freshwater_discharge.ipynb

{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Freshwater discharge in southeast Greenland fjords\n",
    "\n",
    "## Purpose\n",
    "This notebook generates figures and spreadsheets of freshwater discharge for each fjord in the study. The figures show freshwater discharge (from both MAR and RACMO) derived from ice basins, from land basins, and from combined ice+land basins. The spreadsheets describe the cumulative, annual mean, and seasonal mean freshwater discharge in each fjord.\n",
    "\n",
    "## Requirements\n",
    "- `./fjord_discharge/fjord*_discharge_total.txt`\n",
    "\n",
    "If you are generating your own data rather than reproducing our analysis, see instructions in README.\n",
    "\n",
    "## User parameters\n",
    "- list of fjord ID numbers\n",
    "\n",
    "## Output\n",
    "- Figures of freshwater discharge from MAR and RACMO in ice basins, land basins, and combined, for each fjord \\\n",
    "`../figures/fjord_FWdischarge/fjord*_FWdischarge.png`\n",
    "- Spreadsheets of cumulative discharge, mean annual total discharge, and mean seasonal total discharge for all fjords \\\n",
    "`../databases/fjord_FWdischarge_cumulative.csv` \\\n",
    "`../databases/fjord_FWdischarge_annualmean.csv` \\\n",
    "`../databases/fjord_FWdischarge_seasonalmean.csv`\n",
    "\n",
    "## Code"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import pathlib"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# USER PARAMETERS\n",
    "\n",
    "# -- List of fjord ID numbers\n",
    "fjord_number_list = list(range(1,53))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# FUNCTIONS\n",
    "\n",
    "def m3s_to_Gtyr(x):\n",
    "    '''Convert m3/s to Gt/yr for secondary y-axis'''\n",
    "    y = x * (60*60*24*365 / 1E9) # (m3/s) * (60 s/min)(60 min/hr)(24 hr/d)(365 d/yr) / (1E9 m3/Gt for water)\n",
    "    return y\n",
    "\n",
    "def Gtyr_to_m3s(x):\n",
    "    '''Convert Gt/yr to m3/s for secondary y-axis'''\n",
    "    y = x * (1E9 / 60*60*24*365) # (Gt/yr) * (1E9 m3/Gt for water) / (60 s/min)(60 min/hr)(24 hr/d)(365 d/yr)\n",
    "    return y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Example plot of freshwater discharge for a single fjord\n",
    "\n",
    "# priority fjord numbers: 15, 18, 31, 37, 40, 43, 45, 48\n",
    "fjord_number = 1\n",
    "fjord_discharge = pd.read_csv(f'./fjord_discharge/fjord{fjord_number:02}_discharge_total.txt').set_index('time')\n",
    "fjord_discharge.index = pd.to_datetime(fjord_discharge.index)\n",
    "fjord_discharge\n",
    "\n",
    "# Initialize axes\n",
    "fig, axs = plt.subplots(nrows=1, ncols=3, figsize=(15, 5), layout='constrained')\n",
    "axs = axs.flatten()\n",
    "\n",
    "# Plot discharge from ice basins\n",
    "if ('RACMO_ice' in fjord_discharge.keys()) & ('MAR_ice' in fjord_discharge.keys()):\n",
    "    fjord_discharge['RACMO_ice'].plot(ax=axs[0], color='tab:blue', label='RACMO')\n",
    "    fjord_discharge['MAR_ice'].plot(ax=axs[0], alpha=0.6, color='tab:orange', label='MAR')\n",
    "axs[0].set_ylabel('Discharge ($\\mathrm{m^3\\ s^{-1}}$)')\n",
    "axs[0].set_title(f'Fjord {fjord_number} ice basin FW discharge')\n",
    "axs[0].set_xlim(pd.to_datetime(['2015-01-01', '2020-01-01']))\n",
    "secax0 = axs[0].secondary_yaxis('right', functions=(m3s_to_Gtyr, Gtyr_to_m3s))\n",
    "\n",
    "# Plot discharge from land basins\n",
    "if ('RACMO_land' in fjord_discharge.keys()) & ('MAR_land' in fjord_discharge.keys()):\n",
    "    fjord_discharge['RACMO_land'].plot(ax=axs[1], color='tab:blue', label='RACMO')\n",
    "    fjord_discharge['MAR_land'].plot(ax=axs[1], alpha=0.6, color='tab:orange', label='MAR')\n",
    "axs[1].set_title(f'Fjord {fjord_number} land basin FW discharge')\n",
    "axs[1].set_xlim(pd.to_datetime(['2015-01-01', '2020-01-01']))\n",
    "secax1 = axs[1].secondary_yaxis('right', functions=(m3s_to_Gtyr, Gtyr_to_m3s))\n",
    "\n",
    "# Plot combined discharge\n",
    "fjord_discharge['RACMO_total'].plot(ax=axs[2], color='tab:blue', label='RACMO')\n",
    "fjord_discharge['MAR_total'].plot(ax=axs[2], alpha=0.6, color='tab:orange', label='MAR')\n",
    "axs[2].set_title(f'Fjord {fjord_number} combined basin FW discharge')\n",
    "axs[2].set_xlim(pd.to_datetime(['2015-01-01', '2020-01-01']))\n",
    "axs[2].legend()\n",
    "secax2 = axs[2].secondary_yaxis('right', functions=(m3s_to_Gtyr, Gtyr_to_m3s))\n",
    "secax2.set_ylabel('Discharge ($\\mathrm{Gt\\ a^{-1}}$)')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create plots of freshwater discharge from MAR and RACMO in ice basins, land basins, and combined, for each fjord\n",
    "\n",
    "for fjord_number in fjord_number_list:\n",
    "    # get fjord data\n",
    "    fjord_discharge = pd.read_csv(f'./fjord_discharge/fjord{fjord_number:02}_discharge_total.txt').set_index('time')\n",
    "    fjord_discharge.index = pd.to_datetime(fjord_discharge.index)\n",
    "\n",
    "    # initialize plot\n",
    "    fig, axs = plt.subplots(nrows=1, ncols=3, figsize=(15, 5), layout='constrained')\n",
    "    axs = axs.flatten()\n",
    "\n",
    "    # Plot discharge from ice basins\n",
    "    if ('RACMO_ice' in fjord_discharge.keys()) & ('MAR_ice' in fjord_discharge.keys()):\n",
    "        fjord_discharge['RACMO_ice'].plot(ax=axs[0], color='tab:blue', label='RACMO')\n",
    "        fjord_discharge['MAR_ice'].plot(ax=axs[0], alpha=0.6, color='tab:orange', label='MAR')\n",
    "    axs[0].set_ylabel('Discharge ($\\mathrm{m^3\\ s^{-1}}$)')\n",
    "    axs[0].set_title(f'Fjord {fjord_number} ice basin FW discharge')\n",
    "    axs[0].set_xlim(pd.to_datetime(['2015-01-01', '2020-01-01']))\n",
    "    secax0 = axs[0].secondary_yaxis('right', functions=(m3s_to_Gtyr, Gtyr_to_m3s))\n",
    "\n",
    "    # Plot discharge from land basins\n",
    "    if ('RACMO_land' in fjord_discharge.keys()) & ('MAR_land' in fjord_discharge.keys()):\n",
    "        fjord_discharge['RACMO_land'].plot(ax=axs[1], color='tab:blue', label='RACMO')\n",
    "        fjord_discharge['MAR_land'].plot(ax=axs[1], alpha=0.6, color='tab:orange', label='MAR')\n",
    "    axs[1].set_title(f'Fjord {fjord_number} land basin FW discharge')\n",
    "    axs[1].set_xlim(pd.to_datetime(['2015-01-01', '2020-01-01']))\n",
    "    secax1 = axs[1].secondary_yaxis('right', functions=(m3s_to_Gtyr, Gtyr_to_m3s))\n",
    "\n",
    "    # Plot combined discharge\n",
    "    fjord_discharge['RACMO_total'].plot(ax=axs[2], color='tab:blue', label='RACMO')\n",
    "    fjord_discharge['MAR_total'].plot(ax=axs[2], alpha=0.6, color='tab:orange', label='MAR')\n",
    "    axs[2].set_title(f'Fjord {fjord_number} combined basin FW discharge')\n",
    "    axs[2].set_xlim(pd.to_datetime(['2015-01-01', '2020-01-01']))\n",
    "    axs[2].legend()\n",
    "    secax2 = axs[2].secondary_yaxis('right', functions=(m3s_to_Gtyr, Gtyr_to_m3s))\n",
    "    secax2.set_ylabel('Discharge ($\\mathrm{Gt\\ a^{-1}}$)')\n",
    "\n",
    "    # save output\n",
    "    pathlib.Path('../figures/fjord_FWdischarge/').mkdir(parents=True, exist_ok=True)\n",
    "    plt.savefig(f'../figures/fjord_FWdischarge/fjord{fjord_number:02}_FWdischarge.png', bbox_inches='tight', dpi=300, facecolor='white')\n",
    "    plt.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create spreadsheets of cumulative discharge, mean discharge, and seasonal discharge for each fjord\n",
    "\n",
    "cumulative_fjord_discharge = pd.DataFrame(\n",
    "    index=fjord_number_list, \n",
    "    columns=['MAR_land', 'MAR_ice', 'MAR_total', 'RACMO_land', 'RACMO_ice', 'RACMO_total'])\n",
    "mean_fjord_discharge = pd.DataFrame(\n",
    "    index=fjord_number_list, \n",
    "    columns=['MAR_land', 'MAR_ice', 'MAR_total', 'RACMO_land', 'RACMO_ice', 'RACMO_total'])\n",
    "seasonal_fjord_discharge = pd.DataFrame(\n",
    "    index=fjord_number_list, \n",
    "    columns=['WIN_MAR_total', 'SPR_MAR_total', 'SUM_MAR_total', 'AUT_MAR_total', 'WIN_RACMO_total', 'SPR_RACMO_total', 'SUM_RACMO_total', 'AUT_RACMO_total'])\n",
    "\n",
    "def get_season(date):\n",
    "    if date.month in [1, 2]:\n",
    "        season = 'WIN'\n",
    "    elif date.month in [3, 4, 5]:\n",
    "        season = 'SPR'\n",
    "    elif date.month in [6, 7, 8]:\n",
    "        season = 'SUM'\n",
    "    elif date.month in [9, 10, 11]:\n",
    "        season = 'AUT'\n",
    "    elif date.month in [12]:\n",
    "        season = 'WIN'\n",
    "    return season\n",
    "\n",
    "for fjord_number in fjord_number_list:\n",
    "    # get fjord data\n",
    "    fjord_discharge = pd.read_csv(f'./fjord_discharge/fjord{fjord_number:02}_discharge_total.txt').set_index('time')\n",
    "    fjord_discharge.index = pd.to_datetime(fjord_discharge.index)\n",
    "\n",
    "    # convert m3/s to m3/day\n",
    "    fjord_discharge_daily = fjord_discharge * 86400\n",
    "    \n",
    "    # sum daily fjord data (m3) over entire time series to get cumulative discharge from fjord and add it to summary dataframe\n",
    "    fjord_discharge_sum = fjord_discharge_daily.sum().round(decimals=3)\n",
    "    cumulative_fjord_discharge.loc[fjord_number] = fjord_discharge_sum\n",
    "\n",
    "    # get mean annual discharge (m3) from fjord and add it to summary dataframe\n",
    "    fjord_discharge_mean = fjord_discharge_daily.groupby(fjord_discharge_daily.index.year).sum().mean().round(decimals=3)\n",
    "    mean_fjord_discharge.loc[fjord_number] = fjord_discharge_mean\n",
    "\n",
    "    # get mean seasonal discharge (m3) from fjord and add it to summary dataframe\n",
    "    fjord_discharge_daily['Season'] = fjord_discharge_daily.apply(lambda row: get_season(row.name), axis=1)\n",
    "    fds = fjord_discharge_daily.groupby([fjord_discharge_daily.index.year, 'Season']).sum().groupby('Season').mean().round(decimals=3)\n",
    "    seasonal_fjord_discharge.loc[fjord_number] = [\n",
    "        fds.loc['WIN']['MAR_total'], fds.loc['SPR']['MAR_total'], fds.loc['SUM']['MAR_total'], fds.loc['AUT']['MAR_total'],\n",
    "        fds.loc['WIN']['RACMO_total'], fds.loc['SPR']['RACMO_total'], fds.loc['SUM']['RACMO_total'], fds.loc['AUT']['RACMO_total']]\n",
    "\n",
    "# save each to CSV\n",
    "pathlib.Path('../databases/').mkdir(parents=True, exist_ok=True)\n",
    "cumulative_fjord_discharge.to_csv(\n",
    "    '../databases/fjord_FWdischarge_cumulative.csv', \n",
    "    index_label='fjord', \n",
    "    header=['MAR_land (m3)', 'MAR_ice (m3)', 'MAR_total (m3)', 'RACMO_land (m3)', 'RACMO_ice (m3)', 'RACMO_total (m3)'])\n",
    "mean_fjord_discharge.to_csv(\n",
    "    '../databases/fjord_FWdischarge_annualmean.csv', \n",
    "    index_label='fjord', \n",
    "    header=['MAR_land (m3)', 'MAR_ice (m3)', 'MAR_total (m3)', 'RACMO_land (m3)', 'RACMO_ice (m3)', 'RACMO_total (m3)'])\n",
    "seasonal_fjord_discharge.to_csv(\n",
    "    '../databases/fjord_FWdischarge_seasonalmean.csv', \n",
    "    index_label='fjord', \n",
    "    header=['MAR_WIN (m3)', 'MAR_SPR (m3)', 'MAR_SUM (m3)', 'MAR_AUT (m3)', 'RACMO_WIN (m3)', 'RACMO_SPR (m3)', 'RACMO_SUM (m3)', 'RACMO_AUT (m3)'])\n"
   ]
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "31f2aee4e71d21fbe5cf8b01ff0e069b9275f58929596ceb00d14d90e3e16cd6"
  },
  "kernelspec": {
   "display_name": "Python 3.8.10 64-bit",
   "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": "3.8.10"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}