consolidated version of the cmorisation process for DWD

dwd_consolidated.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "220e5b04-c4b4-48ec-9e70-e4e493d88680",
   "metadata": {},
   "source": [
    "DWD Amon\n",
    "========\n",
    "\n",
    "This notebook contains consolidated code from the earlier example, and demonstrates how to produce a number of monthly atmospheric (`Amon`) variables."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "827ac9bd-68c5-4521-9f61-df6cd05ac217",
   "metadata": {},
   "outputs": [],
   "source": [
    "import cftime\n",
    "import cmor\n",
    "import iris\n",
    "import numpy as np\n",
    "import re\n",
    "import os\n",
    "import cf_units\n",
    "\n",
    "from cmor_utils import create_cmor_json_config, generate_dataset_info, parse_filename, parse_time_unit, load_mip_table, DATASET_ROOT, BASE_TIME_UNIT, MIP_ERA, MIP_TABLES_DIR"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "1920c04b-51ff-4a1b-b60f-bdf4947afe52",
   "metadata": {},
   "outputs": [],
   "source": [
    "filename_template  = '{}_Amon_DWD_s{}_r1-16i1p1f1_gn_climatology-1991-2020.nc'\n",
    "\n",
    "variable = 'psl'\n",
    "init_year = 2021\n",
    "\n",
    "filename = filename_template.format(variable, init_year)\n",
    "dataset = iris.load(os.path.join(DATASET_ROOT, filename))[0]\n",
    "\n",
    "calendar = dataset.coord('time').units.calendar\n",
    "dataset.coord('time').convert_units(cf_units.Unit(BASE_TIME_UNIT, calendar=calendar))\n",
    "cftime_points = cftime.num2date(dataset.coord('time').points,\n",
    "                str(dataset.coord('time').units),\n",
    "                calendar=calendar)\n",
    "time_bounds = []\n",
    "for time_point in cftime_points:\n",
    "    # this needs to handle time bounds at monthly resolution\n",
    "    start_date = cftime.date2num(cftime.datetime(time_point.year, time_point.month, 1, calendar=calendar), units=BASE_TIME_UNIT, calendar=calendar)\n",
    "    if time_point.month < 12:\n",
    "        end_year = time_point.year\n",
    "        end_month = time_point.month + 1\n",
    "    else:\n",
    "        end_year = time_point.year + 1\n",
    "        end_month = 1\n",
    "    end_date = cftime.date2num(cftime.datetime(end_year, end_month, 1, calendar=calendar), units=BASE_TIME_UNIT, calendar=calendar)\n",
    "    time_bounds.append((start_date, end_date,))\n",
    "dataset.coord('time').bounds = time_bounds\n",
    "dataset.coord('longitude').guess_bounds()\n",
    "dataset.coord('latitude').guess_bounds()\n",
    "time_coord = dataset.coord('time')\n",
    "lon_coord = dataset.coord('longitude')\n",
    "lat_coord = dataset.coord('latitude')\n",
    "realization_coord = dataset.coord('realization')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "01c73ffb-8001-46de-9eb4-abfecc6a6558",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "\u001b[2;34;47mC Traceback:\n",
      "In function: cmor_close_variable\n",
      "! \u001b[0m\n",
      "\n",
      "\u001b[1;34;47m!!!!!!!!!!!!!!!!!!!!!!!!!\n",
      "!\n",
      "! Warning: while closing variable 0 (pslAnom, table Amon)\n",
      "! we noticed you wrote 0 time steps for the variable,\n",
      "! but its time axis 0 (time) has 122 time steps\n",
      "!\n",
      "!!!!!!!!!!!!!!!!!!!!!!!!!\u001b[0m\n",
      "\n"
     ]
    }
   ],
   "source": [
    "mip_table, institution_id, ensemble_label, grid_label, climatology, initialization_index, physics_index, forcing_index = parse_filename(filename)\n",
    "activity_id = \"DCPP\"\n",
    "calendar = calendar\n",
    "experiment_id = \"dcppB-forecast\"\n",
    "source_id = \"ICON-ESM-LR\"\n",
    "dataset_info = generate_dataset_info(\n",
    "    mip_era=MIP_ERA,\n",
    "    activity_id=activity_id,\n",
    "    calendar=calendar,\n",
    "    ensemble_label=ensemble_label,\n",
    "    experiment_id=experiment_id,\n",
    "    institution_id=institution_id,\n",
    "    source_id=source_id,\n",
    "    source_type='AOGCM',\n",
    "    grid='grid',\n",
    "    grid_label=grid_label,\n",
    "    nominal_resolution='100 km',\n",
    "    forcing_index=forcing_index,\n",
    "    initialization_index=initialization_index,\n",
    "    physics_index=physics_index,\n",
    "    realization_index=1\n",
    ")\n",
    "config_filepath = 'CMOR_input_dwd.json'\n",
    "create_cmor_json_config(dataset_info, config_filepath)\n",
    "cmor.setup(inpath=MIP_TABLES_DIR, netcdf_file_action=cmor.CMOR_REPLACE, create_subdirectories=False)\n",
    "cmor.dataset_json(config_filepath)\n",
    "table = os.path.join(MIP_TABLES_DIR, '{}_{}.json'.format(MIP_ERA, mip_table))\n",
    "cmor.load_table(table)\n",
    "\n",
    "mip_table_json = load_mip_table(MIP_TABLES_DIR, MIP_ERA, mip_table)\n",
    "\n",
    "time = cmor.axis(table_entry='time', units=str(time_coord.units), coord_vals=time_coord.points, cell_bounds=time_coord.bounds)\n",
    "\n",
    "base_unit_year, base_unit_month, base_unit_day = parse_time_unit()\n",
    "\n",
    "date1 = cftime.date2num(cftime.datetime(init_year, 1, 1, calendar=calendar), units=BASE_TIME_UNIT, calendar=calendar)\n",
    "date2 = cftime.date2num(cftime.datetime(base_unit_year, base_unit_month, base_unit_day, calendar=calendar), units=BASE_TIME_UNIT, calendar=calendar)\n",
    "date_offset = date1 - date2\n",
    "\n",
    "reftime = cmor.axis(table_entry=\"reftime1\", units=str(time_coord.units), coord_vals=np.array([date_offset]))\n",
    "latitude = cmor.axis(table_entry=\"latitude\", units=str(lat_coord.units), coord_vals=lat_coord.points, cell_bounds=lat_coord.bounds)\n",
    "longitude = cmor.axis(table_entry=\"longitude\", units=str(lon_coord.units), coord_vals=lon_coord.points, cell_bounds=lon_coord.bounds)\n",
    "realization = cmor.axis(table_entry=\"realization\", units=\"1\", coord_vals=realization_coord.points)\n",
    "\n",
    "data = dataset.data\n",
    "\n",
    "if variable == 'tas':\n",
    "    height2m = cmor.axis(table_entry=\"height2m\", units=\"m\", coord_vals=np.array((2.0,)))\n",
    "    expanded = np.ma.expand_dims(data, axis=[4,5])\n",
    "    axis_ids = [realization, time, latitude, longitude, reftime, height2m]\n",
    "else:\n",
    "    expanded = np.ma.expand_dims(data, axis=[4])\n",
    "    axis_ids = [realization, time, latitude, longitude, reftime]\n",
    "\n",
    "variableAnom = '{}Anom'.format(variable)\n",
    "var_id = cmor.variable(table_entry=variableAnom, axis_ids=axis_ids, units=mip_table_json['variable_entry'][variableAnom]['units'])\n",
    "cmor.set_deflate(var_id, True, True, 1) # enabling netCDF4 compression\n",
    "cmor.write(var_id, expanded)\n",
    "filename = cmor.close(var_id, file_name=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "2311728c-f78d-4e18-a142-bef3531a48a5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ".//pslAnom_Amon_ICON-ESM-LR_dcppB-forecast_r1-16i1p1f1_gn_202111-203112.nc\n"
     ]
    }
   ],
   "source": [
    "print(filename)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e7620308-8081-40ae-85ca-ff997ac2b9fd",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}