{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# North Atlantic Subpolar Gyre Index from EOF analysis\n",
    "### for VIKING20X paper\n",
    "\n",
    "Use SSH to compute PC1 and PC 2 following Hatun and Chafik (2018) as well as Koul et al. (2020)\n",
    "\n",
    "Defining an SPG index based on SSH EOF analysis:\n",
    "\"The first index (PC1 SSH) is defined as the principal component of the leading Empirical Orthogonal Function (EOF) of annual mean SSH anomalies in the subpolar North Atlantic, defined in the domain 20°N to 70°N, 0°W to 80°W (see Hakkinen and Rhines, 2004), and defined for the altimeter period 1993–2016. Similarly, the second index (PC2 SSH) is defined as the principal component of the second EOF of annual mean SSH anomalies.\" (Koul et al., 2020)\n",
    "\n",
    "**The subpolar North Atlantic is defined as the region 20-70N, 0-80W** (Koul et al., 2020).\n",
    "In fact, Hakkinen and Rhines used data that did not extend north of about 65N and not into the Hudson Bay. (cf H&R04 as well as Hatun and Chafik, 2018)\n",
    "\n",
    "[EOF-software by Andrew Dawson](https://ajdawson.github.io/eofs/latest/):\n",
    "Dawson, A., 2016. eofs: A Library for EOF Analysis of Meteorological, Oceanographic, and Climate Data. Journal of Open Research Software, 4(1), p.e14. DOI: http://doi.org/10.5334/jors.122\n",
    "\n",
    "Monthly SPG index of H&C18 at https://bolin.su.se/data/chafik-2019-3\n",
    "\n",
    "Time series of the subpolar gyre index in \\Oj (thin black), \\Vjs (thick black dotted), \\Vjl (thick black solid) and \\Vc (blue), and based on observations (orange).\n",
    "\n",
    "Also provide mean and std.dev for 1990-2009 and correlations between time series\n",
    "\n",
    "Vielleicht macht die Korrelation sowohl auf kurzen (1990-2009, dann mit Beobachtungen) als auch langen (1958-2009, dann nur Modelle; ohne die Trends werden die auch gut korreliert sein) Sinn. Letztendlich möchte ich damit zwei Aussagen dokumentieren: Die Modelle stimmen hinsichtlich der interannualen Variabilität gut mit den Beobachtungen überein. Die dekadische Variabilität ist über die einzelnen Experimente relativ robust. Beides dokumentiert die Wichtigkeit des Windantriebs. \n",
    "\n",
    "### Model runs\n",
    "| Model | pen style | data path |\n",
    "| --- | --- | --- |\n",
    "| OJo (ORCA025-JRA-OMIP) | solid thin red | scalc01:/data/user/tomartin/Models/NEMO/orca025.l46/experiments/ORCA025.L46-KFS003-V/derived |\n",
    "| OJo2 (ORCA025-JRA-OMIP-2nd) | dashed thin red | scalc01:/data/user/tomartin/Models/NEMO/orca025.l46/experiments/ORCA025.L46-KFS003-V-2nd/derived |\n",
    "| OJ (ORCA025-JRA) | solid thin green | nesh-fe:/sfs/fs1/work-geomar3/smomw091/SDIR/ORCA025.L46/ORCA025.L46-KFS001-V/1m/ |\n",
    "| OJst (ORCA025-JRA-strong) | dashed thin green | blogin:/scratch/usr/shkifmfs/shared/ORCA025.L46-KFS006_monthly_SSH |\n",
    "| VJo (VIKING20X-JRA-OMIP) | solid red | scalc01:/data/user/tomartin/Models/NEMO/viking20x.l46/experiments/VIKING20X.L46-KFS003/derived |\n",
    "| VJl (VIKING20X-JRA-long) | solid blue | nesh-fe:/sfs/fs1/work-geomar3/smomw091/SDIR/VIKING20X.L46/VIKING20X.L46-KFS001-S/1m/ |\n",
    "| VJs (VIKING20X-JRA-short) | dashed blue | nesh-fe:/sfs/fs1/work-geomar3/smomw091/SDIR/VIKING20X.L46/VIKING20X.L46-KKG36107B-S/1m/ |\n",
    "| VC (VIKING20X-CORE) | solid black | nesh-fe:/sfs/fs1/work-geomar3/smomw091/SDIR/VIKING20X.L46/VIKING20X.L46-KKG36013H-S/1m/ |\n",
    "| Observations | solid orange | scalc01:/data/user/tomartin/Observations/SSH/ |\n",
    "\n",
    "### AVISO+ observations\n",
    "\n",
    "adt:long_name = \"Absolute dynamic topography\" ;\n",
    "adt:standard_name = \"sea_surface_height_above_geoid\" ;\n",
    "adt:units = \"m\" ;\n",
    "adt:comment = \"The absolute dynamic topography is the sea surface height above geoid; the adt is obtained as follows: adt=sla+mdt where mdt is the mean dynamic topography; see the product user manual for details\" ;\n",
    "\n",
    "sla:long_name = \"Sea level anomaly\" ;\n",
    "sla:standard_name = \"sea_surface_height_above_sea_level\" ;\n",
    "sla:units = \"m\" ;\n",
    "sla:comment = \"The sea level anomaly is the sea surface height above mean sea surface; it is referenced to the [1993, 2012] period; see the product user manual for details\" ;\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Initalization:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import xarray as xr\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from scipy import stats\n",
    "from eofs.xarray import Eof"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "save_figure=True\n",
    "\n",
    "projname='figure10'\n",
    "projpath='/home/tomartin/Projects/VIKING20X/'\n",
    "figpath=projpath+'Figures/'\n",
    "\n",
    "aviso_varname='adt' # adt or sla where adt is the better match for NEMO's diagnosed sossheig"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "SMALL_SIZE=18\n",
    "MED_SIZE=24\n",
    "BIG_SIZE=28\n",
    "plt.rc('font',   size     =SMALL_SIZE)    # controls default text sizes\n",
    "plt.rc('axes',   titlesize=SMALL_SIZE)    # fontsize of the axes title\n",
    "plt.rc('axes',   labelsize=MED_SIZE)      # fontsize of the x and y labels\n",
    "plt.rc('xtick',  labelsize=SMALL_SIZE)    # fontsize of the tick labels\n",
    "plt.rc('ytick',  labelsize=SMALL_SIZE)    # fontsize of the tick labels\n",
    "plt.rc('legend', fontsize =SMALL_SIZE)    # legend fontsize\n",
    "plt.rc('figure', titlesize=BIG_SIZE)      # fontsize of the figure title"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "workdir='/sfs/fs1/work-geomar5/smomw135/Models/NEMO/' # NESH\n",
    "workdir='/data/user/tomartin/Models/NEMO/' # GEOMAR\n",
    "\n",
    "obspath='/data/user/tomartin/Observations/SSH/MonthlyMean/dt_global_allsat_phy_l4_mm_1993-2018_'+aviso_varname+'.nc'\n",
    "\n",
    "shortname=['OJo','OJo2','Oj','Ojst',\n",
    "           'VJo','Vjl','Vjs','Vc','Obs']\n",
    "longname=['ORCA025-JRA-OMIP','ORCA025-JRA-OMIP-2nd','ORCA025-JRA','ORCA025-JRA-strong',\n",
    "          'VIKING20X-JRA-OMIP','VIKING20-JRA-long','VIKING20-JRA-short','VIKING20-CORE','Observations']\n",
    "modelname=['orca025.l46','orca025.l46','orca025.l46','orca025.l46',\n",
    "           'viking20x.l46','viking20x.l46','viking20x.l46','viking20x.l46','aviso.']\n",
    "expname=['ORCA025.L46-KFS003-V','ORCA025.L46-KFS003-V-2nd','ORCA025.L46-KFS001-V','ORCA025.L46-KFS006',\n",
    "         'VIKING20X.L46-KFS003','VIKING20X.L46-KFS001','VIKING20X.L46-KKG36107B','VIKING20X.L46-KKG36013H','AVISO']\n",
    "year1=['1958','1958','1958','1958',\n",
    "       '1958','1958','1980','1958','1993']\n",
    "year2=['2019','2019','2019','2019',\n",
    "       '2019','2019','2019','2009','2018']\n",
    "\n",
    "penwid=[1,1,1,1,3,3,3,3,3]\n",
    "pensty=['-','--','-','--','-','-','--','-','-']\n",
    "dshsty=[[1,0],[6,4],[1,0],[6,4],[1,0],[1,0],[6,4],[1,0],[1,0]]\n",
    "pencol=['r','r',(0,.6,0),(0,.6,0),'r','b','b','k',(1,.7,0)]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Functions:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_cellarea(gridname):\n",
    "    \"\"\"\n",
    "    returns grid-cell areas computed from e1t and e2t\n",
    "    \"\"\"\n",
    "    gridname=gridname.lower()\n",
    "    if gridname == 'orca05':\n",
    "        meshfile='/home/tomartin/ModelGrids/NEMO-ORCA05/ORCA05.L46_mesh_mask.nc'\n",
    "    elif gridname == 'orca025':\n",
    "        meshfile='/home/tomartin/ModelGrids/NEMO-ORCA025/mesh_hgr.nc'\n",
    "    elif gridname == 'viking20x':\n",
    "        meshfile='/home/tomartin/ModelGrids/NEMO-VIKING20X/mesh_mask.nc'\n",
    "    elif gridname == '1_viking20x':\n",
    "        meshfile='/home/tomartin/ModelGrids/NEMO-VIKING20X/1_mesh_mask.nc'\n",
    "    else:\n",
    "        print('ERROR in get_cellarea: gridname',gridname,'not implemented')\n",
    "        return\n",
    "    ds=xr.open_dataset(meshfile)\n",
    "    area=(ds.e1t*ds.e2t).squeeze()\n",
    "    return area.values\n",
    "\n",
    "def get_tmask0(gridname):\n",
    "    \"\"\"\n",
    "    returns land-sea mask for surface layer of T-grid\n",
    "    \"\"\"\n",
    "    gridname=gridname.lower()\n",
    "    if gridname == 'orca05':\n",
    "        meshfile='/home/tomartin/ModelGrids/NEMO-ORCA05/ORCA05.L46_mesh_mask.nc'\n",
    "    elif gridname == 'orca025':\n",
    "        meshfile='/home/tomartin/ModelGrids/NEMO-ORCA025/mask.nc'\n",
    "    elif gridname == 'viking20x':\n",
    "        meshfile='/home/tomartin/ModelGrids/NEMO-VIKING20X/mesh_mask.nc'\n",
    "    elif gridname == '1_viking20x':\n",
    "        meshfile='/home/tomartin/ModelGrids/NEMO-VIKING20X/1_mesh_mask.nc'\n",
    "    else:\n",
    "        print('ERROR in get_cellarea: gridname',gridname,'not implemented')\n",
    "        return\n",
    "    ds=xr.open_dataset(meshfile)\n",
    "    tmask0=ds.tmask.isel(z=0)\n",
    "    tmask0=tmask0.where(tmask0>0).squeeze()\n",
    "    return tmask0.values\n",
    "\n",
    "def get_ssh_glbm(dataset,gridname):\n",
    "    \"\"\"\n",
    "    returns global mean SSH using grid-cell area weighted averaging\n",
    "    \"\"\"\n",
    "#    gridname=dataset.name[dataset.name.rfind('/')+1:dataset.name.find('.L46')].lower()\n",
    "    gridname=gridname.lower()\n",
    "    mask=get_tmask0(gridname)\n",
    "    area=get_cellarea(gridname)*mask\n",
    "    ssh=(dataset.sossheig*area).sum(('y','x'),skipna=True)/np.nansum(area)\n",
    "    try:\n",
    "        ssh=shh.reset_coords('time_centered',drop=True)\n",
    "        print('time_centered removed')\n",
    "    except:\n",
    "        print('no time_centered found')\n",
    "    return ssh #.values\n",
    "\n",
    "def get_ssh_spg(dataset,gridname):\n",
    "    \"\"\"\n",
    "    returns the mean SSH at (57˚N,52˚W) corrected by the global mean SSH\n",
    "      a spatial mean over a box of approx. 2˚x2˚ is computed\n",
    "      using grid-cell area weighted averaging\n",
    "    \"\"\"\n",
    "    # (i,j) locations of (57N,52W)\n",
    "#    gridname=dataset.name[dataset.name.rfind('/')+1:dataset.name.find('.L46')].lower()\n",
    "    gridname=gridname.lower()\n",
    "    if gridname == 'orca05':\n",
    "        j57n=388; i52w=475; nspan=2; jfac=2 # LAT 57.03027 LON -51.841125\n",
    "    elif (gridname == 'orca025') or (gridname == 'viking20x'):\n",
    "        j57n=776; i52w=950; nspan=4; jfac=2 # LAT 57.03027 LON -51.841125\n",
    "    elif gridname == '1_viking20x':\n",
    "        j57n=np.nan\n",
    "        i52w=np.nan\n",
    "        nspan=np.nan\n",
    "    elif gridname == 'aviso':\n",
    "        j57n=588; i52w=1232; nspan=4; jfac=1 # LAT 57.125 LON -51.875\n",
    "    else:\n",
    "        print('ERROR in get_ssh_spg: gridname',gridname,'not implemented')\n",
    "        return\n",
    "    jslice=slice(j57n-nspan*jfac,j57n+nspan*jfac+1)\n",
    "    islice=slice(i52w-nspan,i52w+nspan+1)\n",
    "    #\n",
    "    if gridname != 'aviso':\n",
    "        # get masked grid-cell area\n",
    "        mask=get_tmask0(gridname)[jslice,islice]\n",
    "        area=get_cellarea(gridname)[jslice,islice]*mask\n",
    "        # get global mean SSH\n",
    "        ssh_glbm=get_ssh_glbm(dataset,gridname)\n",
    "        # compute SSH at SPG center\n",
    "        ssh=(dataset.sossheig.isel(y=jslice,x=islice).squeeze()*area).sum(('y','x'),skipna=True)/np.nansum(area)\n",
    "    else:\n",
    "        # get masked grid-cell area\n",
    "        mask=get_mask_aviso()[jslice,islice]\n",
    "        area=get_cellarea_aviso()[jslice,islice]*mask\n",
    "        # get global mean SSH\n",
    "        ssh_glbm=get_ssh_glbm_aviso(dataset)\n",
    "        # compute SSH at SPG center\n",
    "        ssh=eval('(dataset.'+aviso_varname+'.isel(latitude=jslice,longitude=islice).squeeze()*area).sum((\\'latitude\\',\\'longitude\\'),skipna=True)/np.nansum(area)')\n",
    "    ssh=ssh-ssh_glbm.values\n",
    "    return ssh #.values\n",
    "\n",
    "def get_ssh_NA(dataset,gridname):\n",
    "    \"\"\"\n",
    "    returns a cropped SSH array with only the region 30-70N, 80W-10E\n",
    "    and grid cell areas for same region\n",
    "    \"\"\"\n",
    "    gridname=gridname.lower()\n",
    "    if gridname == 'orca05':\n",
    "        yslice=slice(289,430)\n",
    "        xslice=slice(412,574)\n",
    "    elif gridname == 'orca025':\n",
    "        yslice=slice(578,860)\n",
    "        xslice=slice(824,1148)\n",
    "    elif gridname == 'viking20x':\n",
    "        yslice=slice(578,860)\n",
    "        xslice=slice(824,1148)\n",
    "    elif gridname == '1_viking20x':\n",
    "        print('ERROR: subdomain for EOF not yet specified')\n",
    "        return\n",
    "    elif gridname == 'aviso':\n",
    "        xslice=slice(1120,1440)\n",
    "        yslice=slice(440,640)    #\n",
    "    if gridname != 'aviso':\n",
    "        mask=get_tmask0(gridname); mask=mask[yslice,xslice]\n",
    "        area=get_cellarea(gridname); area=area[yslice,xslice]\n",
    "        ssh=dataset.sossheig.isel(y=yslice,x=xslice)*mask\n",
    "        try:\n",
    "            ssh=shh.reset_coords('time_centered',drop=True)\n",
    "            print('time_centered removed')\n",
    "        except:\n",
    "            print('no time_centered found')\n",
    "    else:\n",
    "        mask=get_mask_aviso()[yslice,xslice]\n",
    "        area=get_cellarea_aviso()[yslice,xslice]*mask\n",
    "        ssh=eval('dataset.'+aviso_varname+'.isel(latitude=yslice,longitude=xslice).squeeze()')\n",
    "    #\n",
    "    return ssh,area #.values\n",
    "\n",
    "def get_cellarea_aviso():\n",
    "    \"\"\"\n",
    "    Read grid-cell area for AVISO data\n",
    "    computed with cdo gridarea\n",
    "    \"\"\"\n",
    "    area=xr.open_dataset('/data/user/tomartin/Observations/SSH/gridcellarea.nc').cell_area.squeeze().values\n",
    "    return area\n",
    "\n",
    "def get_mask_aviso():\n",
    "    \"\"\"\n",
    "    Derive land-sea mask from data\n",
    "    \"\"\"\n",
    "    ssh=eval('xr.open_dataset(obspath,decode_times=False).'+aviso_varname+'.sum(\\'time\\').squeeze()')\n",
    "    mask=ssh*0.0+1.0\n",
    "    return mask.values\n",
    "\n",
    "def get_ssh_glbm_aviso(dataset):\n",
    "    mask=get_mask_aviso()\n",
    "    area=get_cellarea_aviso()*mask\n",
    "    ssh=eval('(dataset.'+aviso_varname+'*area).sum((\\'latitude\\',\\'longitude\\'),skipna=True)/np.nansum(area)')\n",
    "    return ssh #.values\n",
    "\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### SPG index plots"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "use_ym=True   # compute annual means\n",
    "use_ts=True   # use already computed timeseries (False for reading from original data)\n",
    "\n",
    "eof_n=2\n",
    "eof_sign=[1,-1,-1,1,-1,-1,-1,1,1] # set sign of PC to align all data sets (EOF analysis may yield mirrored patterns)\n",
    "imode=1\n",
    "\n",
    "ref_period=(1990,2009)\n",
    "\n",
    "fig,ax=plt.subplots(figsize=(12,5))\n",
    "\n",
    "#for irun in range(0,9):\n",
    "for irun in np.array([0,4,5,6,7,8]):\n",
    "#    print('irun=',irun)\n",
    "    if use_ts:\n",
    "        npzfile = np.load('ProcessedData/'+projname+'_ts_'+str(irun)+'.npz')\n",
    "        ssh_pc=npzfile['pc']\n",
    "        time=npzfile['time']\n",
    "    else:\n",
    "        gridname=modelname[irun][0:modelname[irun].find('.')]\n",
    "        if irun == len(shortname)-1:\n",
    "            inpath=obspath\n",
    "            ds=xr.open_dataset(inpath)\n",
    "            ssh_gm_ltm=get_ssh_glbm_aviso(ds).mean()\n",
    "        else:\n",
    "            inpath=workdir+modelname[irun]+'/experiments/'+expname[irun]+'/derived/'+expname[irun]+'_1m_'+year1[irun]+'0101_'+year2[irun]+'1231_sossheig.nc'\n",
    "            ds=xr.open_dataset(inpath).rename({'time_counter':'time'})\n",
    "            ssh_gm_ltm=get_ssh_glbm(ds,gridname).mean()\n",
    "        ssh,wgts=get_ssh_NA(ds,gridname)\n",
    "        wgts=wgts/np.sum(wgts)\n",
    "        if use_ym:\n",
    "            ssh=ssh.groupby('time.year').mean('time')\n",
    "            time=np.arange(int(year1[irun]),int(year2[irun])+1,1)\n",
    "        else:\n",
    "            time=np.arange(int(year1[irun]),int(year2[irun])+1,1/12)\n",
    "        data=(ssh-ssh_gm_ltm).rename({'year':'time'})\n",
    "        # EOF:\n",
    "        eof_solver = Eof(data, weights=wgts)\n",
    "        ssh_eof = eof_solver.eofs(neofs=eof_n)\n",
    "        ssh_pc  = eof_solver.pcs(npcs=eof_n,pcscaling=1) # scaled to unit variance\n",
    "        ssh_var = eof_solver.varianceFraction(neigs=eof_n)*100 # in %\n",
    "        # save result:\n",
    "        np.savez('ProcessedData/'+projname+'_ts_'+str(irun)+'.npz',eof=ssh_eof,pc=ssh_pc,var=ssh_var,time=time)\n",
    "        ssh_pc=ssh_pc.values\n",
    "    #            \n",
    "    ssh_spg=ssh_pc[:,imode] # 2nd mode is SPG index according to Koul et al\n",
    "    #\n",
    "    ax.plot(time,ssh_spg*eof_sign[irun],\n",
    "        color=pencol[irun],linestyle=pensty[irun],dashes=dshsty[irun],linewidth=penwid[irun],\n",
    "        label=longname[irun])\n",
    "\n",
    "ax.legend(bbox_to_anchor=(1.05, 1.), loc='upper left',\n",
    "    ncol=1, borderaxespad=0., fontsize=14, handlelength=3.5)#, mode=\"expand\")\n",
    "\n",
    "plt.xlabel('year')\n",
    "plt.ylabel('standardized SSH index')\n",
    "plt.xlim([1958,2020])\n",
    "plt.ylim([-3,3])\n",
    "\n",
    "if save_figure:\n",
    "    plt.savefig(figpath+projname+'.pdf',bbox_inches='tight')\n",
    "    plt.savefig(figpath+projname+'.png',dpi=300,bbox_inches='tight')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### compute correlations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "correlationsfor 1993-2009\n",
      "Observations with ORCA025-JRA-OMIP    0.86 at signif.level  1.00\n",
      "Observations with VIKING20X-JRA-OMIP  0.71 at signif.level  1.00\n",
      "Observations with VIKING20-JRA-long   0.76 at signif.level  1.00\n",
      "Observations with VIKING20-JRA-short  0.34 at signif.level  0.80\n",
      "Observations with VIKING20-CORE       0.85 at signif.level  1.00\n",
      "\n",
      "correlations for 1958-2009\n",
      "VIKING20-CORE with ORCA025-JRA-OMIP    0.05 at signif.level  0.26\n",
      "VIKING20-CORE with ORCA025-JRA-OMIP-2nd  0.48 at signif.level  1.00\n",
      "VIKING20-CORE with ORCA025-JRA         0.52 at signif.level  1.00\n",
      "VIKING20-CORE with ORCA025-JRA-strong -0.18 at signif.level  0.78\n",
      "VIKING20-CORE with VIKING20X-JRA-OMIP  0.14 at signif.level  0.69\n",
      "VIKING20-CORE with VIKING20-JRA-long   0.09 at signif.level  0.47\n"
     ]
    }
   ],
   "source": [
    "print('correlationsfor 1993-2009')\n",
    "ref_period=(1993,2009)\n",
    "# observations as reference:\n",
    "for i,irun in enumerate((8,0,4,5,6,7)):\n",
    "    npzfile = np.load('ProcessedData/'+projname+'_ts_'+str(irun)+'.npz')\n",
    "    ssh_pc=npzfile['pc']\n",
    "    ssh_spg=ssh_pc[:,imode]\n",
    "    time=npzfile['time']\n",
    "    t1=np.min(np.where(time>=ref_period[0]))\n",
    "    t2=np.max(np.where(time<=ref_period[1]))\n",
    "    if i == 0:\n",
    "        ref=ssh_spg[t1:t2]*eof_sign[irun]\n",
    "    else:\n",
    "        exp=ssh_spg[t1:t2]*eof_sign[irun]\n",
    "        r,p=stats.pearsonr(ref,exp)\n",
    "        print('Observations with','{:<18}'.format(longname[irun]),'{:5.2f}'.format(r),'at signif.level','{:5.2f}'.format(1-p))\n",
    "        \n",
    "        \n",
    "print('')\n",
    "print('correlations for 1958-2009')\n",
    "ref_period=(1958,2009)\n",
    "# ORCA025-JRA as reference:\n",
    "for i,irun in enumerate((7,0,1,2,3,4,5)):\n",
    "    npzfile = np.load('ProcessedData/'+projname+'_ts_'+str(irun)+'.npz')\n",
    "    ssh_pc=npzfile['pc']\n",
    "    ssh_spg=ssh_pc[:,imode]\n",
    "    time=npzfile['time']\n",
    "    t1=np.min(np.where(time>=ref_period[0]))\n",
    "    t2=np.max(np.where(time<=ref_period[1]))\n",
    "    if i == 0:\n",
    "        ref=ssh_spg[t1:t2]*eof_sign[irun]\n",
    "        refname=longname[irun]\n",
    "    else:\n",
    "        exp=ssh_spg[t1:t2]*eof_sign[irun]\n",
    "        r,p=stats.pearsonr(ref,exp)\n",
    "        print('{:<12}'.format(refname),'with','{:<18}'.format(longname[irun]),'{:5.2f}'.format(r),'at signif.level','{:5.2f}'.format(1-p))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [conda env:py3_std_maps_eof]",
   "language": "python",
   "name": "conda-env-py3_std_maps_eof-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}