data is easier without having to read the whole dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "xarray can open multiple files as a single xarray `Dataset` with the `open_mfdataset` command. Pass it either a list of files to open, or a wildcard string that matches the desired files. \n",
    "\n",
    "This is useful as it can skip wasteful steps of creating intermediate datasets which are aggregations of existing data. Just load it all up and select out what is required."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Start with the required imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import xarray"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Wildcards only work when accessing files on a local filesystem, for opendap it is necessary to  pass a list of urls. Below a list of urls are created from the individual filenames which are joined to a prefix."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['http://dapds00.nci.org.au/thredds/dodsC/rr3/CMIP5/output1/CSIRO-BOM/ACCESS1-3/historical/day/atmos/day/r1i1p1/latest/tas/tas_day_ACCESS1-3_historical_r1i1p1_18500101-18741231.nc',\n",
       " 'http://dapds00.nci.org.au/thredds/dodsC/rr3/CMIP5/output1/CSIRO-BOM/ACCESS1-3/historical/day/atmos/day/r1i1p1/latest/tas/tas_day_ACCESS1-3_historical_r1i1p1_18750101-18991231.nc',\n",
       " 'http://dapds00.nci.org.au/thredds/dodsC/rr3/CMIP5/output1/CSIRO-BOM/ACCESS1-3/historical/day/atmos/day/r1i1p1/latest/tas/tas_day_ACCESS1-3_historical_r1i1p1_19000101-19241231.nc']"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "files = ['tas_day_ACCESS1-3_historical_r1i1p1_18500101-18741231.nc', \n",
    "        'tas_day_ACCESS1-3_historical_r1i1p1_18750101-18991231.nc',\n",
    "        'tas_day_ACCESS1-3_historical_r1i1p1_19000101-19241231.nc']\n",
    "prefix = 'http://dapds00.nci.org.au/thredds/dodsC/rr3/CMIP5/output1/CSIRO-BOM/ACCESS1-3/historical/day/atmos/day/r1i1p1/latest/tas/'\n",
    "urls = [prefix+file for file in files]\n",
    "urls"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This list is then passed to `open_mfdataset`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "ds = xarray.open_mfdataset(urls)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The resulting dataset is pretty big, the `nbytes` attribute gives the size of the dataset in bytes. In this case it is over 3GB. The `tas` variable has 27393 daily time steps. It doesn't take a long time to load the dataset, as it is only metadata, but loading the underlying data could be time consuming depending on how much needs to be read."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3.198847672"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ds.nbytes/1000**3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<xarray.Dataset>\n",
       "Dimensions:    (bnds: 2, lat: 145, lon: 192, time: 27393)\n",
       "Coordinates:\n",
       "  * lat        (lat) float64 -90.0 -88.75 -87.5 -86.25 ... 86.25 87.5 88.75 90.0\n",
       "  * lon        (lon) float64 0.0 1.875 3.75 5.625 ... 352.5 354.4 356.2 358.1\n",
       "    height     float64 1.5\n",
       "  * time       (time) datetime64[ns] 1850-01-01T12:00:00 ... 1924-12-31T12:00:00\n",
       "Dimensions without coordinates: bnds\n",
       "Data variables:\n",
       "    time_bnds  (time, bnds) datetime64[ns] dask.array<shape=(27393, 2), chunksize=(9131, 2)>\n",
       "    lat_bnds   (time, lat, bnds) float64 dask.array<shape=(27393, 145, 2), chunksize=(9131, 145, 2)>\n",
       "    lon_bnds   (time, lon, bnds) float64 dask.array<shape=(27393, 192, 2), chunksize=(9131, 192, 2)>\n",
       "    tas        (time, lat, lon) float32 dask.array<shape=(27393, 145, 192), chunksize=(9131, 145, 192)>\n",
       "Attributes:\n",
       "    institution:                     CSIRO (Commonwealth Scientific and Indus...\n",
       "    institute_id:                    CSIRO-BOM\n",
       "    experiment_id:                   historical\n",
       "    source:                          ACCESS1.3 2011. Atmosphere: AGCM v1.0 (N...\n",
       "    model_id:                        ACCESS1.3\n",
       "    forcing:                         GHG, Oz, SA, Sl, Vl, BC, OC, (GHG = CO2,...\n",
       "    parent_experiment_id:            piControl\n",
       "    parent_experiment_rip:           r1i1p1\n",
       "    branch_time:                     90945.0\n",
       "    contact:                         The ACCESS wiki: http://wiki.csiro.au/co...\n",
       "    history:                         CMIP5 compliant file produced from raw A...\n",
       "    references:                      See http://wiki.csiro.au/confluence/disp...\n",
       "    initialization_method:           1\n",
       "    physics_version:                 1\n",
       "    tracking_id:                     d9824016-bde8-4883-b111-9385b3d023cd\n",
       "    version_number:                  v20130227\n",
       "    product:                         output\n",
       "    experiment:                      historical\n",
       "    frequency:                       day\n",
       "    creation_date:                   2013-02-26T01:40:45Z\n",
       "    Conventions:                     CF-1.4\n",
       "    project_id:                      CMIP5\n",
       "    table_id:                        Table day (01 February 2012) b6353e99198...\n",
       "    title:                           ACCESS1.3 model output prepared for CMIP...\n",
       "    parent_experiment:               pre-industrial control\n",
       "    modeling_realm:                  atmos\n",
       "    realization:                     1\n",
       "    cmor_version:                    2.8.0\n",
       "    DODS_EXTRA.Unlimited_Dimension:  time"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ds"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Note also that extra information in the data variables section\n",
    "\n",
    "`dask.array<shape=(27393, 145, 192), chunksize=(9131, 145, 192)>`\n",
    "\n",
    "Dask is beyond the scope of this introduction, but xarray has assigned a dask chunking to the data variables as they are split across files. In particular the `time` dimension is split into three with 9131 values per file.\n",
    "\n",
    "Regardless of chunking, this data set can be used as if it were a single file for the purposes of manipulations and calculations."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now it is possible to specify a spatial subset containing Australia, using `sel`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<xarray.DataArray 'tas' (time: 27393, lat: 37, lon: 27)>\n",
       "dask.array<shape=(27393, 37, 27), dtype=float32, chunksize=(9131, 37, 27)>\n",
       "Coordinates:\n",
       "  * lat      (lat) float64 -50.0 -48.75 -47.5 -46.25 ... -8.75 -7.5 -6.25 -5.0\n",
       "  * lon      (lon) float64 110.6 112.5 114.4 116.2 ... 153.8 155.6 157.5 159.4\n",
       "    height   float64 1.5\n",
       "  * time     (time) datetime64[ns] 1850-01-01T12:00:00 ... 1924-12-31T12:00:00\n",
       "Attributes:\n",
       "    standard_name:     air_temperature\n",
       "    long_name:         Near-Surface Air Temperature\n",
       "    units:             K\n",
       "    cell_methods:      time: mean\n",
       "    cell_measures:     area: areacella\n",
       "    associated_files:  baseURL: http://cmip-pcmdi.llnl.gov/CMIP5/dataLocation..."
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tas_aus = ds.tas.sel(lat=slice(-50,-5), lon=slice(110,160))\n",
    "tas_aus"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Confirm it is correct by plotting the first time point"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.QuadMesh at 0x7f6941f53d30>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 648x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "tas_aus.isel(time=0).plot(size=6)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So `tas_aus` is a 75 year daily temperature dataset of Australia. Any calculation can now be performed, e.g. analysing variability in space, time, season ... there are many possibilities"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [conda env:analysis3-19.07]",
   "language": "python",
   "name": "conda-env-analysis3-19.07-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}