LRF19_regime_diagram.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Regime diagram\n",
    "\n",
    "This notebook reproduces Figure 1 of [Li et al., 2019](https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2019MS001810).\n",
    "It plots the regime diagrams in parameter spaces La$_t$ - $h/L_L$ following [Belcher et al, 2012](https://doi.org/10.1029/2012GL052932) and the extension to stable surface forcing regime.\n",
    "\n",
    "In particular, parameter space covered by JRA55-do forced GOTM runs is shown by the joint PDFs.\n",
    "Overlaied are some representative LES studies of Langmuir turbulence."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import scipy.io as sio\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Functions\n",
    "\n",
    "Auxiliary functions for plotting."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "def loglog_hist2d(xdata, ydata, xpr, ypr):\n",
    "    hist, xi, yi, c = plt.hist2d(xdata, ydata,\n",
    "                                 range=(xpr, ypr), bins=100)\n",
    "    # clean the figure\n",
    "    plt.clf()\n",
    "    # get the centers from the edges\n",
    "    xi = 0.5*(xi[0:-1]+xi[1:])\n",
    "    yi = 0.5*(yi[0:-1]+yi[1:])\n",
    "    # convert back to actual values\n",
    "    xi = 10**xi\n",
    "    yi = 10**yi\n",
    "    return hist, xi, yi\n",
    "\n",
    "def plot_regime_diagram_background_BG12(axis=None):\n",
    "    \"\"\"Plot the background of the regime diagram\n",
    "       following Fig. 3 of Belcher et al., 2012\n",
    "\n",
    "    :axis: (matplotlib.axes, optional) axis to plot figure on\n",
    "\n",
    "    \"\"\"\n",
    "    if axis is None:\n",
    "        axis = plt.gca()\n",
    "\n",
    "    # range of power\n",
    "    xpr = [-1, 1]\n",
    "    ypr = [-3, 3]\n",
    "    # range\n",
    "    xlims = [10**i for i in xpr] \n",
    "    ylims = [10**i for i in ypr] \n",
    "    # size of x and y\n",
    "    nx = 500\n",
    "    ny = 500\n",
    "    xx = np.logspace(xpr[0], xpr[1], nx)\n",
    "    yy = np.logspace(ypr[0], ypr[1], ny)\n",
    "    zz1 = np.zeros([nx, ny]) \n",
    "    zz2 = np.zeros([nx, ny]) \n",
    "    zz3 = np.zeros([nx, ny]) \n",
    "    for i in np.arange(nx):\n",
    "        for j in np.arange(ny):\n",
    "            zz1[i,j] = 2*(1-np.exp(-0.5*xx[i]))\n",
    "            zz2[i,j] = 0.22*xx[i]**(-2)\n",
    "            zz3[i,j] = 0.3*xx[i]**(-2)*yy[j]\n",
    "    zz = zz1 + zz2 + zz3\n",
    "    axis.contourf(xx, yy, np.transpose(np.log10(zz)),\n",
    "                  levels=[-0.1, 0, 0.1, 0.25, 0.5, 1, 2, 3, 4],\n",
    "                  cmap='summer', extend='both')\n",
    "    axis.contour(xx, yy, np.transpose(np.log10(zz)),\n",
    "                  levels=[-0.1, 0, 0.1, 0.25, 0.5, 1, 2, 3, 4],\n",
    "                  colors='darkgray')\n",
    "    axis.contour(xx, yy, np.transpose(zz1/zz), levels=[0.9], colors='k',\n",
    "                linestyles='-', linewidths=2)\n",
    "    axis.contour(xx, yy, np.transpose(zz2/zz), levels=[0.9], colors='k',\n",
    "                linestyles='-', linewidths=2)\n",
    "    axis.contour(xx, yy, np.transpose(zz3/zz), levels=[0.9], colors='k',\n",
    "                linestyles='-', linewidths=2)\n",
    "    axis.set_xlim(xlims)\n",
    "    axis.set_ylim(ylims)\n",
    "    axis.set_xscale('log')\n",
    "    axis.set_yscale('log')\n",
    "    axis.set_xlabel('La$_t$')\n",
    "    axis.set_ylabel('$h/L_L$')\n",
    "    axis.set_aspect(aspect=1/3)\n",
    "    axis.text(0.85, 3e-3, '0', color='k', fontsize=8, rotation=-90)\n",
    "    axis.text(1.6, 1e-2, '0.1', color='k', fontsize=8, rotation=-90)\n",
    "    axis.text(3.8, 1e-1, '0.25', color='k', fontsize=8, rotation=-90)\n",
    "    axis.text(4, 1e2, '0.5', color='k', fontsize=8, rotation=33)\n",
    "    axis.text(3.2, 3e2, '1', color='k', fontsize=8, rotation=36)\n",
    "    axis.text(0.53, 1e2, '2', color='k', fontsize=8, rotation=38)\n",
    "    axis.text(0.3, 3.1e2, '3', color='k', fontsize=8, rotation=39)\n",
    "    axis.text(0.12, 5e2, '4', color='k', fontsize=8, rotation=40)\n",
    "    axis.text(0.11, 4e-3, 'Langmuir', bbox=dict(boxstyle=\"square\",ec='k',fc='w'))\n",
    "    axis.text(3, 4e-3, 'Shear', bbox=dict(boxstyle=\"square\",ec='k',fc='w'))\n",
    "    axis.text(0.13, 1e2, 'Convection', bbox=dict(boxstyle=\"square\",ec='k',fc='w'))\n",
    "\n",
    "def plot_dist_xp(hst, xi, yi, axis=None, levels=None, filled=False, fcolors=None, **kwargs):\n",
    "    \"\"\"Plot bi-dimensional histogram. Show the contours of the\n",
    "       histogram which enclose the highest p1%, p2%, ... and pN%\n",
    "       centered distribution.\n",
    "\n",
    "    :his: (2D numpy array) bi-dimensional histogram\n",
    "    :xi: (1D numpy array) centers of x dimension\n",
    "    :yi: (1D numpy array) centers of y dimension\n",
    "    :axis: (matplotlib.axes, optional) axis to plot figure on\n",
    "    :levels: (list of float, optional) contour levels, 0.0-1.0\n",
    "    :filled: (bool) filled contour if True\n",
    "    :fcolors: (list, optional) color string or sequence of colors\n",
    "    :return: (matplotlib figure object) figure\n",
    "\n",
    "    \"\"\"\n",
    "    # use curret axis if not specified\n",
    "    if axis is None:\n",
    "        axis = plt.gca()\n",
    "    hsum = np.sum(hst)\n",
    "    hlist = -np.sort(-hst.flatten())/hsum\n",
    "    hcum = np.cumsum(hlist)\n",
    "    vl = levels\n",
    "    nv = len(vl)\n",
    "    vlev = np.zeros(nv)\n",
    "    for i in np.arange(nv):\n",
    "        ind = np.argmin(abs(hcum-vl[i]))\n",
    "        vlev[i] = hlist[ind]\n",
    "    pdfData = hst/hsum\n",
    "    pdfData[pdfData==0] = 1e-12\n",
    "    if not filled:\n",
    "        fig = axis.contour(xi, yi, np.log10(np.transpose(pdfData)),\n",
    "                           levels=np.log10(vlev[::-1]), **kwargs)\n",
    "    else:\n",
    "        if fcolors is None:\n",
    "            cmap = cm.get_cmap('bone')\n",
    "            fcolors = cmap(np.linspace(1.0, 0.0, 11)[0:nv+1])\n",
    "        else:\n",
    "            nfc = len(fcolors)\n",
    "            if nfc != nv+1:\n",
    "                raise ValueError('Length of fcolors should equal to number of levels + 1.')\n",
    "        fig = axis.contourf(xi, yi, np.log10(np.transpose(pdfData)),\n",
    "                            levels=np.log10(vlev[::-1]),\n",
    "                            colors=fcolors, extend='both', **kwargs)\n",
    "    return fig\n",
    "    \n",
    "def plot_regime_diagram_BG12(hist, xi, yi, axis=None):\n",
    "    \"\"\"Plot regime diagram following Belcher et al., 2012\n",
    "    \n",
    "    \"\"\"\n",
    "    # background following Fig. 3 of Belcher et al., 2012\n",
    "    plot_regime_diagram_background_BG12(axis=axis)\n",
    "    \n",
    "    # plot bi-dimensional histogram\n",
    "    vl = [0.3, 0.6, 0.9, 0.99]\n",
    "    plot_dist_xp(hist, xi, yi, axis=axis, levels=vl, colors='white',\n",
    "                 linestyles='-')\n",
    "\n",
    "def plot_regime_diagram_heating(hist, xi, yi, axis=None):\n",
    "    if axis is None:\n",
    "        axis = plt.gca()\n",
    "        \n",
    "    # plot bi-dimensional histogram\n",
    "    vl = [0.3, 0.6, 0.9, 0.99]\n",
    "    plot_dist_xp(hist, xi, yi, axis=axis, levels=vl, colors='black',\n",
    "                 linestyles='-')\n",
    "    # range of power\n",
    "    xpr = [-1, 1]\n",
    "    ypr = [-3, 3]\n",
    "    # range\n",
    "    xlims = [10**i for i in xpr]\n",
    "    ylims = [10**i for i in ypr]\n",
    "    axis.set_xlim(xlims)\n",
    "    axis.set_ylim(ylims)\n",
    "    axis.set_xscale('log')\n",
    "    axis.set_yscale('log')\n",
    "    axis.set_xlabel('La$_t$')\n",
    "    axis.set_ylabel('$-h/L_L$')\n",
    "    axis.set_aspect(aspect=1/3)\n",
    "    axis.text(0.11, 4e-3, 'Langmuir', bbox=dict(boxstyle=\"square\",ec='k',fc='w'))\n",
    "    axis.text(3, 4e-3, 'Shear', bbox=dict(boxstyle=\"square\",ec='k',fc='w'))\n",
    "    axis.text(0.13, 1e2, 'Stable', bbox=dict(boxstyle=\"square\",ec='k',fc='w'))\n",
    "    \n",
    "def add_scatter(la, hLL, axis=None, **kwargs):\n",
    "    if axis is None:\n",
    "        axis = plt.gca()\n",
    "    # range of power\n",
    "    xpr = [-1, 1]\n",
    "    ypr = [-3, 3]\n",
    "    # range\n",
    "    xlims = [10**i for i in xpr]\n",
    "    ylims = [10**i for i in ypr]\n",
    "    xx = np.where(la<xlims[0], xlims[0], la)\n",
    "    xx = np.where(xx>xlims[1], xlims[1], xx)\n",
    "    yy = np.where(hLL<ylims[0], ylims[0], hLL)\n",
    "    yy = np.where(yy>ylims[1], ylims[1], yy)\n",
    "    axis.scatter(xx, yy, s=12, clip_on=False, zorder=10, **kwargs)\n",
    "\n",
    "def add_range(xrange, yrange, axis=None, **kwargs):\n",
    "    if axis is None:\n",
    "        axis = plt.gca()\n",
    "    # range of power\n",
    "    xpr = [-1, 1]\n",
    "    ypr = [-3, 3]\n",
    "    # range\n",
    "    xlims = [10**i for i in xpr]\n",
    "    ylims = [10**i for i in ypr]\n",
    "    xrange = np.maximum(xlims[0], xrange)\n",
    "    xrange = np.minimum(xlims[1], xrange)\n",
    "    yrange = np.maximum(ylims[0], yrange)\n",
    "    yrange = np.minimum(ylims[1], yrange)\n",
    "    axis.plot(xrange, yrange, '--', linewidth=2., clip_on=False, zorder=9, **kwargs)\n",
    "\n",
    "def add_mean_and_range(xm, xmin, xmax, ym, ymin, ymax, axis=None, **kwargs):\n",
    "    if axis is None:\n",
    "        axis = plt.gca()\n",
    "    # range of power\n",
    "    xpr = [-1, 1]\n",
    "    ypr = [-3, 3]\n",
    "    # range\n",
    "    xlims = [10**i for i in xpr]\n",
    "    ylims = [10**i for i in ypr]\n",
    "    xm = np.maximum(xlims[0], xm)\n",
    "    xm = np.minimum(xlims[1], xm)\n",
    "    xmin = np.maximum(xlims[0], xmin)\n",
    "    xmin = np.minimum(xlims[1], xmin)\n",
    "    xmax = np.maximum(xlims[0], xmax)\n",
    "    xmax = np.minimum(xlims[1], xmax)\n",
    "    ym = np.maximum(ylims[0], ym)\n",
    "    ym = np.minimum(ylims[1], ym)\n",
    "    ymin = np.maximum(ylims[0], ymin)\n",
    "    ymin = np.minimum(ylims[1], ymin)\n",
    "    ymax = np.maximum(ylims[0], ymax)\n",
    "    ymax = np.minimum(ylims[1], ymax)\n",
    "    axis.scatter(xm, ym, s=12, marker='s', clip_on=False, zorder=10, **kwargs)\n",
    "    ndim = xm.size\n",
    "    for i in np.arange(ndim):\n",
    "        axis.plot([xmin[i], xmax[i]], [ym[i], ym[i]], linewidth=1.5,\n",
    "                  clip_on=False, zorder=10, **kwargs)\n",
    "        axis.plot([xm[i], xm[i]], [ymin[i], ymax[i]], linewidth=1.5,\n",
    "                  clip_on=False, zorder=10, **kwargs)\n",
    "        \n",
    "def add_legend(axis, xloc, yloc, color, text):\n",
    "    axis.scatter(xloc, yloc, s=16, color=color, transform=axis.transAxes)\n",
    "    axis.text(xloc+0.015, yloc, text, fontsize=9,\n",
    "              transform=axis.transAxes, color='k',\n",
    "              va='center')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### $h/L_L$ and La$_t$ in some previous LES studies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# McWilliams et al., 1997\n",
    "c_msm97 = 'navy'\n",
    "msm97_la = np.array([0.3, 1000])\n",
    "msm97_ustar = np.ones(msm97_la.size)*6.1e-3\n",
    "msm97_B0 = np.ones(msm97_la.size)*5/4200/1000*2e-4*9.8\n",
    "msm97_mld = np.ones(msm97_la.size)*35\n",
    "msm97_hLL = msm97_la**2*msm97_B0*msm97_mld/msm97_ustar**3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Li et al., 2005\n",
    "# h/LL = Ho * \\beta * H / 4\n",
    "c_l05 = 'skyblue'\n",
    "l05_la_range_s1 = [0.2, 10]\n",
    "l05_hLL_s1 = 0.0\n",
    "l05_la_s2 = 0.34\n",
    "l05_hLL_range_s2 = [0, 100]\n",
    "# surface heating cases\n",
    "l05_ho_s3 = np.array([0.25, 0.45])\n",
    "l05_hiLL_s3 = l05_ho_s3*8.4/4\n",
    "l05_hLL_s3 = l05_hiLL_s3/(1+3.*l05_hiLL_s3) \n",
    "l05_la_s3 = np.ones(l05_hLL_s3.size)*0.34"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Harcourt and D'Asaro 2008\n",
    "c_hd08 = 'orange'\n",
    "hd08_la_s1 = np.array([0.331, 0.341, 0.351, 0.360, 0.366, 0.375, 0.381, 0.387,\n",
    "                       0.306, 0.316, 0.325, 0.332, 0.338, 0.346, 0.351, 0.357,\n",
    "                       0.291, 0.301, 0.309, 0.316, 0.322, 0.329, 0.333, 0.339,\n",
    "                       0.270, 0.278, 0.286, 0.292, 0.298, 0.304, 0.308, 0.313])\n",
    "hd08_la_s2 = np.array([0.275, 0.288, 0.300, 0.310, 0.320, 0.331, 0.339, 0.347])\n",
    "hd08_la_s3 = np.array([0.404, 0.417, 0.435, 0.364, 0.364, 0.364, 0.364])\n",
    "hd08_la_s4 = np.array([0.270, 0.278, 0.286, 0.292, 0.298, 0.304, 0.308, 0.313])\n",
    "hd08_hLL_s1 = np.ones(hd08_la_s1.size)*1e-8\n",
    "hd08_hLL_s2 = np.ones(hd08_la_s2.size)*1e-8\n",
    "hd08_hLL_s3 = np.ones(hd08_la_s3.size)*1e-8\n",
    "hd08_hLL_s4 = np.ones(hd08_la_s4.size)*1e-8"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Grant and Belcher, 2009\n",
    "c_gb09 = 'royalblue'\n",
    "gb09_la = np.array([0.15, 0.2, 0.3, 0.4, 0.8, 1.5])\n",
    "gb09_hLL = np.ones(gb09_la.size)*1e-8"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Van Roekel et al., 2012\n",
    "c_vr12 = 'orchid'\n",
    "vr12_la = np.array([0.18, 0.23, 0.29, 0.37, 0.49])\n",
    "vr12_us0 = np.array([0.022, 0.038, 0.063, 0.100, 0.162])\n",
    "vr12_ustar = np.ones(vr12_la.size)*5.3e-3\n",
    "vr12_mld = np.ones(vr12_la.size)*35\n",
    "vr12_B0 = np.ones(vr12_la.size)*5/4200/1000*2e-4*9.8\n",
    "vr12_hLL = vr12_la**2*vr12_B0*vr12_mld/vr12_ustar**3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Pearson et al., 2015 (surface heating)\n",
    "c_p15 = 'firebrick'\n",
    "p15_LL_s1 = np.array([62, 93, 124, 186, 248, 372, 496])\n",
    "p15_hi_s1 = 53\n",
    "p15_hiLL_s1 = p15_hi_s1/p15_LL_s1\n",
    "p15_hLL_s1 = p15_hiLL_s1/(1+3.5*p15_hiLL_s1)\n",
    "p15_la_s1 = np.ones(p15_LL_s1.size)*0.3\n",
    "p15_LL_s2 = np.array([93, 124, 186, 248, 372])\n",
    "p15_hi_s2 = 36\n",
    "p15_hiLL_s2 = p15_hi_s2/p15_LL_s2\n",
    "p15_hLL_s2 = p15_hiLL_s2/(1+3.*p15_hiLL_s2)\n",
    "p15_la_s2 = np.ones(p15_LL_s2.size)*0.3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Min and Noh, 2004 (surface heating)\n",
    "c_mn04 = 'green'\n",
    "mn04_ustar_i = np.array([0.01, 0.01, 0.01, 0.01, 0.005, 0.01, 0.01])\n",
    "mn04_us0_i = np.array([0.082, 0.082, 0.195, 0.05, 0.082, 0.082, 0.082])\n",
    "mn04_la_i = np.array([0.35, 0.35, 0.23, 0.45, 0.25, 0.35, 0.35])\n",
    "mn04_hi = 50\n",
    "mn04_B0_j = np.array([1, 2, 4, 6, 8, 10])*1e-7\n",
    "mn04_ni = mn04_ustar_i.size\n",
    "mn04_nj = mn04_B0_j.size\n",
    "mn04_la_arr = np.zeros([mn04_ni, mn04_nj])\n",
    "mn04_hiLL_arr = np.zeros([mn04_ni, mn04_nj])\n",
    "for j in np.arange(mn04_nj):\n",
    "    mn04_la_arr[:,j] = mn04_la_i\n",
    "    mn04_hiLL_arr[:,j] = mn04_la_i[:]**2 / mn04_ustar_i[:]**3 * mn04_B0_j[j] * mn04_hi\n",
    "mn04_la = mn04_la_arr.flatten()\n",
    "mn04_hiLL = mn04_hiLL_arr.flatten()\n",
    "mn04_hLL = mn04_hiLL/(1+3.*mn04_hiLL)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Reichl et al., 2016\n",
    "c_r16 = 'mediumpurple'\n",
    "r16_data_s1 = sio.loadmat('data/RWHGK16_log10_ParamSpace_5.mat')\n",
    "r16_data_s2 = sio.loadmat('data/RWHGK16_log10_ParamSpace_10.mat')\n",
    "r16_la_s1_mean = r16_data_s1['LAt_mean'].flatten()\n",
    "r16_la_s1_std = r16_data_s1['LAt_std'].flatten()\n",
    "r16_la_s2_mean = r16_data_s2['LAt_mean'].flatten()\n",
    "r16_la_s2_std = r16_data_s2['LAt_std'].flatten()\n",
    "r16_hLL_s1_mean = r16_data_s1['H_LL_mean'].flatten() \n",
    "r16_hLL_s1_std = r16_data_s1['H_LL_std'].flatten() \n",
    "r16_hLL_s2_mean = r16_data_s2['H_LL_mean'].flatten() \n",
    "r16_hLL_s2_std = r16_data_s2['H_LL_std'].flatten()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Li and Fox-Kemper, 2017\n",
    "c_lf17 = 'pink'\n",
    "lf17_data = sio.loadmat('data/LF17_La_hLL.mat')\n",
    "lf17_la_tmp = lf17_data['la']\n",
    "lf17_la_tmp = np.where(np.isinf(lf17_la_tmp), np.nan, lf17_la_tmp)\n",
    "lf17_la = lf17_la_tmp[~np.isnan(lf17_la_tmp)]\n",
    "lf17_hLL_tmp = lf17_data['hL']\n",
    "lf17_hLL_tmp = np.where(np.isinf(lf17_hLL_tmp), np.nan, lf17_hLL_tmp)\n",
    "lf17_hLL = lf17_hLL_tmp[~np.isnan(lf17_la_tmp)]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### $h/L_L$ and La$_t$ in JRA55-do"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# load data\n",
    "# data = np.load('data/LRF19_ParamSpace_JRA55do_KPP-CVMix.npz')\n",
    "# fld_laturb_s1 = data['laturb_unstable']\n",
    "# fld_laturb_s2 = data['laturb_stable']\n",
    "# fld_hLL_s1    = data['hLL_unstable']\n",
    "# fld_hLL_s2    = data['hLL_stable']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Destabilizing h/L_L:  62.31%\n",
      "Stabilizing h/L_L:  37.69%\n"
     ]
    }
   ],
   "source": [
    "# load data\n",
    "data = np.load('data/LRF19_Parameters_JRA55do_KPP-CVMix.npz')\n",
    "fld_laturb = data['laturb']\n",
    "fld_bflux  = data['bflux']\n",
    "fld_ustar  = data['ustar']\n",
    "fld_hbl    = data['hbl']\n",
    "fld_hLL    = -fld_bflux*fld_hbl/(fld_ustar**3)*fld_laturb**2\n",
    "\n",
    "def filter_hLL(laturb, hLL, condition='Destabilizing'):\n",
    "    xdata = np.copy(laturb)\n",
    "    ydata = np.copy(hLL)\n",
    "    if condition == 'Destabilizing':\n",
    "        # remove data points where h/L_L<0 \n",
    "        inds = ydata < 0\n",
    "    elif condition == 'Stabilizing':\n",
    "        # remove data points where h/L_L>0\n",
    "        inds = ydata > 0\n",
    "    else:\n",
    "        raise ValueError('Condition {} not supported. Should be \\'Destabilizing\\' or \\'Stabilizing\\'')\n",
    "    ncon = np.sum(inds) \n",
    "    print(condition+' h/L_L: {:6.2f}%'.format(100-ncon/ydata.size*100))\n",
    "    xdata[inds] = np.nan\n",
    "    ydata[inds] = np.nan\n",
    "    xdata = xdata[~np.isnan(xdata)]\n",
    "    ydata = ydata[~np.isnan(ydata)]\n",
    "    return xdata, ydata\n",
    "\n",
    "# Surface cooling\n",
    "fld_laturb_s1, fld_hLL_s1 = filter_hLL(fld_laturb, fld_hLL, 'Destabilizing')\n",
    "fld_laturb_s2, fld_hLL_s2 = filter_hLL(fld_laturb, fld_hLL, 'Stabilizing')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 432x288 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# get the bi-dimensional histogram in log-log space for dataset 1\n",
    "xdata = np.log10(fld_laturb_s1)\n",
    "ydata = np.log10(fld_hLL_s1)\n",
    "hist_s1, xi_s1, yi_s1 = loglog_hist2d(xdata, ydata,\n",
    "                                      xpr=[-1, 1], ypr=[-3, 3])\n",
    "\n",
    "# same for dataset 2\n",
    "xdata = np.log10(fld_laturb_s2)\n",
    "ydata = np.log10(-fld_hLL_s2)\n",
    "hist_s2, xi_s2, yi_s2 = loglog_hist2d(xdata, ydata,\n",
    "                                      xpr=[-1, 1], ypr=[-3, 3])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Plot figure: Figure 1 of [Li et al., 2019](https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2019MS001810)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Figure: regime diagram following Belcher et al., 2012\n",
    "\n",
    "fig, axarr = plt.subplots(1, 2)\n",
    "fig.set_size_inches(8, 5)\n",
    "\n",
    "# plot regime diagram with pdf overlaid\n",
    "plot_regime_diagram_BG12(hist_s1, xi_s1, yi_s1, axis=axarr[0])\n",
    "    \n",
    "axarr[0].text(5, 2e2, '(a)', fontsize=14, color='black')\n",
    "axarr[0].tick_params(axis='both', which='major', length=6)\n",
    "axarr[0].tick_params(axis='both', which='minor', length=4)\n",
    "\n",
    "# plot bi-dimensional histogram\n",
    "plot_regime_diagram_heating(hist_s2, xi_s2, yi_s2, axis=axarr[1])\n",
    "axarr[1].text(5, 2e2, '(b)', fontsize=14, color='black')\n",
    "axarr[1].tick_params(axis='both', which='major', length=6)\n",
    "axarr[1].tick_params(axis='both', which='minor', length=4)\n",
    "\n",
    "# y-shift of legend\n",
    "yshift = -0.13\n",
    "\n",
    "# Li and Fox-Kemper, 2017\n",
    "add_scatter(lf17_la, lf17_hLL, color=c_lf17, axis=axarr[0])\n",
    "\n",
    "# McWilliams et al., 1997\n",
    "add_scatter(msm97_la, msm97_hLL, color=c_msm97, axis=axarr[0])\n",
    "\n",
    "# Min and Noh, 2004\n",
    "add_scatter(mn04_la, mn04_hLL, color=c_mn04, axis=axarr[1]) # heating\n",
    "\n",
    "# Li et al., 2005\n",
    "add_range(l05_la_range_s1, [l05_hLL_s1, l05_hLL_s1], color=c_l05, axis=axarr[0])\n",
    "add_range([l05_la_s2, l05_la_s2], l05_hLL_range_s2, color=c_l05, axis=axarr[0])\n",
    "add_scatter(l05_la_s3, l05_hLL_s3, color=c_l05, axis=axarr[1])\n",
    "\n",
    "# Harcourt and D'Asaro, 2008\n",
    "add_scatter(hd08_la_s1, hd08_hLL_s1, color=c_hd08, axis=axarr[0])\n",
    "add_scatter(hd08_la_s2, hd08_hLL_s2, color=c_hd08, axis=axarr[0])\n",
    "add_scatter(hd08_la_s3, hd08_hLL_s3, color=c_hd08, axis=axarr[0])\n",
    "add_scatter(hd08_la_s4, hd08_hLL_s4, color=c_hd08, axis=axarr[0])\n",
    "\n",
    "# Grant and Belcher, 2009\n",
    "add_scatter(gb09_la, gb09_hLL, color=c_gb09, axis=axarr[0])\n",
    "\n",
    "# Van Roekel et al., 2012\n",
    "add_scatter(vr12_la, vr12_hLL, color=c_vr12, axis=axarr[0])\n",
    "\n",
    "# Pearson et al., 2015\n",
    "add_scatter(p15_la_s1, p15_hLL_s1, color=c_p15, axis=axarr[1]) # heating\n",
    "add_scatter(p15_la_s2, p15_hLL_s2, color=c_p15, axis=axarr[1]) # heating\n",
    "\n",
    "# Reichl et al., 2016\n",
    "r16_la_s1_m = 10**r16_la_s1_mean\n",
    "r16_la_s1_min = 10**(r16_la_s1_mean-r16_la_s1_std)\n",
    "r16_la_s1_max = 10**(r16_la_s1_mean+r16_la_s1_std)\n",
    "r16_hLL_s1_m = 10**r16_hLL_s1_mean\n",
    "r16_hLL_s1_min = 10**(r16_hLL_s1_mean-r16_hLL_s1_std)\n",
    "r16_hLL_s1_max = 10**(r16_hLL_s1_mean+r16_hLL_s1_std)\n",
    "add_mean_and_range(r16_la_s1_m, r16_la_s1_min, r16_la_s1_max,\n",
    "                   r16_hLL_s1_m, r16_hLL_s1_min, r16_hLL_s1_max,\n",
    "                   color=c_r16, axis=axarr[0])\n",
    "r16_la_s2_m = 10**r16_la_s2_mean\n",
    "r16_la_s2_min = 10**(r16_la_s2_mean-r16_la_s2_std)\n",
    "r16_la_s2_max = 10**(r16_la_s2_mean+r16_la_s2_std)\n",
    "r16_hLL_s2_m = 10**r16_hLL_s2_mean\n",
    "r16_hLL_s2_min = 10**(r16_hLL_s2_mean-r16_hLL_s2_std)\n",
    "r16_hLL_s2_max = 10**(r16_hLL_s2_mean+r16_hLL_s2_std)\n",
    "add_mean_and_range(r16_la_s2_m, r16_la_s2_min, r16_la_s2_max,\n",
    "                   r16_hLL_s2_m, r16_hLL_s2_min, r16_hLL_s2_max,\n",
    "                   color=c_r16, axis=axarr[0])\n",
    "\n",
    "axarr[1].plot([1e-1, 1e1], [1/3, 1/3], '--', color='gray', linewidth=1)\n",
    "\n",
    "# reduce margin\n",
    "plt.subplots_adjust(left=0.1, right=0.95, bottom=0.2, top=0.96, wspace=0.3, hspace=0.1)\n",
    "\n",
    "# plot legend\n",
    "# x- and y-labels\n",
    "ax = fig.add_axes([0.1,0.05,0.8,0.1])\n",
    "ax.spines['bottom'].set_color('white')\n",
    "ax.spines['top'].set_color('white') \n",
    "ax.spines['right'].set_color('white')\n",
    "ax.spines['left'].set_color('white')\n",
    "ax.axes.get_yaxis().set_visible(False)\n",
    "ax.axes.get_xaxis().set_visible(False)\n",
    "x_c1 = 0.01\n",
    "x_c2 = 0.34\n",
    "x_c3 = 0.67\n",
    "y_r1 = 0.83\n",
    "y_r2 = 0.5\n",
    "y_r3 = 0.17\n",
    "add_legend(ax, x_c1, y_r1, c_msm97, 'McWilliams et al., 1997')\n",
    "add_legend(ax, x_c2, y_r1, c_mn04,  'Min and Noh, 2004')\n",
    "add_legend(ax, x_c3, y_r1, c_l05,   'Li et al., 2005')\n",
    "add_legend(ax, x_c1, y_r2, c_hd08,  'Harcourt and D’Asaro, 2008')\n",
    "add_legend(ax, x_c2, y_r2, c_gb09,  'Grant and Belcher, 2009')\n",
    "add_legend(ax, x_c3, y_r2, c_vr12,  'Van Roekel et al., 2012')\n",
    "add_legend(ax, x_c1, y_r3, c_p15,   'Pearson et al., 2015')\n",
    "add_legend(ax, x_c2, y_r3, c_r16,   'Reichl et al., 2016')\n",
    "add_legend(ax, x_c3, y_r3, c_lf17,  'Li and Fox-Kemper, 2017')\n",
    "\n",
    "# save figure\n",
    "figname = 'fig_regime_diagram_LRF19.png'\n",
    "plt.savefig(figname, dpi = 300, facecolor='w')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "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": "3.9.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}