fig2.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Figure 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from netCDF4 import Dataset\n",
    "import glob, os, ols, numpy as np\n",
    "from collections import namedtuple\n",
    "\n",
    "from matplotlib.pyplot import *\n",
    "import matplotlib.pyplot as plt\n",
    "import xarray as xr\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Read in models\n",
    "modout=xr.open_dataset('historical/globalavg/historicalEnsemble.nc')\n",
    "\n",
    "modout['time']=pd.to_datetime(modout.time, format='%Y%m%d.%f')\n",
    "modout=modout.groupby('time.year').mean(dim='time') # annual average\n",
    "\n",
    "flux=np.array(modout.srad0+modout.trad0).squeeze()\n",
    "ts=np.array(modout.temp2).squeeze()\n",
    "ta=np.array(modout.t_trop).squeeze()\n",
    "tag=np.array(modout.t).squeeze()\n",
    "tst=np.array(modout.temp2_trop).squeeze()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "delta F = 2.20 W/m^2\n"
     ]
    }
   ],
   "source": [
    "## forcing_aed_ensemble described in notes on 170426\n",
    "# forcing is already annually averaged\n",
    "forcing=xr.open_dataset('historical/globalavg/forcing_aed_ensemble.nc')\n",
    "forcing=np.array(forcing.f).squeeze()\n",
    "forcing=forcing[:156];forcing=forcing.reshape(156,1) # make same length as model runs\n",
    "\n",
    "print('delta F = {:.2f} W/m^2'.format(np.average(forcing[-10:]-forcing[:10])))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "## calculate lambda and theta\n",
    "## output is m1all; m1all[0] is lambda, while m1all[1] is theta\n",
    "s=len(forcing)\n",
    "\n",
    "## analyze\n",
    "length=10 # length of period (in years) at beginning and end to average\n",
    "m1all=[]\n",
    "for tempx in [ts,ta,tag,tst]:\n",
    "\n",
    "    flux1=flux-forcing \n",
    "    \n",
    "    drg=np.average(flux1[-length:,]-flux1[:length,],axis=0)\n",
    "    dtg=np.average(tempx[-length:,]-tempx[:length,],axis=0)\n",
    "\n",
    "    m1all.append(drg/dtg)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "## read in 4xCO2 run of MPI-ESM\n",
    "mpi = namedtuple('mpi', ('ts','ttrop','flux','t'))\n",
    "modelname='historical/globalavg/model0003.nc'\n",
    "ovar=[]\n",
    "with Dataset(modelname) as fni:\n",
    "    for var in ['temp2','trad0','srad0','t_trop','t']:\n",
    "        fx=fni[var][:].squeeze()\n",
    "        ovar.append(fx)\n",
    "\n",
    "mpiout=mpi(ts=ovar[0],flux=ovar[1]+ovar[2],ttrop=ovar[3],t=ovar[4])\n",
    "\n",
    "modelname='historical/globalavg/model0001.nc'\n",
    "ovar=[]\n",
    "with Dataset(modelname) as fni:\n",
    "    for var in ['temp2','trad0','srad0','t_trop','t']:\n",
    "        fx=fni[var][:].squeeze()\n",
    "        ovar.append(fx)\n",
    "\n",
    "mpicontrol=mpi(ts=ovar[0],flux=ovar[1]+ovar[2],ttrop=ovar[3],t=ovar[4])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "_calculate values from the abrupt 4xCO2 model_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "abrupt 4xCO2 $\\lambda$: -1.36 +/- 0.07 W/m2/K\n",
      "abrupt 4xCO2 ECS: 2.87 K\n",
      "abrupt 4xCO2 $\\Theta$: -1.03 +/- 0.04 W/m2/K\n",
      "abrupt 4xCO2 $\\Theta_a$: -1.13 +/- 0.04 W/m2/K\n",
      "\n",
      "ECS from all 2600 years: 3.37\n",
      "lambda from all 2600 years: -0.95 +/- 0.01\n",
      "theta from all 2600 years: -1.00 +/- 0.01\n"
     ]
    }
   ],
   "source": [
    "# calculate lambda and theta for MPI model, along with 5-95% confidence interval\n",
    "# f1 = annual average flux\n",
    "f1=(mpiout.flux-np.average(mpicontrol.flux)).reshape(len(mpiout.ts)//12,12).mean(axis=1)\n",
    "\n",
    "# fit vs. Ts\n",
    "t1=(mpiout.ts-np.average(mpicontrol.ts)).reshape(len(mpiout.ts)//12,12).mean(axis=1)\n",
    "mpifitTs=ols.ols(f1[:150],t1[:150]);mpifitTs.est_auto()\n",
    "\n",
    "# fit vs. Ta,trop\n",
    "t2=(mpiout.ttrop-np.average(mpicontrol.ttrop)).reshape(len(mpiout.ttrop)//12,12).mean(axis=1)\n",
    "mpifitTtrop=ols.ols(f1[:150],t2[:150]);mpifitTtrop.est_auto()\n",
    "\n",
    "# fit vs. Ta,global\n",
    "t3=(mpiout.t-np.average(mpicontrol.t)).reshape(len(mpiout.t)//12,12).mean(axis=1)\n",
    "mpifitTa=ols.ols(f1[:150],t3[:150]);mpifitTa.est_auto()\n",
    "\n",
    "mpiforcing=3.9 # from 1% runs\n",
    "print('abrupt 4xCO2 $\\lambda$: {:.2f} +/- {:.2f} W/m2/K'.format(mpifitTs.b[1],mpifitTs.conf90[1]))\n",
    "print('abrupt 4xCO2 ECS: {:.2f} K'.format(-mpiforcing/mpifitTs.b[1]))\n",
    "print('abrupt 4xCO2 $\\Theta$: {:.2f} +/- {:.2f} W/m2/K'.format(mpifitTtrop.b[1],mpifitTtrop.conf90[1]))\n",
    "print('abrupt 4xCO2 $\\Theta_a$: {:.2f} +/- {:.2f} W/m2/K'.format(mpifitTa.b[1],mpifitTa.conf90[1]))\n",
    "print('')\n",
    "\n",
    "print('ECS from all 2600 years: {:.2f}'.format(np.average(t1[-30:])/2))\n",
    "fit1=ols.ols(f1,t1);fit1.est_auto()\n",
    "print('lambda from all 2600 years: {:.2f} +/- {:.2f}'.format(fit1.b[1],fit1.conf90[1]))\n",
    "fit1=ols.ols(f1,t2);fit1.est_auto()\n",
    "print('theta from all 2600 years: {:.2f} +/- {:.2f}'.format(fit1.b[1],fit1.conf90[1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "def colorplot(ax):\n",
    "    \n",
    "    almost_black = '#262626' # tweek from Thorsten\n",
    "    \n",
    "    for spx in ['top','right']:\n",
    "            ax.spines[spx].set_visible(False)\n",
    "\n",
    "    # Only show ticks on the left and bottom spines\n",
    "    ax.yaxis.set_ticks_position('left')\n",
    "    ax.xaxis.set_ticks_position('bottom')\n",
    "    \n",
    "    # set axis positions\n",
    "#     ax.yaxis.set_label_position(pspines[0]) \n",
    "#     ax.xaxis.set_label_position(pspines[1]) \n",
    "    \n",
    "    for spx in ['bottom','left']:\n",
    "        ax.spines[spx].set_linewidth(0.5)\n",
    "        ax.spines[spx].set_color(almost_black)\n",
    "    \n",
    "    for ticks in ax.xaxis.get_ticklines() + ax.yaxis.get_ticklines():\n",
    "        ticks.set_color(almost_black)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7ff30f87d1d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib as mpl\n",
    "\n",
    "mpl.rcParams['xtick.labelsize'] = 14\n",
    "mpl.rcParams['ytick.labelsize'] = 14\n",
    "\n",
    "# figure 2 — histogram version\n",
    "fig = figure(figsize=(18,5))\n",
    "\n",
    "rect=0.13, 0.13, 0.5, 0.8\n",
    "ax=fig.add_axes(rect) \n",
    "\n",
    "## left-hand panel\n",
    "m1=m1all[1]\n",
    "ax.hist(m1,color='0.6',normed=True,edgecolor='black', linewidth=1.2)\n",
    "for ii in [5,50,95]:\n",
    "    ax.axvline(np.percentile(m1,ii),color='k',ls='dotted',lw=2) \n",
    "\n",
    "ax.text(0.06,0.9,'(a)',fontsize=20,transform=ax.transAxes)\n",
    "\n",
    "m1=m1all[0]\n",
    "ax.hist(m1,color='0.8',normed=True,alpha=0.6,edgecolor='black', linewidth=1.2)\n",
    "colorplot(ax)\n",
    "for ii in [5,50,95]:\n",
    "    ax.axvline(np.percentile(m1,ii),color='k',ls='dashed',lw=1) \n",
    "\n",
    "ax.set_xlim(-1.9,-0.8)\n",
    "# x1=axes[0].get_ylim();axes[0].set_ylim(x1[0],x1[1]*1.1)\n",
    "    \n",
    "ax.set_xlabel('$\\lambda$ or $\\Theta$ ($W/m^2/K$)',fontsize=18)\n",
    "ax.set_ylabel('1/($W/m^2/K$)',fontsize=18)\n",
    "\n",
    "ax.text(-1.5,3.,'$\\lambda$',fontsize=28)\n",
    "ax.text(-1.1,6.,'$\\Theta$',fontsize=28)\n",
    "\n",
    "x1=ax.get_ylim();ax.set_ylim(x1[0],x1[1]*1.08)\n",
    "\n",
    "## right panel\n",
    "rect=0.7, 0.13, 0.25, 0.8\n",
    "ax=fig.add_axes(rect) \n",
    "ecs=-mpiforcing/np.array(m1all[0]) # calculate ECS with known forcing (see MPI forcing calc for first 70 years)\n",
    "ax.hist(ecs,color='0.8',bins=15,normed=True,edgecolor='black', linewidth=1.2)\n",
    "colorplot(ax)\n",
    "for ii in [5,50,95]:\n",
    "    ax.axvline(np.percentile(ecs,ii),color='k',ls='dashed',lw=1) \n",
    "\n",
    "# ax.text(0.5,0.85,'{:.2f}; {:.2f} to {:.2f}'.format(np.median(ecs),np.percentile(ecs,5),np.percentile(ecs,95)),\\\n",
    "#         fontsize=16,transform=ax.transAxes)\n",
    "ax.text(0.06,0.9,'(b)',fontsize=20,transform=ax.transAxes)\n",
    "\n",
    "x1=ax.get_ylim();ax.set_ylim(x1[0],x1[1]*1.08)\n",
    "ax.set_xlabel('ECS ($K$)',fontsize=18)\n",
    "ax.set_ylabel('$1/K$',fontsize=18);\n",
    "\n",
    "# savefig('fig2.pdf')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "range of ECS: 2.08 to 3.87\n",
      "5-95% range of ECS: 2.39 to 3.34\n",
      "median of ECS: 2.72\n",
      "average of ECS: 2.76\n"
     ]
    }
   ],
   "source": [
    "print('range of ECS: {:.2f} to {:.2f}'.format(np.min(ecs),np.max(ecs)))\n",
    "print('5-95% range of ECS: {:.2f} to {:.2f}'.format(np.percentile(ecs,5),np.percentile(ecs,95)))\n",
    "print('median of ECS: {:.2f}'.format(np.median(ecs)))\n",
    "print('average of ECS: {:.2f}'.format(np.average(ecs)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "range of lambda: -1.88 to -1.01\n",
      "5-95% range of lambda: -1.63 to -1.17\n",
      "median of lambda: -1.43\n",
      "average of lambda: -1.42\n"
     ]
    }
   ],
   "source": [
    "print('range of lambda: {:.2f} to {:.2f}'.format(np.min(m1all[0]),np.max(m1all[0])))\n",
    "print('5-95% range of lambda: {:.2f} to {:.2f}'.format(np.percentile(m1all[0],5),np.percentile(m1all[0],95)))\n",
    "print('median of lambda: {:.2f}'.format(np.median(m1all[0])))\n",
    "print('average of lambda: {:.2f}'.format(np.average(m1all[0])))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5-95% confidence interval of median: -1.45 to -1.40 W/m^2/K\n"
     ]
    }
   ],
   "source": [
    "## calculate uncertainty of median\n",
    "## Monte Carlo analysis\n",
    "med1=[]\n",
    "for ii in range(1000):\n",
    "    med1.append(np.median(m1all[0][np.random.randint(len(m1all[0]),size=len(m1all[0]))]))\n",
    "\n",
    "print('5-95% confidence interval of median: {:.2f} to {:.2f} W/m^2/K'.format(np.percentile(med1,5),np.percentile(med1,95)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "twice the standard error: 0.03 W/m^2/K\n"
     ]
    }
   ],
   "source": [
    "## check using more standard approach: standard deviation divided by sqrt(N)\n",
    "print('twice the standard error: {:.2f} W/m^2/K'.format(np.std(m1all[0])*2/10.))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "range of theta: -1.18 to -0.89\n",
      "5-95% range of theta: -1.16 to -0.92\n",
      "median of theta: -1.04\n",
      "average of theta: -1.04\n"
     ]
    }
   ],
   "source": [
    "print('range of theta: {:.2f} to {:.2f}'.format(np.min(m1all[1]),np.max(m1all[1])))\n",
    "print('5-95% range of theta: {:.2f} to {:.2f}'.format(np.percentile(m1all[1],5),np.percentile(m1all[1],95)))\n",
    "print('median of theta: {:.2f}'.format(np.median(m1all[1])))\n",
    "print('average of theta: {:.2f}'.format(np.average(m1all[1])))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5-95% confidence interval of median: -1.05 to -1.03 W/m^2/K\n"
     ]
    }
   ],
   "source": [
    "## calculate uncertainty of median using monte carlo\n",
    "med1=[]\n",
    "for ii in range(1000):\n",
    "    med1.append(np.median(m1all[1][np.random.randint(len(m1all[1]),size=len(m1all[1]))]))\n",
    "\n",
    "print('5-95% confidence interval of median: {:.2f} to {:.2f} W/m^2/K'.format(np.percentile(med1,5),np.percentile(med1,95)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## estimate impact of other lengths"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "this code calculates how the statistics of the ensemble change if the averaging period changes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10 years/ts: 0.8685903549194336\n",
      "20 years/ts: 0.48006415367126465\n",
      "30 years/ts: 0.530222475528717\n",
      "50 years/ts: 0.4876049757003784\n",
      "\n",
      "10 years/ta: 0.29287928342819214\n",
      "20 years/ta: 0.2165355086326599\n",
      "30 years/ta: 0.22595465183258057\n",
      "50 years/ta: 0.17968690395355225\n",
      "\n"
     ]
    }
   ],
   "source": [
    "## change both beginning and end periods\n",
    "\n",
    "## analyze\n",
    "labx=iter('ts,ta'.split(','))\n",
    "flux1=flux-forcing \n",
    "for tempx in [ts,ta]:\n",
    "    labxx=next(labx)\n",
    "    for length in [10,20,30,50]: # length of period (in years) at beginning and end to average\n",
    "        \n",
    "        drg=np.average(flux1[-length:,]-flux1[:length,],axis=0)\n",
    "        dtg=np.average(tempx[-length:,]-tempx[:length,],axis=0)\n",
    "        mx=np.max(drg/dtg);mn=np.min(drg/dtg)\n",
    "        print('{} years/{}: {}'.format(length,labxx,mx-mn))\n",
    "    print()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10 years/ts: 0.8686163425445557\n",
      "20 years/ts: 0.5445048809051514\n",
      "30 years/ts: 0.5525064468383789\n",
      "50 years/ts: 0.5369447469711304\n",
      "\n",
      "10 years/ta: 0.29289841651916504\n",
      "20 years/ta: 0.22870737314224243\n",
      "30 years/ta: 0.227361261844635\n",
      "50 years/ta: 0.2223949432373047\n",
      "\n"
     ]
    }
   ],
   "source": [
    "## keep end period 10 years, but change beginning period\n",
    "\n",
    "## analyze\n",
    "labx=iter('ts,ta'.split(','))\n",
    "flux1=flux-forcing \n",
    "for tempx in [ts,ta]:\n",
    "    labxx=next(labx)\n",
    "    for length in [10,20,30,50]: # length of period (in years) at beginning and end to average\n",
    "        \n",
    "        drg=np.average(flux1[-10:,],axis=0)-np.average(flux1[:length,],axis=0)\n",
    "        dtg=np.average(tempx[-10:,],axis=0)-np.average(tempx[:length,],axis=0)\n",
    "        mx=np.max(drg/dtg);mn=np.min(drg/dtg)\n",
    "        print('{} years/{}: {}'.format(length,labxx,mx-mn))\n",
    "    print()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10 years/ts: 0.8686163425445557\n",
      "20 years/ts: 0.7916418313980103\n",
      "30 years/ts: 0.9580596685409546\n",
      "50 years/ts: 1.168889045715332\n",
      "\n",
      "10 years/ta: 0.29289841651916504\n",
      "20 years/ta: 0.29040080308914185\n",
      "30 years/ta: 0.33711206912994385\n",
      "50 years/ta: 0.5001328587532043\n",
      "\n"
     ]
    }
   ],
   "source": [
    "## keep beginning period 10 years, but change end period\n",
    "\n",
    "## analyze\n",
    "labx=iter('ts,ta'.split(','))\n",
    "flux1=flux-forcing\n",
    "for tempx in [ts,ta]:\n",
    "    labxx=next(labx)\n",
    "    for length in [10,20,30,50]: # length of period (in years) at beginning and end to average\n",
    "\n",
    "        drg=np.average(flux1[-length:,],axis=0)-np.average(flux1[:10,],axis=0)\n",
    "        dtg=np.average(tempx[-length:,],axis=0)-np.average(tempx[:10,],axis=0)\n",
    "        mx=np.max(drg/dtg);mn=np.min(drg/dtg)\n",
    "        print('{} years/{}: {}'.format(length,labxx,mx-mn))\n",
    "    print()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "_changing the beginning period reduces the uncertainty for both $\\lambda$ and $\\Theta$_.  Changing the end period has little impact — if anything, it increases the spread."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "## older stuff\n",
    "\n",
    "old version w/ values from 4xCO2 run"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x7f0c7f427c90>"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABE0AAAFZCAYAAABt4NELAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f0c7f4c8ad0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib as mpl\n",
    "label_size = 8\n",
    "mpl.rcParams['xtick.labelsize'] = 14\n",
    "mpl.rcParams['ytick.labelsize'] = 14\n",
    "\n",
    "# figure 2 — histogram version\n",
    "fig = figure(figsize=(18,5))\n",
    "\n",
    "rect=0.13, 0.13, 0.5, 0.8\n",
    "ax=fig.add_axes(rect) \n",
    "\n",
    "## left-hand panel\n",
    "m1=m1all[1]\n",
    "ax.hist(m1,color='0.6',normed=True)\n",
    "ax.axvline(mpifitTtrop.slope,color='k',ls='dashed',lw=2) # from MPI climate sensitivity calc\n",
    "\n",
    "# ax.text(0.06,0.7,'$\\Theta$: {:.2f};{:.2f} to {:.2f}'.format(np.median(m1),np.percentile(m1,5),np.percentile(m1,95)),\\\n",
    "#         fontsize=16,transform=ax.transAxes)\n",
    "ax.text(0.06,0.9,'(a)',fontsize=20,transform=ax.transAxes)\n",
    "\n",
    "m1=m1all[0]\n",
    "ax.hist(m1,color='0.8',normed=True,alpha=0.6)\n",
    "colorplot(ax)\n",
    "ax.axvline(mpifitTs.slope,color='k',ls='dashed',lw=2) # from MPI climate sensitivity calc\n",
    "\n",
    "ax.text(mpifitTtrop.slope-0.005,6.65,'{:.2f}'.format(mpifitTtrop.slope),rotation=90,fontsize=14,backgroundcolor='w',ha='center') \n",
    "ax.text(mpifitTs.slope,6.65,'{:.2f}'.format(mpifitTs.slope),rotation=90,fontsize=14,backgroundcolor='w',ha='center') \n",
    "\n",
    "ax.set_xlim(-1.9,-0.8)\n",
    "# x1=axes[0].get_ylim();axes[0].set_ylim(x1[0],x1[1]*1.1)\n",
    "    \n",
    "ax.set_xlabel('$\\lambda$ or $\\Theta$',fontsize=22)\n",
    "\n",
    "# ax.text(0.061,0.78,'$\\lambda$: {:.2f};{:.2f} to {:.2f}'.format(np.median(m1),np.percentile(m1,5),np.percentile(m1,95)),\\\n",
    "#         fontsize=16,transform=ax.transAxes)\n",
    "\n",
    "ax.text(-1.5,3.,'$\\lambda$',fontsize=28)\n",
    "ax.text(-0.95,3.,'$\\Theta$',fontsize=28)\n",
    "\n",
    "x1=ax.get_ylim();ax.set_ylim(x1[0],x1[1]*1.08)\n",
    "\n",
    "## right panel\n",
    "rect=0.7, 0.13, 0.25, 0.8\n",
    "ax=fig.add_axes(rect) \n",
    "ecs=-mpiforcing/np.array(m1all[0]) # calculate ECS with known forcing (see MPI forcing calc for first 70 years)\n",
    "ax.hist(ecs,color='0.8',bins=15,normed=True)\n",
    "colorplot(ax)\n",
    "ax.axvline(-mpiforcing/mpifitTs.slope,color='k',ls='dashed',lw=2)\n",
    "ax.text(-mpiforcing/mpifitTs.slope+0.025,1.55,'{:.2f}'.format(-mpiforcing/mpifitTs.slope),\\\n",
    "    rotation=90,fontsize=14,backgroundcolor='w',ha='center') \n",
    "\n",
    "# ax.text(0.5,0.85,'{:.2f}; {:.2f} to {:.2f}'.format(np.median(ecs),np.percentile(ecs,5),np.percentile(ecs,95)),\\\n",
    "#         fontsize=16,transform=ax.transAxes)\n",
    "ax.text(0.06,0.9,'(b)',fontsize=20,transform=ax.transAxes)\n",
    "\n",
    "x1=ax.get_ylim();ax.set_ylim(x1[0],x1[1]*1.08)\n",
    "ax.set_xlabel('ECS',fontsize=22)\n",
    "\n",
    "# savefig('fig2.pdf')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "### Calculation using global-average Ta"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x7f575655cb50>"
      ]
     },
     "execution_count": 92,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABE0AAAFZCAYAAABt4NELAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f57533c4050>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib as mpl\n",
    "label_size = 8\n",
    "mpl.rcParams['xtick.labelsize'] = 14\n",
    "mpl.rcParams['ytick.labelsize'] = 14\n",
    "\n",
    "# figure 2 — histogram version\n",
    "fig = figure(figsize=(18,5))\n",
    "\n",
    "rect=0.13, 0.13, 0.5, 0.8\n",
    "ax=fig.add_axes(rect) \n",
    "\n",
    "## left-hand panel\n",
    "m1=m1all[2]\n",
    "ax.hist(m1,color='0.6',normed=True)\n",
    "ax.axvline(mpifitTa.slope,color='k',ls='dashed',lw=2) # from MPI climate sensitivity calc\n",
    "\n",
    "# ax.text(0.06,0.7,'$\\Theta$: {:.2f}$\\pm${:.2f}, {:.2f}'.format(np.median(m1),np.std(m1),(np.max(m1)-np.min(m1))),\\\n",
    "#         fontsize=20,transform=ax.transAxes)\n",
    "ax.text(0.06,0.7,'$\\Theta$: {:.2f};{:.2f} to {:.2f}'.format(np.median(m1),np.percentile(m1,5),np.percentile(m1,95)),\\\n",
    "        fontsize=16,transform=ax.transAxes)\n",
    "ax.text(0.06,0.9,'(a)',fontsize=20,transform=ax.transAxes)\n",
    "\n",
    "m1=m1all[0]\n",
    "ax.hist(m1,color='0.8',normed=True,alpha=0.6)\n",
    "colorplot(ax)\n",
    "ax.axvline(mpifitTs.slope,color='k',ls='dashed',lw=2) # from MPI climate sensitivity calc\n",
    "\n",
    "ax.text(mpifitTa.slope-0.005,6.65,'{:.2f}'.format(mpifitTa.slope),rotation=90,fontsize=14,backgroundcolor='w',ha='center') \n",
    "ax.text(mpifitTs.slope,6.65,'{:.2f}'.format(mpifitTs.slope),rotation=90,fontsize=14,backgroundcolor='w',ha='center') \n",
    "\n",
    "ax.set_xlim(-1.9,-0.8)\n",
    "# x1=axes[0].get_ylim();axes[0].set_ylim(x1[0],x1[1]*1.1)\n",
    "    \n",
    "ax.set_xlabel('$\\lambda$ or $\\Theta$',fontsize=22)\n",
    "\n",
    "# ax.text(0.06,0.8,'$\\lambda$: {:.2f}$\\pm${:.2f}, {:.2f}'.format(np.median(m1),np.std(m1),(np.max(m1)-np.min(m1))),\\\n",
    "#         fontsize=20,transform=ax.transAxes)\n",
    "ax.text(0.061,0.78,'$\\lambda$: {:.2f};{:.2f} to {:.2f}'.format(np.median(m1),np.percentile(m1,5),np.percentile(m1,95)),\\\n",
    "        fontsize=16,transform=ax.transAxes)\n",
    "\n",
    "ax.text(-1.5,3.,'$\\lambda$',fontsize=28)\n",
    "ax.text(-0.95,3.,'$\\Theta$',fontsize=28)\n",
    "\n",
    "x1=ax.get_ylim();ax.set_ylim(x1[0],x1[1]*1.08)\n",
    "\n",
    "## right panel\n",
    "rect=0.7, 0.13, 0.25, 0.8\n",
    "ax=fig.add_axes(rect) \n",
    "ecs=-mpiforcing/np.array(m1all[0]) # calculate ECS with known forcing (see MPI forcing calc for first 70 years)\n",
    "ax.hist(ecs,color='0.8',bins=15,normed=True)\n",
    "colorplot(ax)\n",
    "ax.axvline(-mpiforcing/mpifitTs.slope,color='k',ls='dashed',lw=2)\n",
    "ax.text(-mpiforcing/mpifitTs.slope+0.025,1.55,'{:.2f}'.format(-mpiforcing/mpifitTs.slope),\\\n",
    "    rotation=90,fontsize=14,backgroundcolor='w',ha='center') \n",
    "\n",
    "# ax.text(0.65,0.92,'%5.2f$\\pm$%4.2f' % (np.average(ecs),np.std(ecs)),\\\n",
    "#         fontsize=16,transform=ax.transAxes)\n",
    "# ax.text(0.8,0.82,'%4.2f' % (np.max(ecs)-np.min(ecs)),\\\n",
    "#         fontsize=16,transform=ax.transAxes)\n",
    "# ax.text(0.5,0.85,'{:.2f}$\\pm${:.2f}, {:.1f}'.format(np.median(ecs),np.std(ecs),(np.max(ecs)-np.min(ecs))),\\\n",
    "#         fontsize=20,transform=ax.transAxes)\n",
    "ax.text(0.5,0.85,'{:.2f}; {:.2f} to {:.2f}'.format(np.median(ecs),np.percentile(ecs,5),np.percentile(ecs,95)),\\\n",
    "        fontsize=16,transform=ax.transAxes)\n",
    "ax.text(0.06,0.9,'(b)',fontsize=20,transform=ax.transAxes)\n",
    "\n",
    "x1=ax.get_ylim();ax.set_ylim(x1[0],x1[1]*1.08)\n",
    "ax.set_xlabel('ECS',fontsize=22)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "Calculation using tropical-average Ts"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x7f5753022810>"
      ]
     },
     "execution_count": 95,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABFoAAAFZCAYAAACokkuaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nkiRJktQSEy2SJEmSJEkt+f8qDUreElCIeAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f575343b590>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib as mpl\n",
    "label_size = 8\n",
    "mpl.rcParams['xtick.labelsize'] = 14\n",
    "mpl.rcParams['ytick.labelsize'] = 14\n",
    "\n",
    "# figure 2 — histogram version\n",
    "fig = figure(figsize=(18,5))\n",
    "\n",
    "rect=0.13, 0.13, 0.5, 0.8\n",
    "ax=fig.add_axes(rect) \n",
    "\n",
    "## left-hand panel\n",
    "m1=m1all[3]\n",
    "ax.hist(m1,color='0.6',normed=True)\n",
    "\n",
    "# ax.text(0.06,0.7,'$\\Theta$: {:.2f}$\\pm${:.2f}, {:.2f}'.format(np.median(m1),np.std(m1),(np.max(m1)-np.min(m1))),\\\n",
    "#         fontsize=20,transform=ax.transAxes)\n",
    "ax.text(0.06,0.7,'$\\lambda_t$: {:.2f};{:.2f} to {:.2f}'.format(np.median(m1),np.percentile(m1,5),np.percentile(m1,95)),\\\n",
    "        fontsize=16,transform=ax.transAxes)\n",
    "ax.text(0.06,0.9,'(a)',fontsize=20,transform=ax.transAxes)\n",
    "\n",
    "m1=m1all[0]\n",
    "ax.hist(m1,color='0.8',normed=True,alpha=0.6)\n",
    "colorplot(ax)\n",
    "\n",
    "ax.set_xlim(-2.1,-0.8)\n",
    "# x1=axes[0].get_ylim();axes[0].set_ylim(x1[0],x1[1]*1.1)\n",
    "    \n",
    "ax.set_xlabel('$\\lambda$ or $\\Theta$',fontsize=22)\n",
    "\n",
    "# ax.text(0.06,0.8,'$\\lambda$: {:.2f}$\\pm${:.2f}, {:.2f}'.format(np.median(m1),np.std(m1),(np.max(m1)-np.min(m1))),\\\n",
    "#         fontsize=20,transform=ax.transAxes)\n",
    "ax.text(0.061,0.78,'$\\lambda$: {:.2f};{:.2f} to {:.2f}'.format(np.median(m1),np.percentile(m1,5),np.percentile(m1,95)),\\\n",
    "        fontsize=16,transform=ax.transAxes)\n",
    "\n",
    "ax.text(-1.5,3.,'$\\lambda$',fontsize=28)\n",
    "ax.text(-0.95,3.,'$\\Theta$',fontsize=28)\n",
    "\n",
    "x1=ax.get_ylim();ax.set_ylim(x1[0],x1[1]*1.08)\n",
    "\n",
    "## right panel\n",
    "rect=0.7, 0.13, 0.25, 0.8\n",
    "ax=fig.add_axes(rect) \n",
    "ecs=-mpiforcing/np.array(m1all[0]) # calculate ECS with known forcing (see MPI forcing calc for first 70 years)\n",
    "ax.hist(ecs,color='0.8',bins=15,normed=True)\n",
    "colorplot(ax)\n",
    "ax.axvline(-mpiforcing/mpifitTs.slope,color='k',ls='dashed',lw=2)\n",
    "ax.text(-mpiforcing/mpifitTs.slope+0.025,1.55,'{:.2f}'.format(-mpiforcing/mpifitTs.slope),\\\n",
    "    rotation=90,fontsize=14,backgroundcolor='w',ha='center') \n",
    "\n",
    "# ax.text(0.65,0.92,'%5.2f$\\pm$%4.2f' % (np.average(ecs),np.std(ecs)),\\\n",
    "#         fontsize=16,transform=ax.transAxes)\n",
    "# ax.text(0.8,0.82,'%4.2f' % (np.max(ecs)-np.min(ecs)),\\\n",
    "#         fontsize=16,transform=ax.transAxes)\n",
    "# ax.text(0.5,0.85,'{:.2f}$\\pm${:.2f}, {:.1f}'.format(np.median(ecs),np.std(ecs),(np.max(ecs)-np.min(ecs))),\\\n",
    "#         fontsize=20,transform=ax.transAxes)\n",
    "ax.text(0.5,0.85,'{:.2f}; {:.2f} to {:.2f}'.format(np.median(ecs),np.percentile(ecs,5),np.percentile(ecs,95)),\\\n",
    "        fontsize=16,transform=ax.transAxes)\n",
    "ax.text(0.06,0.9,'(b)',fontsize=20,transform=ax.transAxes)\n",
    "\n",
    "x1=ax.get_ylim();ax.set_ylim(x1[0],x1[1]*1.08)\n",
    "ax.set_xlabel('ECS',fontsize=22)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "### calculate stats for $T_{500}$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0101  0102  0103  0104  0105  0106  0107  0108  0109  0110  0111  0112  0113  0114  0115  0116  0117  0118  0119  0120  0121  0122  0123  0124  0125  0126  0127  0128  0129  0130  0131  0132  0133  0134  0135  0136  0137  0138  0139  0140  0141  0142  0143  0144  0145  0146  0147  0148  0149  0150  0151  0152  0153  0154  0155  0156  0157  0158  0159  0160  0161  0162  0163  0164  0165  0166  0167  0168  0169  0170  0171  0172  0173  0174  0175  0176  0177  0178  0179  0180  0181  0182  0183  0184  0185  0186  0187  0188  0189  0190  0191  0192  0193  0194  0195  0196  0197  0198  0199  0200 \n"
     ]
    }
   ],
   "source": [
    "s=len(forcing)\n",
    "\n",
    "## analyze\n",
    "length=10;m1all2=[]\n",
    "m1=[]\n",
    "for mn in range(101,201):\n",
    "\tprint '%04i ' % mn,\n",
    "\tsys.stdout.flush()\n",
    "\n",
    "\tmod1=modout[mn]\n",
    "\tflux=mod1.flux-forcing # fluxes are the same for all calcs\n",
    "\tt1=mod1.t\n",
    "\n",
    "\tdrg.append(np.average(flux[-length*12:]-flux[:length*12]))\n",
    "\tdtg.append(np.average(t1[-length*12:]-t1[:length*12]))\n",
    "\tm1.append(drg[-1]/dtg[-1])\n",
    "\n",
    "m1all2.append(m1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "average and standard deviation of $\\Theta$: -1.09 +/- 0.07\n"
     ]
    }
   ],
   "source": [
    "print 'average and standard deviation of $\\Theta$: {:.2f} +/- {:.2f}'.format(np.average(m1all2[0]),np.std(m1all2[0]))"
   ]
  }
 ],
 "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": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}