{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "f30aa9dc",
   "metadata": {},
   "source": [
    "# ocean: fx\n",
    "\n",
    "fx ocean grid data\n",
    "\n",
    "```{dropdown}  **Branding suffix of datasets**\n",
    "Datasets with the folloing branding suffix are shown in this page: \\\n",
    "[time]-[vertical]-[horizontal]-[domain]\n",
    "```\n",
    "* ti-u-hxy-sea\n",
    "* ti-u-hxy-u\n",
    "* ti-ol-hxy-sea\n",
    "    - ti: time independent\n",
    "    - u: ocean surface\n",
    "    - ol: ocean levels\n",
    "    - hxy: horizontal curvlinear grids\n",
    "    - sea: ocean domain"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f1f107ce",
   "metadata": {
    "tags": [
     "remove-input"
    ]
   },
   "outputs": [],
   "source": [
    "## Import libraries\n",
    "import glob\n",
    "import logging\n",
    "import os\n",
    "import sys\n",
    "\n",
    "import cartopy.crs as ccrs\n",
    "import cartopy.feature as cfeature\n",
    "import cmocean\n",
    "import matplotlib as mpl\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import xarray as xr\n",
    "from IPython.display import HTML, display\n",
    "from matplotlib.colors import BoundaryNorm, ListedColormap\n",
    "\n",
    "from utils import read_compound_names, read_variables\n",
    "\n",
    "logging.basicConfig(\n",
    "    filename=\"execute.log\",\n",
    "    level=logging.INFO,\n",
    "    format=\"%(asctime)s %(message)s\",\n",
    "    force=True,\n",
    ")\n",
    "log = logging.getLogger(\"ocean_fx\")\n",
    "log.info(\"Starting ocean_fx.ipynb\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f824913d",
   "metadata": {
    "tags": [
     "remove-input"
    ]
   },
   "outputs": [],
   "source": [
    "# load parameters\n",
    "import yaml\n",
    "\n",
    "with open('params.yml', 'r') as f:\n",
    "    param = yaml.safe_load(f)\n",
    "    cmorout = param['cmorout']\n",
    "    source_id = param['source_id']\n",
    "    experiment_id = param['experiment_id']\n",
    "    variant_label = param['variant_label']\n",
    "    version = param['version']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8314513c",
   "metadata": {
    "tags": [
     "remove-input"
    ]
   },
   "outputs": [],
   "source": [
    "# grid label\n",
    "grid_label = 'g???'         # grid label matches 'g202', etc\n",
    "\n",
    "# data path\n",
    "data_path = os.path.join(cmorout, source_id, experiment_id, version)\n",
    "\n",
    "# load grid\n",
    "grid_file = f\"areacello_ti-u-hxy-u_fx_glb_g202_{source_id}_{experiment_id}_{variant_label}.nc\"\n",
    "if os.path.exists(os.path.join(data_path, grid_file)):\n",
    "    grid_file = os.path.join(data_path, grid_file)\n",
    "    with xr.open_dataset(grid_file) as ds:\n",
    "        parea = ds['areacello']\n",
    "else:\n",
    "    grid_file = 'assets/grid.nc'\n",
    "    with xr.open_dataset(grid_file) as ds:\n",
    "        parea = ds['parea']\n",
    "        parea = parea.rename({'y': 'j', 'x': 'i'})\n",
    "\n",
    "grid_file = f\"sftof_ti-u-hxy-u_fx_glb_g202_{source_id}_{experiment_id}_{variant_label}.nc\"\n",
    "if os.path.exists(os.path.join(data_path, grid_file)):\n",
    "    grid_file = os.path.join(data_path, grid_file)\n",
    "    with xr.open_dataset(grid_file) as ds:\n",
    "        pmask = ds['sftof']\n",
    "else:\n",
    "    grid_file = 'assets/grid.nc'\n",
    "    with xr.open_dataset(grid_file) as ds:\n",
    "        pmask = ds['pmask']\n",
    "        pmask = pmask.rename({'y': 'j', 'x': 'i'})\n",
    "\n",
    "# load methods for plotting and set defaults\n",
    "methods =read_variables('assets/methods.txt')\n",
    "#print(methods.keys())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f06c51b7",
   "metadata": {},
   "source": [
    "---\n",
    "```{dropdown}  **List of datasets:** \n",
    "List of compound names of datasets that needs to be cmorized. \\\n",
    "Those datasets which are not cmorized have no link.\n",
    "```\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "67a5a359",
   "metadata": {
    "tags": [
     "remove-input"
    ]
   },
   "outputs": [],
   "source": [
    "# load compound names\n",
    "\n",
    "coords = ('ti-u-hxy-sea', 'ti-u-hxy-u', 'ti-ol-hxy-sea')\n",
    "cnames = read_compound_names('assets/variables.nml')\n",
    "# examples of compound names:\n",
    "# cnames = ['ocean.tos.tavg-u-hxy-sea.mon.glb', 'ocean.ficeberg.tavg-u-hxy-sea.mon.glb']\n",
    "for cname in cnames:\n",
    "    log.info(f\"processing {cname}\")\n",
    "    realm = cname.split('.')[0]\n",
    "    var = cname.split('.')[1]\n",
    "    coord = cname.split('.')[2]\n",
    "    freq = cname.split('.')[3]\n",
    "    region = cname.split('.')[4]\n",
    "\n",
    "    if realm != 'ocean' or coord not in coords:\n",
    "        continue\n",
    "    else:\n",
    "        file_pattern = f\"{var}_{coord}_{freq}_{region}_{grid_label}_{source_id}_{experiment_id}_{variant_label}.nc\"\n",
    "        if not glob.glob(os.path.join(data_path, file_pattern)):\n",
    "            continue\n",
    "        else:\n",
    "            display(HTML(f'<a href=\"#{cname}\">{cname}</a>'))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a2cb0804",
   "metadata": {},
   "source": [
    "---\n",
    "```{dropdown}  **Dataset validated:**\n",
    "Each validated dataset has some text information and some plots (2D and/or 1D)\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "25072041",
   "metadata": {
    "tags": [
     "remove-input"
    ]
   },
   "outputs": [],
   "source": [
    "# loop through compound names and plot\n",
    "for cname in cnames:\n",
    "    mth_vert = 'mean'\n",
    "    mth_ts = 'mean'\n",
    "    mth_cmap = 'mpl.colormaps[\"viridis\"]'\n",
    "    if cname not in methods:\n",
    "        print(f\"{cname} not found in methods.txt, using default methods for plotting.\")\n",
    "    else:\n",
    "        if methods[cname] is not None:\n",
    "            if 'vertical' in methods[cname]:\n",
    "                mth_vert = methods[cname]['vertical']\n",
    "\n",
    "            if 'timeseries' in methods[cname]:\n",
    "                mth_ts = methods[cname]['timeseries']\n",
    "\n",
    "            if 'cmap' in methods[cname]:\n",
    "                mth_cmap = methods[cname]['cmap']\n",
    "\n",
    "    realm = cname.split('.')[0]\n",
    "    var = cname.split('.')[1]\n",
    "    coord = cname.split('.')[2]\n",
    "    freq = cname.split('.')[3]\n",
    "    region = cname.split('.')[4]\n",
    "\n",
    "    if realm != 'ocean' or coord not in coords:\n",
    "        continue\n",
    "\n",
    "    file_pattern = f\"{var}_{coord}_{freq}_{region}_{grid_label}_{source_id}_{experiment_id}_{variant_label}.nc\"\n",
    "    data_files = sorted(glob.glob(os.path.join(data_path, file_pattern)))\n",
    "    if data_files:\n",
    "        log.info(f\"{cname}\")\n",
    "        with xr.open_mfdataset(data_files) as ds:\n",
    "            if var in ds:\n",
    "                data = ds[var]\n",
    "                lat = ds['latitude']\n",
    "                lon = ds['longitude']\n",
    "            else:\n",
    "                continue\n",
    "    else:\n",
    "        continue\n",
    "\n",
    "    # remove the las row of grid info if the data has not the last row\n",
    "    if data.sizes['j'] == 384:\n",
    "        parea = parea.isel(j=slice(0, 384))\n",
    "        pmask = pmask.isel(j=slice(0, 384))\n",
    "\n",
    "    display(HTML(f'<div id=\"{cname}\"></div>'))\n",
    "    print(f'\\033[1m{cname}\\033[0m')\n",
    "    print(f'long name: {data.long_name} ({data.units})')\n",
    "#    print(f'timeseries aggregation method for 3D->1D plot: {mth_ts}')\n",
    "    print(f'original_name: {data.attrs[\"original_name\"]} -> {var}')\n",
    "    if 'history' in data.attrs:\n",
    "        print(f'history: {data.attrs[\"history\"]}')\n",
    "    if 'comment' in data.attrs:\n",
    "        print(f'comment: {data.attrs[\"comment\"]}')\n",
    "\n",
    "    if coord == 'ti-ol-hxy-sea':\n",
    "        if mth_vert == 'mean':\n",
    "            data2d = data.mean(dim='lev',keep_attrs=True).where(pmask == 1)\n",
    "        elif mth_vert == 'sum':\n",
    "            data2d = data.sum(dim='lev',keep_attrs=True).where(pmask == 1)\n",
    "        else:\n",
    "            raise ValueError(f'Unsupported vertical aggregation method: {mth_vert}')\n",
    "        print(f'vertical aggregation method for 3D->2D plot: {mth_vert}')\n",
    "    else:\n",
    "        data2d = data \n",
    "\n",
    "\n",
    "    #ax1 = fig.add_subplot(1, 2, 1, projection=proj)\n",
    "    #fig, (ax1, ax2) = plt.subplots( nrows=1, ncols=2, figsize=(12, 5), gridspec_kw={'width_ratios': [2, 1]}, subplot_kw={'projection': proj})\n",
    "\n",
    "    proj = ccrs.PlateCarree(central_longitude=90.0)\n",
    "    fig, ax = plt.subplots(1, figsize=(11.7, 4), dpi=96, subplot_kw={\"projection\": proj})\n",
    "\n",
    "    # plot 2D map\n",
    "    #ax1 = fig.add_subplot(gs[0], projection=proj)\n",
    "    # rescale\n",
    "    vmin, vmax = data2d.quantile([0.001, 0.999], dim=None).values\n",
    "    if mth_cmap == 'cmocean.cm.balance':\n",
    "        vmax = max(abs(vmin),abs(vmax))\n",
    "        vmin = -vmax\n",
    "    # Define the discrete levels (boundaries) and the corresponding colors\n",
    "    levels = np.linspace(vmin, vmax, 21)\n",
    "    n_levels = 20\n",
    "    colors = eval(mth_cmap)(np.linspace(0,1,n_levels))\n",
    "    cmap = ListedColormap(colors)\n",
    "\n",
    "    # Create a BoundaryNorm to map data to discrete color indices\n",
    "    norm = BoundaryNorm(levels, ncolors=len(cmap.colors), clip=False)\n",
    "    \n",
    "    # Plot the mesh\n",
    "    if vmin == vmax:\n",
    "        pm = ax.pcolormesh(lon, lat, data2d, cmap=cmap,\n",
    "                            transform=ccrs.PlateCarree(), shading='auto', rasterized=True)\n",
    "    else:\n",
    "        pm = ax.pcolormesh(lon, lat, data2d, cmap=cmap, norm=norm,\n",
    "                            transform=ccrs.PlateCarree(), shading='auto', rasterized=True)\n",
    "\n",
    "    # Add map features\n",
    "    #ax.stock_img()\n",
    "    ax.add_feature(cfeature.LAND, facecolor='lightgray')\n",
    "\n",
    "    gl = ax.gridlines(ylocs=range(-90, 90, 30), draw_labels=True)\n",
    "    gl.ylocator = mpl.ticker.FixedLocator(range(-90,90,30))\n",
    "\n",
    "    # Add colorbar\n",
    "    cb = plt.colorbar(pm, ax=ax, fraction=0.4, shrink=0.8, label='[yr]')\n",
    "    cb.set_label(label=data.units, size=14)\n",
    "    cb.ax.tick_params(labelsize=12)\n",
    "    plt.tight_layout()\n",
    "        \n",
    "    #fig, ax = plot_map2d(lon, lat, data2d.mean(dim='time'), proj='PlateCarree', norm=None, cmap=eval(mth_cmap))\n",
    "    ax.coastlines(resolution='110m')\n",
    "    ax.add_feature(cfeature.LAND, facecolor='lightgray')\n",
    "    plt.title(data2d.long_name)\n",
    "\n",
    "\n",
    "    plt.title(data2d.long_name)\n",
    "    plt.show()\n",
    "\n",
    "    del data, data2d\n",
    "    del ax, pm, cb, gl, fig"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "cmip7validate-env",
   "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": "3.13.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}