JRA55_preparation_TROPICS_input_for_perturbations_1972-2017.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table style=\"border: 2px solid white;\">\n",
       "<tr>\n",
       "<td style=\"vertical-align: top; border: 0px solid white\">\n",
       "<h3 style=\"text-align: left;\">Client</h3>\n",
       "<ul style=\"text-align: left; list-style: none; margin: 0; padding: 0;\">\n",
       "  <li><b>Scheduler: </b>tcp://127.0.0.1:34649</li>\n",
       "  <li><b>Dashboard: </b><a href='/proxy/8787/status' target='_blank'>/proxy/8787/status</a></li>\n",
       "</ul>\n",
       "</td>\n",
       "<td style=\"vertical-align: top; border: 0px solid white\">\n",
       "<h3 style=\"text-align: left;\">Cluster</h3>\n",
       "<ul style=\"text-align: left; list-style:none; margin: 0; padding: 0;\">\n",
       "  <li><b>Workers: </b>4</li>\n",
       "  <li><b>Cores: </b>16</li>\n",
       "  <li><b>Memory: </b>68.72 GB</li>\n",
       "</ul>\n",
       "</td>\n",
       "</tr>\n",
       "</table>"
      ],
      "text/plain": [
       "<Client: 'tcp://127.0.0.1:34649' processes=4 threads=16, memory=68.72 GB>"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Purpose: Preparing the perturbation forcing for the tropics simulation\n",
    "#          i.e. climatological forcing everywhere, except over the +/-25°N/S\n",
    "#          tropics with tapering off across 5° latitude at each side from\n",
    "#          sin(pi/2) to sin(pi)\n",
    "#          -> only doing it for surface air temperature (SAT) for now\n",
    "# ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #                         \n",
    "#     Maurice F. Huguenin, m.huguenin-virchaux@unsw.edu.au, 11. 05. 2020      #                                                \n",
    "# ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #\n",
    "\n",
    "# % preamble ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #                         \n",
    "save = '/g/data/e14/mv7494/HeatDist_figures/'\n",
    "\n",
    "import xarray as xr # for working with labelled multi-dimensional arrays                                                                \n",
    "import numpy as np # for numerical operations                                                           \n",
    "import cmocean as cm\n",
    "from datetime import datetime # package for printing time in plot                                       \n",
    "import string # for subplot annotation                                                                  \n",
    "import matplotlib.pyplot as plt # for matlab-like plotting                                              \n",
    "import cartopy.crs as ccrs # for maps                                                                   \n",
    "import os.path # to check if file exists\n",
    "from matplotlib import gridspec\n",
    "import cartopy.mpl.ticker as cticker\n",
    "import string # for subplot annotation\n",
    "import cartopy.mpl.ticker as cticker\n",
    "from matplotlib import ticker\n",
    "\n",
    "import cartopy.feature as cfeature\n",
    "\n",
    "import nc_time_axis\n",
    "import cftime\n",
    "\n",
    "# for displaying multiple output in one cell\n",
    "from IPython.display import display as dp\n",
    "\n",
    "# colours                                                                                               \n",
    "ned_blue = [.12,.27,.54] # blue colour from the Dutch flag                                              \n",
    "ned_red  = [.68,.10,.15] # red  \"    \"    \"    \"    \"    \"\n",
    "myblue = [.19, .21, .58]\n",
    "\n",
    "from dask.distributed import Client\n",
    "client = Client(n_workers=4)\n",
    "client"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 411 ms, sys: 154 ms, total: 565 ms\n",
      "Wall time: 3.69 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "# load JRA55-do data and create the masks\n",
    "# climatological mask = everywhere 1962-1972 data except the tropics where there are zero values\n",
    "# tropics mask = 1972-2018 data only in the tropics with zero values everywhere else\n",
    "\n",
    "base = '/g/data/qv56/replicas/input4MIPs/CMIP6/OMIP/MRI/MRI-JRA55-do-1-4-0/atmos/3hrPt/tas/gr/v20190429/'\n",
    "fname = 'tas_input4MIPs_atmosphericState_OMIP_MRI-JRA55-do-1-4-0_gr_'\n",
    "\n",
    "#a = np.linspace(1958,2017,60) # years 1958, 1959, 1960, ..., 2017\n",
    "a = np.linspace(1958,1958,1)\n",
    "\n",
    "for year in a:                 # loop over all years\n",
    "    year = str(int(year))\n",
    "    ds1 = xr.open_mfdataset(base + fname + year + '*',combine = 'by_coords', chunks = {'time': 1},decode_times = True) # open dataset\n",
    "#    ds2 = xr.open_mfdataset(base + fname + '2017*',combine = 'by_coords', chunks = {'time': 1},decode_times = True) # open dataset\n",
    "\n",
    "ds = ds1.tas[1,:,:] # plot everywhere zero values except over land where we have nan\n",
    "ds = ds.where(np.isnan(ds),0)\n",
    "\n",
    "# now the next step is tapering from np.sin(pi) to np.sin(0)\n",
    "smoothing = [1, 0.8, 0.6, 0.4, .2, 0]\n",
    "\n",
    "u = ds.where((ds.lat>46) | (ds.lat<-46) | np.isnan(ds), smoothing[5]) \n",
    "u = u.where((u.lat>45) | (u.lat<-45) | np.isnan(u),smoothing[4]) \n",
    "u = u.where((u.lat>44) | (u.lat<-44) | np.isnan(u),smoothing[3])\n",
    "u = u.where((u.lat>43) | (u.lat<-43) | np.isnan(u),smoothing[2])\n",
    "u = u.where((u.lat>42) | (u.lat<-42) | np.isnan(u),smoothing[1])\n",
    "u = u.where((u.lat>41) | (u.lat<-41) | np.isnan(u),smoothing[0])\n",
    "iaf_mask = u.where((u.lat>40) | (u.lat<-40),1)\n",
    "#iaf_mask.plot()\n",
    "\n",
    "# save to netcdf\n",
    "#iaf_mask.to_netcdf(save + 'basin_mask_tropics_pert.nc', encoding=tropics.encoding.update({'zlib': True, 'complevel': 4}))\n",
    "\n",
    "# next step, do the oppsoite for the climatological part of the forcing\n",
    "# it needs to be 1 everywhere, except in the tropics and tapering from zero to 1 to the 25° latitude border with the tropics\n",
    "\n",
    "ds = ds1.tas[1,:,:]\n",
    "ds = ds.where(np.isnan(ds),1) # plot everywhere zero values except over land where we have nan\n",
    "\n",
    "u = ds.where((ds.lat>46) | (ds.lat<-46) | np.isnan(ds),smoothing[0]) \n",
    "u = u.where((u.lat>45) | (u.lat<-45) | np.isnan(u),smoothing[1])\n",
    "u = u.where((u.lat>44) | (u.lat<-44) | np.isnan(u),smoothing[2])\n",
    "u = u.where((u.lat>43) | (u.lat<-43) | np.isnan(u),smoothing[3])\n",
    "u = u.where((u.lat>42) | (u.lat<-42) | np.isnan(u),smoothing[4])\n",
    "u = u.where((u.lat>41) | (u.lat<-41) | np.isnan(u),smoothing[5])\n",
    "rdf_mask = u.where((u.lat>40) | (u.lat<-40) | np.isnan(u),0)\n",
    "\n",
    "\n",
    "# save back into xarray structure and save as .netcdf file  \n",
    "ds3 = xr.Dataset({'tas': ds, \n",
    "                  'iaf_mask_tropics': (('lat', 'lon'), iaf_mask),\n",
    "                  'iaf_mask_extratropics': (('lat', 'lon'), rdf_mask)}) \n",
    "\n",
    "\n",
    "os.system('rm -r /g/data/e14/mv7494/HeatDist_input/JRA55-do_basin_masks_for_tropics_extratropics_runs.nc')\n",
    "ds3.to_netcdf('/g/data/e14/mv7494/HeatDist_input/JRA55-do_basin_masks_for_tropics_extratropics_runs.nc') \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Plotting routine for masks"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/g/data3/hh5/public/apps/miniconda3/envs/analysis3-20.04/lib/python3.7/site-packages/ipykernel_launcher.py:73: UserWarning: This figure includes Axes that are not compatible with tight_layout, so results might be incorrect.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 2.94 s, sys: 139 ms, total: 3.08 s\n",
      "Wall time: 3.02 s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/g/data3/hh5/public/apps/miniconda3/envs/analysis3-20.04/lib/python3.7/site-packages/IPython/core/pylabtools.py:132: UserWarning: This figure includes Axes that are not compatible with tight_layout, so results might be incorrect.\n",
      "  fig.canvas.print_figure(bytes_io, **kw)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x648 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%%time\n",
    "ds3 = xr.open_dataset('/g/data/e14/mv7494/HeatDist_input/JRA55-do_basin_masks_for_tropics_extratropics_runs.nc')  \n",
    "plot_range=[-90,90]; ticks = np.linspace(0,1,21)\n",
    "# initialise figure\n",
    "fig = plt.figure(figsize=(16, 9),tight_layout=True,facecolor='w',edgecolor='k')\n",
    "gs = gridspec.GridSpec(3,1) \n",
    "# ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n",
    "def fancy_plot(axes_nr):\n",
    "    '''\n",
    "    This function creates my specifics for the global plot, i.e.\n",
    "    coastlines, land colour, gridlines and labels (30°lat, 60°lon)\n",
    "    input:\n",
    "        - axes_nr -> number of subplot, i.e. ax1 for subplot(2,2,1)\n",
    "    '''\n",
    "    axes_nr.coastlines(resolution = '50m') # add coast lines\n",
    "    # ticklabels on left and bottom only\n",
    "    axes_nr.set_xticks([0,90,180,270,360], crs=ccrs.PlateCarree())   \n",
    "    axes_nr.set_yticks([-90,-45,0,45,90], crs=ccrs.PlateCarree())\n",
    "    lon_formatter = cticker.LongitudeFormatter()\n",
    "    lat_formatter = cticker.LatitudeFormatter()\n",
    "    axes_nr.xaxis.set_major_formatter(lon_formatter) # set format of x-axis\n",
    "    axes_nr.yaxis.set_major_formatter(lat_formatter) # \"       \"       \"\n",
    "    axes_nr.xaxis.label.set_visible(False);axes_nr.yaxis.label.set_visible(False)\n",
    "    axes_nr.tick_params(axis='both', which='major', labelsize=16) # fontsize of lon/lat labels\n",
    "\n",
    "# --- Tropics ---------------------------------------------------------- #\n",
    "ax1 = plt.subplot(gs[0,0], projection=ccrs.PlateCarree(central_longitude=180))\n",
    "ax1.set_extent([1, 360, plot_range[0], plot_range[1]], crs=ccrs.PlateCarree())\n",
    "\n",
    "p2 = ds3.iaf_mask_tropics.plot.contourf(levels=ticks, cmap='Reds',\n",
    "                                 add_colorbar=False,extend='neither',transform=ccrs.PlateCarree())\n",
    "\n",
    "plt.title('Tropics Mask', fontsize=16) # title\n",
    "ax1.text(0, 1.05,string.ascii_lowercase[0]+')', transform=ax1.transAxes, size=16, weight='normal') # bold letter a at the top\n",
    "fancy_plot(ax1); ax1.set_xticklabels([], minor=False)\n",
    "\n",
    "# --- Extratropics -------------------------------------------------------- #\n",
    "ax1 = plt.subplot(gs[1,0], projection=ccrs.PlateCarree(central_longitude=180))\n",
    "ax1.set_extent([1, 360, plot_range[0], plot_range[1]], crs=ccrs.PlateCarree())\n",
    "\n",
    "p2 = ds3.iaf_mask_tropics.plot.contourf(levels=ticks, cmap='Reds',\n",
    "                                 add_colorbar=False,extend='neither',transform=ccrs.PlateCarree())\n",
    "\n",
    "plt.title('Extratropics Mask', fontsize=16) # title\n",
    "ax1.text(0, 1.05,string.ascii_lowercase[1]+')', transform=ax1.transAxes, size=16, weight='normal') # bold letter a at the top\n",
    "fancy_plot(ax1)\n",
    "\n",
    "# # --- Tropics -------------------------------------------------------- #\n",
    "# ax1 = plt.subplot(gs[2,0], projection=ccrs.PlateCarree(central_longitude=0))\n",
    "# ax1.set_extent([1, 360, plot_range[0], plot_range[1]], crs=ccrs.PlateCarree())\n",
    "\n",
    "# #p2 = iaf_mask.plot.contourf(levels=ticks, cmap='Reds',\n",
    "# #                                 add_colorbar=False,extend='neither',transform=ccrs.PlateCarree())\n",
    "\n",
    "# plt.title('No Mask', fontsize=16) # title\n",
    "# ax1.text(0, 1.05,string.ascii_lowercase[2]+')', transform=ax1.transAxes, size=16, weight='normal') # bold letter a at the top\n",
    "# fancy_plot(ax1); ax1.set_yticklabels([], minor=False)\n",
    "\n",
    "# add colour bar\n",
    "cax = fig.add_axes([0.38, 0.3, 0.25, 0.015]) # position: [x0, y0, width, height]\n",
    "cb = plt.colorbar(p2, cax = cax, shrink=.25, orientation='horizontal'); \n",
    "cb.set_label(label='Amplitude', size=16)\n",
    "cb.ax.tick_params(labelsize=16); tick_locator = ticker.MaxNLocator(nbins=6)\n",
    "cb.locator = tick_locator;cb.update_ticks()  \n",
    "\n",
    "\n",
    "# --------------------------------------------------------------------------- #\n",
    "# --- saving as 300 dpi .PNG image in specified folder ---------------------- #\n",
    "# --- end of script --------------------------------------------------------- #\n",
    "plt.savefig(save + 'JRA55_basin_mask_preparation_for_forcing_tropics', \n",
    "            dpi=300, facecolor='w',\n",
    "            edgecolor='w', orientation='landscape', papertype=None, \n",
    "            format=None, transparent=False, bbox_inches='tight', \n",
    "            pad_inches=0.1, metadata=None)\n",
    "# --- end of script --------------------------------------------------------- #\n",
    "\n",
    "# Wall time: 52.3 s"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Here the heavy lifting: combine masks with IAF and RDF forcing and save in specified folder\n",
    "##### -> (IAF data * iaf_mask) + (RDF data * rdf_mask)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Year 1972 & uas: leap day added\n",
      "0:01:18.049533\n",
      "~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n",
      "\n",
      "Year 1973 & uas: no adjustment\n",
      "0:00:59.702119\n",
      "~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n",
      "\n",
      "#########################################\n",
      "#           Variable finished           #\n",
      "#########################################\n",
      "\n",
      "Year 1972 & vas: leap day added\n",
      "0:01:27.230224\n",
      "~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n",
      "\n",
      "Year 1973 & vas: no adjustment\n",
      "0:00:57.148540\n",
      "~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n",
      "\n",
      "#########################################\n",
      "#           Variable finished           #\n",
      "#########################################\n",
      "\n",
      "Year 1972 & tas: leap day added\n",
      "0:01:03.644476\n",
      "~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n",
      "\n",
      "Year 1973 & tas: no adjustment\n",
      "0:01:07.659469\n",
      "~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n",
      "\n",
      "#########################################\n",
      "#           Variable finished           #\n",
      "#########################################\n",
      "\n",
      "Year 1972 & huss: leap day added\n",
      "0:01:09.755565\n",
      "~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n",
      "\n",
      "Year 1973 & huss: no adjustment\n",
      "0:01:03.134643\n",
      "~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n",
      "\n",
      "#########################################\n",
      "#           Variable finished           #\n",
      "#########################################\n",
      "\n",
      "Year 1972 & rlds: leap day added\n",
      "0:01:05.871261\n",
      "~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n",
      "\n",
      "Year 1973 & rlds: no adjustment\n",
      "0:00:58.371098\n",
      "~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n",
      "\n",
      "#########################################\n",
      "#           Variable finished           #\n",
      "#########################################\n",
      "\n",
      "Year 1972 & rsds: leap day added\n",
      "0:00:59.357629\n",
      "~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n",
      "\n",
      "Year 1973 & rsds: no adjustment\n",
      "0:01:08.663119\n",
      "~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n",
      "\n",
      "#########################################\n",
      "#           Variable finished           #\n",
      "#########################################\n",
      "\n",
      "Year 1972 & prra: leap day added\n",
      "0:01:12.651923\n",
      "~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n",
      "\n",
      "Year 1973 & prra: no adjustment\n",
      "0:01:11.756110\n",
      "~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n",
      "\n",
      "#########################################\n",
      "#           Variable finished           #\n",
      "#########################################\n",
      "\n",
      "Year 1972 & psl: leap day added\n",
      "0:01:27.605499\n",
      "~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n",
      "\n",
      "Year 1973 & psl: no adjustment\n",
      "0:00:59.772329\n",
      "~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n",
      "\n",
      "#########################################\n",
      "#           Variable finished           #\n",
      "#########################################\n",
      "\n",
      "Year 1972 & prsn: leap day added\n",
      "0:00:53.557422\n",
      "~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n",
      "\n",
      "Year 1973 & prsn: no adjustment\n",
      "0:00:57.023740\n",
      "~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n",
      "\n",
      "#########################################\n",
      "#           Variable finished           #\n",
      "#########################################\n",
      "\n"
     ]
    },
    {
     "ename": "IndexError",
     "evalue": "list index out of range",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mIndexError\u001b[0m                                Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-6-7fd2f77bacae>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m     60\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     61\u001b[0m         \u001b[0;31m# get correct subfolder, either 'atmos', 'land' or 'landIce'\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 62\u001b[0;31m         \u001b[0;32mif\u001b[0m \u001b[0mvar\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m'friver'\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mforcing\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m'1-3'\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     63\u001b[0m             \u001b[0mdomain\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m'ocean/'\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     64\u001b[0m         \u001b[0;32melif\u001b[0m \u001b[0mvar\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m'friver'\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mforcing\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m'1-4-0'\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mIndexError\u001b[0m: list index out of range"
     ]
    }
   ],
   "source": [
    "# combining RDF and IAF forcing into a netcdf file\n",
    "\n",
    "# --- which forcing to use, either JRA55-do-1-3 or JRA55-do-1-4-0\n",
    "forcing = '1-3'\n",
    "\n",
    "\n",
    "# to accommodate the different forcing which has different naming conventions\n",
    "if forcing == '1-3':\n",
    "    base = '/g/data/qv56/replicas/input4MIPs/CMIP6/OMIP/MRI/MRI-JRA55-do-1-3/'\n",
    "    fname = '_input4MIPs_atmosphericState_OMIP_MRI-JRA55-do-1-3_gn_' \n",
    "    version = 'gn/v20180412/'\n",
    "    hrPt_timestep = ['01010000','12312230']\n",
    "    hr_timestep = ['01010130','01010000']\n",
    "    day_timestep = ['0101','0101']\n",
    "    experiment = 'EXP_tropics_40SN_JRA55-do-1-3/'\n",
    "elif forcing == '1-4-0':\n",
    "    base = '/g/data/qv56/replicas/input4MIPs/CMIP6/OMIP/MRI/MRI-JRA55-do-1-4-0/'\n",
    "    fname = '_input4MIPs_atmosphericState_OMIP_MRI-JRA55-do-1-4-0_gr_'\n",
    "    version = 'gr/v20190429/'\n",
    "    hrPt_timestep = ['01010000','12312100']\n",
    "    hr_timestep = ['01010130','12312230']\n",
    "    day_timestep = ['0101','1231']\n",
    "    experiment = 'EXP_tropics_40SN_JRA55-do-1-4-0/'   \n",
    "      \n",
    "# variables of the JRA55 input\n",
    "# friver = water flux into sea water from rivers          # rlds = surface downwelling longwave radiation\n",
    "# huss =  near-surface specific humidity                  # rsds = surface downwelling shortwave radiation\n",
    "# licalvf = land ice calving flux                         # tas = near-surface air temperature\n",
    "# prra = rainfall flux                                    # uas = eastward near-surface wind\n",
    "# prsn = snowfall flux                                    # vas = northward near-surface wind\n",
    "# psl = sea level pressure \n",
    "\n",
    "#var = ['friver','huss','licalvf,','prra','prsn','psl','rlds','rsds','tas','uas','vas'] # full list of input variables\n",
    "\n",
    "var = ['uas','vas','tas','huss','rlds','rsds','prra','psl','prsn']#,'friver'] # only eight of the variables I used in my Msc\n",
    "\n",
    "\n",
    "a = np.linspace(1972,2018,47) # years 1958, 1959, 1960, ..., 2017\n",
    "\n",
    "\n",
    "# repeating the 1962-1971 period 4.6 times to build one repeating list\n",
    "b = np.linspace(1962,1971,10)\n",
    "b = np.concatenate((b, b, b, b, [1962,1963,1964,1965,1966,1967]), axis=0)\n",
    "\n",
    "# ok, this is great. Now I have two lists, one cycling through the 1962-1971 climatological period and the other going\n",
    "# from 1972-2018 with the IAF perturbation in the tropics\n",
    "\n",
    "# testing\n",
    "#dp(a)\n",
    "#dp(b)\n",
    "\n",
    "startTime = datetime.now() # start clock     \n",
    "\n",
    "for i in range(0,10):                 # loop over the JRA55 variables,          range(0,10):\n",
    "    for year in range(0,46):          # loop over all years from 1972-2018,    range(0,46):\n",
    "        year_a = str(int(a[year]))\n",
    "        year_aa = str(int(a[year+1]))\n",
    "        year_b = str(int(b[year]))       \n",
    "        startTime = datetime.now()   # start clock     \n",
    "\n",
    "        # get correct subfolder, either 'atmos', 'land' or 'landIce'\n",
    "        if var[i] == 'friver' and forcing == '1-3':\n",
    "            domain = 'ocean/'\n",
    "        elif var[i] == 'friver' and forcing == '1-4-0':\n",
    "            domain = 'land/'\n",
    "        elif var[i] == 'licalvf':\n",
    "            domain = 'landIce/'\n",
    "        else:\n",
    "            domain = 'atmos/'\n",
    "        \n",
    "        # get correct resolution of forcing field, either '3hr', '3hrPt' or 'day'\n",
    "        if var[i] in ['rsds','rlds','prra','prsn']:\n",
    "            res = '3hr/'\n",
    "        elif var[i] in ['psl','tas','huss','uas','vas']:\n",
    "            res = '3hrPt/'\n",
    "        else:\n",
    "            res = 'day/'\n",
    "            \n",
    "        # load in data for interannual IAF forcing \n",
    "        ds1 = xr.open_mfdataset(base + domain + res + var[i] + '/' + version + var[i] + fname + year_a + '*',combine = 'by_coords', \n",
    "                                chunks = {'time': 1},decode_times = True) # open dataset  \n",
    "        #ds1 = ds1.sel(time = year_a + '-01-01') # only select subset of data\n",
    "    \n",
    "\n",
    "        # load in data for climatological RDF forcing \n",
    "        ds2 = xr.open_mfdataset(base + domain + res + var[i] + '/' + version + var[i] + fname + year_b + '*',combine = 'by_coords', \n",
    "                                chunks = {'time': 1},decode_times = True) # open dataset\n",
    "        \n",
    "\n",
    "        # leap day adjustment, either add or remove 29th of Feb. so the clim. and IAF forcing match\n",
    "        # the adjustment takes place in the ds2, i.e. climatological data\n",
    "        # see also file presentation20200511.ppt in the autumn semester 2020 folder\n",
    "        if year_a in ['1972','1976','1980','1992','1996','2000','2012', '2016']:\n",
    "            # in these yeas I have to add a leap day in the clim. files as a copy of 28th of Feb\n",
    "            #print(year_b)  \n",
    "\n",
    "            # --- in this section I add a leap day\n",
    "\n",
    "            new_attrs = {**ds2.time.attrs} # keep the time attributes for later use\n",
    "            day_length=ds2.time.size//365 # day_length = 8, 6-hourly data\n",
    "\n",
    "            # extract the 28th of February\n",
    "            ds2_29 = ds2.isel(time=slice((31+27)*day_length,(31+28)*day_length)) \n",
    "\n",
    "            # day_shift is 1 as the unit of the time array is day here.\n",
    "            # But it could be more if time was in hours for example.\n",
    "            # See the time attributes for the unit.\n",
    "            day_shift = ds2.time.values[day_length] - ds2.time.values[0] \n",
    "            ds2_29.time.values[:] += day_shift\n",
    "\n",
    "            # Create a 1 day \"hole\" in original time coordinate:\n",
    "            ds2.time.values[-day_length:] += day_shift\n",
    "\n",
    "            ds2 = xr.concat((ds2,ds2_29),dim=\"time\") # yes, great now the time vector has 2928 entries instead of 2920 (no leap days)\n",
    "\n",
    "            # We see the result isn't quite right as the time dimension isn't ordered. Just need a sort:\n",
    "            ds2 = ds2.sortby(ds2.time)\n",
    "\n",
    "            # Now we change the calendar to all_leap and decode the times:\n",
    "            new_attrs[\"calendar\"] = \"all_leap\"\n",
    "            time = ds2.time\n",
    "            time = time.assign_attrs(**new_attrs)\n",
    "            ds2 =ds2.assign_coords({\"time\":time})\n",
    "\n",
    "            print('Year ' + year_a + ' & ' + var[i] + ': leap day added')   \n",
    "            \n",
    "        elif year_a in ['1974', '1978', '1994', '1998', '2014']:\n",
    "            # in these yeas I have to remove the leap day in the clim. files \n",
    "            ds2 = ds2.sel(time=~((ds2.time.dt.month == 2) & (ds2.time.dt.day == 29)))\n",
    "            print('Year ' + year_a + ' & ' + var[i] + ': leap day deleted')   \n",
    "        else:\n",
    "            print('Year ' + year_a + ' & ' + var[i] + ': no adjustment')   \n",
    "        ds2['time'] = ds1.time # overwrite time vector so that the 'interannual' and 'climatological' regions have the same time vector\n",
    "\n",
    "        # iaf_mask has IAF forcing in the tropics and zeroes in the extratropics\n",
    "        # rdf_mask has RDF forcing in the extratropics and zeroes in the tropics\n",
    "        # --------------------------------------------------------------------------------------------------- #\n",
    "        ds1[var[i]] = (ds1[var[i]] * iaf_mask) + (ds2[var[i]] * rdf_mask) # combine the masks with the actual forcing\n",
    "        # --------------------------------------------------------------------------------------------------- #     \n",
    "\n",
    "        # different file names depending on the variables\n",
    "        if var[i] in ['uas','vas','tas','psl','huss']:\n",
    "            # construct filename for output file, e.g.\n",
    "            # filename = '/g/data/e14/mv7494/HeatDist_input/EXP_tropics_JRA55-do-1-3//uas/uas_input4MIPs_atmosphericState_OMIP_MRI-JRA55-do-1-3_gn_197201010000-12312230.nc'\n",
    "            filename = '/g/data/e14/mv7494/HeatDist_input/' + experiment + var[i] + '/' + var[i] + fname + year_a + hrPt_timestep[0] + '-' + year_a + hrPt_timestep[1] + '.nc'\n",
    "            if os.path.isfile(filename) == True: # check if file already exists, if yes then I can skip that iteration which speeds up my script\n",
    "                continue                \n",
    "            else: \n",
    "                ds1.to_netcdf(filename) # if file not yet exists, save array to netcdf file\n",
    "        elif var[i] in ['prra','prsn','rlds','rsds']:\n",
    "            filename = '/g/data/e14/mv7494/HeatDist_input/' + experiment + var[i] + '/' + var[i] + fname + year_a + hr_timestep[0] + '-' + year_aa + hr_timestep[1] + '.nc'\n",
    "            if os.path.isfile(filename) == True:\n",
    "                continue\n",
    "            else: \n",
    "                ds1.to_netcdf(filename)\n",
    "        elif var[i] == 'friver':\n",
    "            if forcing == '1-3':\n",
    "                filename = '/g/data/e14/mv7494/HeatDist_input/' + experiment + var[i]+ '/' + var[i] + fname + year_a + day_timestep[0] + '-' + year_aa + day_timestep[1] + '.nc'\n",
    "            else:\n",
    "                filename = '/g/data/e14/mv7494/HeatDist_input/' + experiment + var[i]+ '/' + var[i] + fname + year_a + day_timestep[0] + '-' + year_a + day_timestep[1] + '.nc'\n",
    "\n",
    "            if os.path.isfile(filename) == True:\n",
    "                continue\n",
    "            else: \n",
    "                ds1.to_netcdf(filename)\n",
    "\n",
    "        # ok this works, but is missing the lat/lon/bnds fields\n",
    "\n",
    "        print(datetime.now() - startTime) # end time of script\n",
    "        print('~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~')    \n",
    "        print('') \n",
    "    # Note: Saving one month of data takes 18 seconds\n",
    "    # ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n",
    "    print('#########################################')    \n",
    "    print('#           Variable finished           #')\n",
    "    print('#########################################')    \n",
    "    print('') \n",
    "    \n",
    "    # it works!!! takes 0:01:08.35 seconds for one variable and for one year using 48 cores and 24 workers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# ----------------------------------------------------------------------------------------------------------------------- #\n",
    "#                                                   Notes down below                                                      #\n",
    "# ----------------------------------------------------------------------------------------------------------------------- #"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [conda env:analysis3-21.04]",
   "language": "python",
   "name": "conda-env-analysis3-21.04-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": "3.8.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}