calculate_permeability_sediment_mixtures.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Calculate permeability of sand-clay mixtures\n",
    "\n",
    "Calculate permeability of sediment mixtures (ie mixtures of clay and granular material (sand, silt) and compare with emasured values. The calculation of permeability uses a series of equations by Luijendijk and Gleeson (2015, Geofluids 15, or Crustal permeability book chapter 2016) ([link](https://onlinelibrary.wiley.com/doi/full/10.1111/gfl.12115))."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Import modules"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [],
   "source": [
    "import string\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import os\n",
    "import matplotlib.pyplot as pl"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Select dataset:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [],
   "source": [
    "# file containing porosity & permeability data\n",
    "# uncomment (remove the # sign) one of the filenames below:\n",
    "\n",
    "# pure sand (Fountainbleau sandstone, Bourbie & Zinsner 1985):\n",
    "#fn = 'data/porperm_Bourbie1985_Fontainebleau_sandstone.csv'\n",
    "\n",
    "# pure clay, compilation of clay permeability data from different data sources (Olsen 1966; Mesri & Olson 1971; Al-Tabbaa & Wood 1987; Vasseur et al. 1995)\n",
    "#fn = 'data/porperm_compilation_clays.csv'\n",
    "\n",
    "# lab-mixed kaolinite & sand by Knoll (1996):\n",
    "#fn = 'data/porperm_kaoline_sand_mix_Knoll1996.csv'\n",
    "\n",
    "# mixed sediments from the Roer Valley Graben:\n",
    "fn = 'data/porperm_data_well_AST02_with_spec_surf.csv'\n",
    "\n",
    "# example dataset using a sample from IODP leg 190:\n",
    "#fn = 'data/example_dataset_seafloor_sediments_with_spec_surf.csv'\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Define some parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "# percolation threshold for porosity, used for calculating effective porosity for Kozeny-Carman equation\n",
    "phi_threshold = 0.025\n",
    "\n",
    "# Kozeny-Carman constant\n",
    "C = 5.0\n",
    "\n",
    "# density of rock matrix and pore water\n",
    "rho_s = 2650.0\n",
    "rho_w = 1025.0\n",
    "\n",
    "# empirical constants for permeability-void ratio equation for kaolinite, illite, smectite\n",
    "# these values were calibrated to a compilation of permeability data for different clay minerals\n",
    "# see Luijendijk & Gleeson (2015, Geofluids) for more info\n",
    "clays = ['kaolinite', 'illite', 'smectite']\n",
    "k0_clays = 10 ** np.array([-16.21014136, -18.81362945, -20.92831503])\n",
    "m_clays = np.array([3.60952787, 3.57637639, 3.01368043])\n",
    "\n",
    "# note that for the Knoll (1996) dataset one should use a much higher reference (k0) permeability for kaolinite. The reason for this is unknown:\n",
    "if 'Knoll1996' in fn:\n",
    "    k0_clays[0] = 10 ** (np.log10(k0_clays[0]) + 0.78)\n",
    "\n",
    "# power mean exponent. \n",
    "# This governs if the permeability of the sediment mixture is calculated as the harmonic mean, geometric mean or arithmetic mean, \n",
    "# or something in between. Use p=-1 for harmonic mean, p=0 for geometric mean or p=1 for arithmetic mean\n",
    "# Luijendijk and Gleeson (2015) suggest that p=0 would be the best value for first order estiamtes of permeability vairation\n",
    "# however this has only been tested on a relatively small dataset, and it may be dependent on the type and size of samples, internal structure of samples,\n",
    "# depositional environments etc...\n",
    "power_mean_exponent = 0.0\n",
    "\n",
    "# colormap, see here for some other option: https://matplotlib.org/users/colormaps.html\n",
    "cmap = 'copper_r'\n",
    "\n",
    "# type of figure to save. you can choose png, svg, pdf, jpg etc.. see matplotlib savefig documentation for more info\n",
    "figure_extension = 'png'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load porosity, clay content, clay type and specific surface data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>well</th>\n",
       "      <th>sample_number</th>\n",
       "      <th>depth</th>\n",
       "      <th>0-2_micron</th>\n",
       "      <th>2-4_micron</th>\n",
       "      <th>4-8_micron</th>\n",
       "      <th>8-16_micron</th>\n",
       "      <th>16-25_micron</th>\n",
       "      <th>25-35_micron</th>\n",
       "      <th>...</th>\n",
       "      <th>kaolinite_fraction</th>\n",
       "      <th>illite_fraction</th>\n",
       "      <th>smectite_fraction</th>\n",
       "      <th>note</th>\n",
       "      <th>specific_surface_granular_fraction</th>\n",
       "      <th>ln_grainsize_mean</th>\n",
       "      <th>ln_grainsize_std</th>\n",
       "      <th>log10_grainsize_mean</th>\n",
       "      <th>log10_grainsize_std</th>\n",
       "      <th>median_grainsize</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>AST-02</td>\n",
       "      <td>79</td>\n",
       "      <td>1517.25</td>\n",
       "      <td>16.1</td>\n",
       "      <td>4.8</td>\n",
       "      <td>6.6</td>\n",
       "      <td>7.1</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2.4</td>\n",
       "      <td>...</td>\n",
       "      <td>0.15</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.35</td>\n",
       "      <td>clay composition estimated</td>\n",
       "      <td>101.068461</td>\n",
       "      <td>-10.255900</td>\n",
       "      <td>1.013547</td>\n",
       "      <td>-4.454081</td>\n",
       "      <td>1.013547</td>\n",
       "      <td>0.000063</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>AST-02</td>\n",
       "      <td>80</td>\n",
       "      <td>1517.75</td>\n",
       "      <td>13.6</td>\n",
       "      <td>5.4</td>\n",
       "      <td>12.8</td>\n",
       "      <td>13.1</td>\n",
       "      <td>9.1</td>\n",
       "      <td>4.4</td>\n",
       "      <td>...</td>\n",
       "      <td>0.15</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.35</td>\n",
       "      <td>clay composition estimated</td>\n",
       "      <td>143.549090</td>\n",
       "      <td>-10.584529</td>\n",
       "      <td>1.090895</td>\n",
       "      <td>-4.596803</td>\n",
       "      <td>1.090895</td>\n",
       "      <td>0.000030</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>AST-02</td>\n",
       "      <td>81</td>\n",
       "      <td>1518.22</td>\n",
       "      <td>56.7</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.6</td>\n",
       "      <td>1.6</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.15</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.35</td>\n",
       "      <td>clay composition estimated</td>\n",
       "      <td>24.265495</td>\n",
       "      <td>-9.832384</td>\n",
       "      <td>0.638496</td>\n",
       "      <td>-4.270150</td>\n",
       "      <td>0.638496</td>\n",
       "      <td>0.000071</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>AST-02</td>\n",
       "      <td>82</td>\n",
       "      <td>1518.75</td>\n",
       "      <td>16.2</td>\n",
       "      <td>6.4</td>\n",
       "      <td>8.1</td>\n",
       "      <td>6.8</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.15</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.35</td>\n",
       "      <td>clay composition estimated</td>\n",
       "      <td>117.839434</td>\n",
       "      <td>-10.321195</td>\n",
       "      <td>1.078734</td>\n",
       "      <td>-4.482438</td>\n",
       "      <td>1.078734</td>\n",
       "      <td>0.000067</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>AST-02</td>\n",
       "      <td>83</td>\n",
       "      <td>1519.25</td>\n",
       "      <td>16.5</td>\n",
       "      <td>6.9</td>\n",
       "      <td>12.6</td>\n",
       "      <td>10.9</td>\n",
       "      <td>8.4</td>\n",
       "      <td>4.6</td>\n",
       "      <td>...</td>\n",
       "      <td>0.15</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.35</td>\n",
       "      <td>clay composition estimated</td>\n",
       "      <td>148.004649</td>\n",
       "      <td>-10.709373</td>\n",
       "      <td>1.140843</td>\n",
       "      <td>-4.651022</td>\n",
       "      <td>1.140843</td>\n",
       "      <td>0.000031</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 45 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0    well  sample_number    depth  0-2_micron  2-4_micron  \\\n",
       "0           0  AST-02             79  1517.25        16.1         4.8   \n",
       "1           1  AST-02             80  1517.75        13.6         5.4   \n",
       "2           2  AST-02             81  1518.22        56.7         0.0   \n",
       "3           3  AST-02             82  1518.75        16.2         6.4   \n",
       "4           4  AST-02             83  1519.25        16.5         6.9   \n",
       "\n",
       "   4-8_micron  8-16_micron  16-25_micron  25-35_micron        ...         \\\n",
       "0         6.6          7.1           5.0           2.4        ...          \n",
       "1        12.8         13.1           9.1           4.4        ...          \n",
       "2         1.6          1.6           1.0           0.0        ...          \n",
       "3         8.1          6.8           5.0           2.0        ...          \n",
       "4        12.6         10.9           8.4           4.6        ...          \n",
       "\n",
       "   kaolinite_fraction  illite_fraction  smectite_fraction  \\\n",
       "0                0.15              0.5               0.35   \n",
       "1                0.15              0.5               0.35   \n",
       "2                0.15              0.5               0.35   \n",
       "3                0.15              0.5               0.35   \n",
       "4                0.15              0.5               0.35   \n",
       "\n",
       "                         note  specific_surface_granular_fraction  \\\n",
       "0  clay composition estimated                          101.068461   \n",
       "1  clay composition estimated                          143.549090   \n",
       "2  clay composition estimated                           24.265495   \n",
       "3  clay composition estimated                          117.839434   \n",
       "4  clay composition estimated                          148.004649   \n",
       "\n",
       "   ln_grainsize_mean  ln_grainsize_std  log10_grainsize_mean  \\\n",
       "0         -10.255900          1.013547             -4.454081   \n",
       "1         -10.584529          1.090895             -4.596803   \n",
       "2          -9.832384          0.638496             -4.270150   \n",
       "3         -10.321195          1.078734             -4.482438   \n",
       "4         -10.709373          1.140843             -4.651022   \n",
       "\n",
       "   log10_grainsize_std  median_grainsize  \n",
       "0             1.013547          0.000063  \n",
       "1             1.090895          0.000030  \n",
       "2             0.638496          0.000071  \n",
       "3             1.078734          0.000067  \n",
       "4             1.140843          0.000031  \n",
       "\n",
       "[5 rows x 45 columns]"
      ]
     },
     "execution_count": 84,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv(fn)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Calculate permeabiltiy of clay fraction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [],
   "source": [
    "def calculate_k_clay_exponential(porosity, k0, m):\n",
    "    \n",
    "    \"\"\"\n",
    "    calculate permeability as an exponential function of void ratio \n",
    "    \"\"\"\n",
    "    \n",
    "    void_ratio = porosity / (1.0 - porosity)\n",
    "    \n",
    "    return k0 * void_ratio**m"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [],
   "source": [
    "## first calculate permeability for each clay mineral:\n",
    "for clay, k0_clay, m_clay in zip(clays, k0_clays, m_clays):\n",
    "    df['k_%s' % clay] = calculate_k_clay_exponential(df['porosity'], k0_clay, m_clay)\n",
    "    \n",
    "# combine estimates and take the geometric mean:\n",
    "df['k_clay'] = 0\n",
    "for clay in clays:\n",
    "    df['k_clay'] += df['%s_fraction' % clay] * np.log10(df['k_%s' % clay])\n",
    "df['k_clay'] = 10**df['k_clay']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Calculate permeability of granular fraction using the Kozeny-Carman eq."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [],
   "source": [
    "def kozeny_carman(phi, C, Sa, rho_s, rho_w, phi_eff):\n",
    "    \n",
    "    \"\"\"\n",
    "    Kozeny-Carman equation\n",
    "    \n",
    "    \"\"\"\n",
    "    #phi_eff = phi - phi_threshold\n",
    "    rho = phi * rho_w + (1.0 - phi) * rho_s\n",
    "    \n",
    "    # calculate specific surface per volume instead of per kg    \n",
    "    Sa_vol = Sa * (1.0 - phi) * rho_s\n",
    "    \n",
    "    # calculate permeability\n",
    "    k = (1.0 / (C * Sa_vol**2)) * phi_eff**3 / (1.0 - phi_eff)**2\n",
    "    \n",
    "    return k"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [],
   "source": [
    "phi = df['porosity']\n",
    "phi_eff = df['porosity'] - phi_threshold\n",
    "Sa = df['specific_surface_granular_fraction']\n",
    "df['k_granular'] = kozeny_carman(phi, C, Sa, rho_s, rho_w, phi_eff)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Calculate the permeability of the sediment mixture"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [],
   "source": [
    "def power_mean(p, w, k1, k2):\n",
    "\n",
    "    \"\"\"\n",
    "    calculate power mean of two components\n",
    "    reduces to arithmetic mean for p=1 and harmonic mean for p=-1\n",
    "    returns the geomteric mean for p=0\n",
    "\n",
    "    \"\"\"\n",
    "    \n",
    "    if p!=0:\n",
    "        k_calc = (w*k1**p + (1-w)*k2**p)**(1.0/p)\n",
    "    \n",
    "    elif p == 0:\n",
    "        k_calc = np.exp(w*np.log(k1) + (1-w)*np.log(k2))\n",
    "        \n",
    "    # find pure clay samples\n",
    "    ind_sand = w == 0\n",
    "    k_calc[ind_sand] = k2\n",
    "    \n",
    "    # find pure sand samples\n",
    "    ind_clay = w == 1.0\n",
    "    k_calc[ind_clay] = k1\n",
    "    \n",
    "    return k_calc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "added column with calculated permeability values to .csv file:\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0    4.660193e-15\n",
       "1    8.266365e-15\n",
       "2    3.449173e-17\n",
       "3    7.881181e-15\n",
       "4    2.047126e-15\n",
       "Name: k_calculated, dtype: float64"
      ]
     },
     "execution_count": 90,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['power_mean_exponent'] = power_mean_exponent\n",
    "df['k_calculated'] = power_mean(power_mean_exponent, df['clay_fraction'], df['k_clay'], df['k_granular'])\n",
    "\n",
    "print('added column with calculated permeability values to .csv file:')\n",
    "df['k_calculated'].head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Save permeability data to a csv file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "saved csv file with calculated permeability as data/porperm_data_well_AST02_with_spec_surf_with_calculated_k.csv\n"
     ]
    }
   ],
   "source": [
    "fno = fn[:-4] + '_with_calculated_k.csv'\n",
    "df.to_csv(fno)\n",
    "print('saved csv file with calculated permeability as %s' % fno)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Make some figures of measured vs calculated permeability"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Measured vs calculated permeability"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [],
   "source": [
    "klims = (np.min(np.log10(df['permeability_measured'])), np.max(np.log10(df['permeability_measured'])))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "saving figure as  figs/porperm_data_well_AST02_with_spec_surf_measured_vs_calculated_k.png\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f36ae913a20>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axs = pl.subplots(1, 2, figsize=(10, 5))\n",
    "\n",
    "\n",
    "\n",
    "sckm = axs[0].scatter(df['porosity'], np.log10(df['permeability_measured']), c=df['clay_fraction'], \n",
    "                      vmin=0.0, vmax=1.0, cmap=cmap)\n",
    "\n",
    "sckc = axs[0].scatter(df['porosity'], np.log10(df['k_calculated']), \n",
    "                      edgecolor='gray', facecolor='None', marker='s')\n",
    "\n",
    "legs = [sckm, sckc]\n",
    "labels = [r'Measured', r'Calculated']\n",
    "axs[0].legend(legs, labels, frameon=True, loc='lower right')\n",
    "\n",
    "axs[0].set_xlabel('Porosity')\n",
    "axs[0].set_ylabel(r'Permeability (m$^2$)')\n",
    "\n",
    "#################\n",
    "# second panel:\n",
    "#################\n",
    "sc = axs[1].scatter(np.log10(df['permeability_measured']), np.log10(df['k_calculated']), c=df['clay_fraction'], \n",
    "                    vmin=0.0, vmax=1.0, cmap=cmap)\n",
    "\n",
    "l11 = axs[1].plot(klims, klims, ls='--', color='black')\n",
    "\n",
    "cb = fig.colorbar(sckm, shrink=0.5)\n",
    "cb.set_label('Clay fraction')\n",
    "\n",
    "axs[1].set_xlabel(r'Measured permeability (m$^2$)')\n",
    "axs[1].set_ylabel(r'Calculated permeability (m$^2$)')\n",
    "\n",
    "axs[1].legend([sc, l11[0]], ['Data', '1:1'], loc='lower right', frameon=True)\n",
    "\n",
    "# make the figure look nicer:\n",
    "for i, panel in enumerate(axs):\n",
    "    panel.spines['top'].set_visible(False)\n",
    "    panel.spines['right'].set_visible(False)\n",
    "    panel.text(0.03, 1.0, string.ascii_lowercase[i], \n",
    "            weight='bold', transform=panel.transAxes, ha='left', va='top', fontsize='medium')\n",
    "\n",
    "fn1 = os.path.split(fn)[-1][:-4]\n",
    "fn2 = fn1 + '_measured_vs_calculated_k.%s' % figure_extension\n",
    "fig_fn = os.path.join('figs', fn2)\n",
    "print('saving figure as ', fig_fn)\n",
    "fig.savefig(fig_fn, dpi=300)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Fig caption** *Measured vs calculated permeability using the power mean of the permeability of the granular and clay fractions*"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Normalized difference of measured permeability and the theoretical permeability of the granular and clay fractions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [],
   "source": [
    "# calculate k offset using clay composition estimates for datasets\n",
    "df['k_offset'] = \\\n",
    "    (np.log10(df['permeability_measured']) - np.log10(df['k_clay'])) / \\\n",
    "    (np.log10(df['k_granular']) - np.log10(df['k_clay']))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [],
   "source": [
    "# load permeability vs clay content model Revil (2002)\n",
    "dfr = pd.read_csv('data/simulated_k_revil1999_model.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "saving figure as  figs/porperm_data_well_AST02_with_spec_surf_normalized_k.png\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f36aeb77780>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = pl.subplots(1, 1, figsize=(8, 6))\n",
    "\n",
    "# plot data\n",
    "sc = ax.scatter(df['clay_fraction'], df['k_offset'], c=df['clay_fraction'], s=100, lw=2.0, \n",
    "                vmin=0.0, vmax=1.0, cmap=cmap, alpha=0.7)\n",
    "\n",
    "sc.set_facecolor('none')\n",
    "\n",
    "# plot theoretical geometric mean k of clay and sand fractions in panel d\n",
    "w_plot = np.arange(0.01, 0.99, 0.01)\n",
    "k0 = 1\n",
    "k1 = 10\n",
    "geom_mean = power_mean(0.0, w_plot, k0, k1)\n",
    "leg_dfg, = ax.plot(w_plot, np.log10(geom_mean), color='black', ls='-')\n",
    "leg_dfm, = ax.plot(dfr['cv'], dfr['k_normalized_Knoll1996'], color='black', ls='--')\n",
    "\n",
    "# add lines for theoretical permeability pure granular or clay components\n",
    "xmax = 1.0\n",
    "leg_k_sd = ax.axhline(y=1, color='orange', ls='-', lw=1.5, zorder=0)\n",
    "leg_k_cl = ax.axhline(y=0, color='brown', ls='-', lw=1.5, zorder=0)\n",
    "ax.text(xmax - 0.02, 0.98, 'pure sand',\n",
    "          fontsize='small', ha='right', va='top')\n",
    "ax.text(xmax - 0.02, -0.02, 'pure clay',\n",
    "          fontsize='small', ha='right', va='top')\n",
    "\n",
    "# add a colorbar\n",
    "cb = fig.colorbar(sc, shrink=0.5)\n",
    "cb.set_label('Clay fraction')\n",
    "\n",
    "ax.set_xlabel('Clay fraction')\n",
    "ax.set_ylabel('Normalized difference\\npermeability')\n",
    "\n",
    "ylim = ax.get_ylim()\n",
    "if ylim[-1] <= 1.25:\n",
    "    ax.set_ylim(ylim[0], 1.25)\n",
    "\n",
    "# add a legend:\n",
    "legs = [sc, leg_dfm, leg_dfg]\n",
    "labels = ['Data', \n",
    "          'Ideal packing model',\n",
    "          'Geometric mean model']\n",
    "legend = ax.legend(legs, labels,\n",
    "                   fontsize='small',\n",
    "                   ncol=1, numpoints=1, loc='upper right', frameon=False)\n",
    "\n",
    "#\n",
    "ax.spines['top'].set_visible(False)\n",
    "ax.spines['right'].set_visible(False)\n",
    "\n",
    "\n",
    "# save figure\n",
    "fn1 = os.path.split(fn)[-1][:-4]\n",
    "fn2 = fn1 + '_normalized_k.%s' % figure_extension\n",
    "fig_fn = os.path.join('figs', fn2)\n",
    "print('saving figure as ', fig_fn)\n",
    "fig.savefig(fig_fn, dpi=300)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Fig. caption** *Normalized permeability of the permeability dataset compared to two theoretical models, 1) the ideal packing model (Revil 1999, 2002) and 2) the geometric mean model. This is a simplified version of Fig. 5 in Luijendijk & Gleeson (2015)*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}