{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "a1b2c3d4",
   "metadata": {},
   "source": [
    "This is part of a tutorial series. We recommend following them in order, starting with [Part 0: Welcome to `musica`](0.%20Welcome%20to%20MUSICA.ipynb)."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b2c3d4e5",
   "metadata": {},
   "source": [
    "# `tuv-x`: Computing a Subset of Photolysis Rates\n",
    "\n",
    "The TS1/TSMLT configuration of TUV-x includes 50+ photolysis reactions covering species from the troposphere through the stratosphere, including halogens, sulfur compounds, and many organic species. For simulations that only need a smaller set of reactions — for example, a tropospheric mechanism that doesn't include halogen chemistry — computing the full set is unnecessary work.\n",
    "\n",
    "TUV-x computes photolysis rate constants by:\n",
    "1. Solving the radiation field (the expensive radiative transfer calculation, which is independent of the photolysis reaction list)\n",
    "2. For each photolysis reaction: integrating the product of the radiation field, the species cross section, and the quantum yield over the wavelength grid\n",
    "\n",
    "Step 2 is repeated for every reaction in the configuration. Removing unused reactions from the config directly reduces the number of these integrals.\n",
    "\n",
    "This tutorial shows how to select only the reactions your mechanism needs using the `config_string` API."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c3d4e5f6",
   "metadata": {},
   "source": [
    "## 1. Listing Available Reactions\n",
    "\n",
    "Let's start by seeing what reactions are defined in the TS1/TSMLT configuration."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "d4e5f6a7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TS1/TSMLT has 73 photolysis reactions:\n",
      "  jo2_a                 (O2 + hv -> O + O1D)\n",
      "  jo2_b                 (O2 + hv -> O + O)\n",
      "  jo3_a                 (O3 + hv -> O2 + O(1D))\n",
      "  jo3_b                 (O3 + hv -> O2 + O(3P))\n",
      "  jn2o                  (N2O + hv -> N2 + O(1D))\n",
      "  jno2                  (NO2 + hv -> NO + O(3P))\n",
      "  jn2o5_a               (N2O5 + hv -> NO2 + NO3)\n",
      "  jn2o5_b               (N2O5 + hv -> NO + O + NO3)\n",
      "  jhno3                 (HNO3 + hv -> OH + NO2)\n",
      "  jno3_a                (NO3 + hv -> NO2 + O(3P))\n",
      "  jno3_b                (NO3 + hv -> NO + O2)\n",
      "  jch3ooh               (CH3OOH + hv -> CH3O + OH)\n",
      "  jch2o_a               (CH2O + hv -> H + HCO)\n",
      "  jch2o_b               (CH2O + hv -> H2 + CO)\n",
      "  jh2o2                 (H2O2 + hv -> OH + OH)\n",
      "  jch3cho               (CH3CHO + hv -> CH3 + HCO)\n",
      "  jpan                  (PAN + hv -> 0.6*CH3CO3 + 0.6*NO2 + 0.4*CH3O2 + 0.4*NO3 + 0.4*CO2)\n",
      "  jmvk                  (MVK + hv -> 0.7*C3H6 + 0.7*CO + 0.3*CH3O2 + 0.3*CH3CO3)\n",
      "  jacet                 (CH3COCH3 + hv -> CH3CO + CH3)\n",
      "  jmgly                 (CH3COCHO + hv -> CH3CO3 + CO + HO2)\n",
      "  jglyald               (GLYALD + hv -> 2*HO2 + CO + CH2O)\n",
      "  jbrcl                 (BrCl + hv -> Br + Cl)\n",
      "  jbro                  (BrO + hv -> Br + O)\n",
      "  jbrono2_a             (BrONO2 + hv -> Br + NO3)\n",
      "  jbrono2_b             (BrONO2 + hv -> BrO + NO2)\n",
      "  jccl4                 (CCl4 + hv -> Products)\n",
      "  jcf2clbr              (CF2BrCl + hv -> Products)\n",
      "  jcf3br                (CF3Br + hv -> Products)\n",
      "  jcfcl3                (CCl3F + hv -> Products)\n",
      "  jcfc113               (CFC-113 + hv -> Products)\n",
      "  jcfc114               (CFC-114 + hv -> Products)\n",
      "  jcfc115               (CFC-115 + hv -> Products)\n",
      "  jcf2cl2               (CCl2F2 + hv -> Products)\n",
      "  jch2br2               (CH2BR2 + hv -> 2*BR)\n",
      "  jch3br                (CH3Br + hv -> Products)\n",
      "  jch3ccl3              (CH3CCl3+hv->Products)\n",
      "  jch3cl                (CH3Cl + hv -> Products)\n",
      "  jchbr3                (CHBr3 + hv -> Products)\n",
      "  jcl2                  (Cl2 + hv -> Cl + Cl)\n",
      "  jcl2o2                (ClOOCl + hv -> Cl + ClOO)\n",
      "  jclo                  (ClO + hv -> Cl + O)\n",
      "  jclono2_a             (ClONO2 + hv -> Cl + NO3)\n",
      "  jclono2_b             (ClONO2 + hv -> ClO + NO2)\n",
      "  jcof2                 (CF2O + hv -> Products)\n",
      "  jcofcl                (CClFO + hv -> Products)\n",
      "  jh2402                (H2402 + hv -> 2*BR + 2*COF2)\n",
      "  jhcfc141b             (HCFC-141b + hv -> Products)\n",
      "  jhcfc142b             (HCFC-142b + hv -> Products)\n",
      "  jhcfc22               (HCFC-22 + hv -> Products)\n",
      "  jhcl                  (HCl + hv -> H + Cl)\n",
      "  jhobr                 (HOBr + hv -> OH + Br)\n",
      "  jhocl                 (HOCl + hv -> HO + Cl)\n",
      "  joclo                 (OClO + hv -> Products)\n",
      "  jho2no2_a             (HNO4 + hv -> OH + NO3)\n",
      "  jho2no2_b             (HNO4 + hv -> HO2 + NO2)\n",
      "  jmacr_a               (CH2=C(CH3)CHO->1.34HO2+0.66MCO3+1.34CH2O+CH3CO3)\n",
      "  jmacr_b               (CH2=C(CH3)CHO->0.66OH+1.34CO)\n",
      "  jhyac                 (CH2(OH)COCH3->CH3CO3+HO2+CH2O)\n",
      "  jh2o_a                (H2O + hv -> OH + H)\n",
      "  jh2o_b                (H2O + hv -> H2 + O1D)\n",
      "  jh2o_c                (H2O + hv -> 2*H + O)\n",
      "  jch4_a                (CH4 + hv -> H + CH3O2)\n",
      "  jch4_b                (CH4 + hv -> 1.44*H2 + 0.18*CH2O + 0.18*O + 0.33*OH + 0.33*H + 0.44*CO2 + 0.38*CO + 0.05*H2O)\n",
      "  jco2                  (CO2 + hv -> CO + O)\n",
      "  jhbr                  (HBR + hv -> BR + H)\n",
      "  jhf                   (HF + hv -> H + F)\n",
      "  jsf6                  (SF6 + hv -> sink)\n",
      "  jh2so4                (H2SO4 + hv -> SO3 + H2O)\n",
      "  jocs                  (OCS + hv -> S + CO)\n",
      "  jso                   (SO + hv -> S + O)\n",
      "  jso2                  (SO2 + hv -> SO + O)\n",
      "  jso3                  (SO3 + hv -> SO2 + O)\n",
      "  jno_i                 (NO + hv -> NOp + e)\n"
     ]
    }
   ],
   "source": [
    "import json\n",
    "from musica.tuvx import vTS1\n",
    "\n",
    "config_path = vTS1.config_file_path()\n",
    "\n",
    "with open(config_path) as f:\n",
    "    full_config = json.load(f)\n",
    "\n",
    "all_reactions = full_config[\"photolysis\"][\"reactions\"]\n",
    "print(f\"TS1/TSMLT has {len(all_reactions)} photolysis reactions:\")\n",
    "for r in all_reactions:\n",
    "    reaction_str = r.get(\"__reaction\", \"\")\n",
    "    print(f\"  {r['name']:20s}  ({reaction_str})\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e5f6a7b8",
   "metadata": {},
   "source": [
    "## 2. Filtering the Configuration\n",
    "\n",
    "To run TUV-x with a subset of these reactions, we:\n",
    "\n",
    "1. Load the full TS1/TSMLT JSON configuration\n",
    "2. Filter `config[\"photolysis\"][\"reactions\"]` to keep only the reactions we need\n",
    "3. Pass the modified configuration as a JSON string to the `TUVX` constructor via `config_string`\n",
    "\n",
    "The radiative transfer calculation (the dominant cost) is unaffected — it depends on the grids and atmospheric profiles, not the photolysis reaction list. Only the per-reaction integration step is reduced.\n",
    "\n",
    "> **Note on file paths**: When using `config_string`, relative paths in the configuration are resolved against the current working directory, not the original config file location. The helper function below handles this by temporarily changing to the config directory."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "f6a7b8c9",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import json\n",
    "from musica.tuvx import vTS1\n",
    "from musica.tuvx.tuvx import TUVX\n",
    "from musica.tuvx.grid_map import GridMap\n",
    "from musica.tuvx.profile_map import ProfileMap\n",
    "from musica.tuvx.radiator_map import RadiatorMap\n",
    "\n",
    "\n",
    "def get_tuvx_with_subset(reaction_names: list[str]) -> TUVX:\n",
    "    \"\"\"Create a TUV-x TS1/TSMLT instance that computes only the specified reactions.\"\"\"\n",
    "    config_path = vTS1.config_file_path()\n",
    "\n",
    "    with open(config_path) as f:\n",
    "        config = json.load(f)\n",
    "\n",
    "    all_reactions = config[\"photolysis\"][\"reactions\"]\n",
    "    all_names = {r[\"name\"] for r in all_reactions}\n",
    "\n",
    "    unknown = set(reaction_names) - all_names\n",
    "    if unknown:\n",
    "        raise ValueError(\n",
    "            f\"Unknown reaction names: {sorted(unknown)}.\\n\"\n",
    "            f\"Available: {sorted(all_names)}\"\n",
    "        )\n",
    "\n",
    "    requested = set(reaction_names)\n",
    "    config[\"photolysis\"][\"reactions\"] = [\n",
    "        r for r in all_reactions if r[\"name\"] in requested\n",
    "    ]\n",
    "\n",
    "    # Build grids, profiles, and radiators (same as the full TS1/TSMLT setup)\n",
    "    grids = GridMap()\n",
    "    grids[\"height\", \"km\"] = vTS1.height_grid()\n",
    "    grids[\"wavelength\", \"nm\"] = vTS1.wavelength_grid()\n",
    "\n",
    "    height = grids[\"height\", \"km\"]\n",
    "    wavelength = grids[\"wavelength\", \"nm\"]\n",
    "\n",
    "    profiles = ProfileMap()\n",
    "    profiles[\"air\", \"molecule cm-3\"] = vTS1.profile(\"air\", height)\n",
    "    profiles[\"O3\", \"molecule cm-3\"] = vTS1.profile(\"O3\", height)\n",
    "    profiles[\"O2\", \"molecule cm-3\"] = vTS1.profile(\"O2\", height)\n",
    "    profiles[\"temperature\", \"K\"] = vTS1.profile(\"temperature\", height)\n",
    "    profiles[\"surface albedo\", \"none\"] = vTS1.profile(\"surface albedo\", wavelength)\n",
    "    profiles[\"extraterrestrial flux\", \"photon cm-2 s-1\"] = vTS1.profile(\n",
    "        \"extraterrestrial flux\", wavelength)\n",
    "\n",
    "    radiators = RadiatorMap()\n",
    "    radiators[\"aerosol\"] = vTS1.radiator(\"aerosol\", height, wavelength)\n",
    "\n",
    "    # Temporarily change to the config directory so relative paths in the config work\n",
    "    config_dir = os.path.dirname(os.path.abspath(config_path))\n",
    "    original_cwd = os.getcwd()\n",
    "    os.chdir(config_dir)\n",
    "    try:\n",
    "        tuvx = TUVX(\n",
    "            grid_map=grids,\n",
    "            profile_map=profiles,\n",
    "            radiator_map=radiators,\n",
    "            config_string=json.dumps(config)\n",
    "        )\n",
    "    finally:\n",
    "        os.chdir(original_cwd)\n",
    "\n",
    "    return tuvx"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a7b8c9d0",
   "metadata": {},
   "source": [
    "## 3. Running with a Tropospheric Subset\n",
    "\n",
    "Here we select 11 reactions relevant to a basic tropospheric mechanism, skipping the halogen, sulfur, and stratospheric-only reactions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "b8c9d0e1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creating TUV-x with 11 of 73 reactions...\n",
      "\n",
      "Surface-level photolysis rate constants (altitude = 0 km):\n",
      "  jo2_a       : 0.000e+00 s⁻¹\n",
      "  jo2_b       : 3.608e-27 s⁻¹\n",
      "  jo3_a       : 4.882e-05 s⁻¹\n",
      "  jo3_b       : 5.015e-04 s⁻¹\n",
      "  jno2        : 1.105e-02 s⁻¹\n",
      "  jno3_a      : 2.108e-01 s⁻¹\n",
      "  jno3_b      : 2.213e-02 s⁻¹\n",
      "  jhno3       : 8.321e-07 s⁻¹\n",
      "  jh2o2       : 9.150e-06 s⁻¹\n",
      "  jch2o_a     : 4.145e-05 s⁻¹\n",
      "  jch2o_b     : 5.913e-05 s⁻¹\n"
     ]
    }
   ],
   "source": [
    "import xarray as xr\n",
    "\n",
    "tropospheric_reactions = [\n",
    "    \"jo2_a\",    # O2 + hv -> O + O(1D)\n",
    "    \"jo2_b\",    # O2 + hv -> O + O\n",
    "    \"jo3_a\",    # O3 + hv -> O2 + O(1D)\n",
    "    \"jo3_b\",    # O3 + hv -> O2 + O(3P)\n",
    "    \"jno2\",     # NO2 + hv -> NO + O(3P)\n",
    "    \"jno3_a\",   # NO3 + hv -> NO2 + O(3P)\n",
    "    \"jno3_b\",   # NO3 + hv -> NO + O2\n",
    "    \"jhno3\",    # HNO3 + hv -> OH + NO2\n",
    "    \"jh2o2\",    # H2O2 + hv -> OH + OH\n",
    "    \"jch2o_a\",  # CH2O + hv -> H + HCO\n",
    "    \"jch2o_b\",  # CH2O + hv -> H2 + CO\n",
    "]\n",
    "\n",
    "print(f\"Creating TUV-x with {len(tropospheric_reactions)} of {len(all_reactions)} reactions...\")\n",
    "tuvx_subset = get_tuvx_with_subset(tropospheric_reactions)\n",
    "\n",
    "results: xr.Dataset = tuvx_subset.run(sza=0.0, earth_sun_distance=1.0)\n",
    "\n",
    "print(\"\\nSurface-level photolysis rate constants (altitude = 0 km):\")\n",
    "for name in tropospheric_reactions:\n",
    "    rate = float(results[\"photolysis_rate_constants\"].sel(reaction=name).values[0])\n",
    "    print(f\"  {name:12s}: {rate:.3e} s\\u207b\\u00b9\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c9d0e1f2",
   "metadata": {},
   "source": [
    "## 4. Comparing Against the Full Configuration\n",
    "\n",
    "The subset results should be identical to the full TS1/TSMLT configuration for the same reactions, because the radiative transfer calculation is unchanged. Let's verify this."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "d0e1f2a3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Checking that subset results match the full TS1/TSMLT configuration:\n",
      "  jo2_a       : OK\n",
      "  jo2_b       : OK\n",
      "  jo3_a       : OK\n",
      "  jo3_b       : OK\n",
      "  jno2        : OK\n",
      "  jno3_a      : OK\n",
      "  jno3_b      : OK\n",
      "  jhno3       : OK\n",
      "  jh2o2       : OK\n",
      "  jch2o_a     : OK\n",
      "  jch2o_b     : OK\n",
      "\n",
      "All rates match: True\n"
     ]
    }
   ],
   "source": [
    "from musica.tuvx import vTS1\n",
    "\n",
    "tuvx_full = vTS1.get_tuvx_calculator()\n",
    "results_full: xr.Dataset = tuvx_full.run(sza=0.0, earth_sun_distance=1.0)\n",
    "\n",
    "print(\"Checking that subset results match the full TS1/TSMLT configuration:\")\n",
    "all_match = True\n",
    "for name in tropospheric_reactions:\n",
    "    subset_vals = results[\"photolysis_rate_constants\"].sel(reaction=name).values\n",
    "    full_vals = results_full[\"photolysis_rate_constants\"].sel(reaction=name).values\n",
    "    import numpy as np\n",
    "    match = np.allclose(subset_vals, full_vals, rtol=1e-10)\n",
    "    print(f\"  {name:12s}: {'OK' if match else 'MISMATCH'}\")\n",
    "    all_match = all_match and match\n",
    "\n",
    "print(f\"\\nAll rates match: {all_match}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "g3h4i5j6",
   "metadata": {},
   "source": [
    "## 5. Timing Comparison\n",
    "\n",
    "We can use the `%timeit` magic to measure how much faster the `run()` call is with the subset configuration.\n",
    "Both instances are already initialised — the grids, profiles, and radiative transfer setup happen at construction time.\n",
    "What we are timing here is the per-call cost: solving the radiation field plus the per-reaction integrations.\n",
    "The radiation field solve is identical in both cases; the difference comes entirely from the number of\n",
    "cross section / quantum yield integrals."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "h4i5j6k7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Timing full TS1/TSMLT (73 reactions)...\n",
      "14.3 ms ± 113 μs per loop (mean ± std. dev. of 7 runs, 100 loops each)\n",
      "\n",
      "Timing subset (11 reactions)...\n",
      "3.9 ms ± 116 μs per loop (mean ± std. dev. of 7 runs, 100 loops each)\n"
     ]
    }
   ],
   "source": [
    "print(f\"Timing full TS1/TSMLT ({len(tuvx_full.photolysis_rate_names)} reactions)...\")\n",
    "%timeit tuvx_full.run(sza=0.0, earth_sun_distance=1.0)\n",
    "\n",
    "print(f\"\\nTiming subset ({len(tuvx_subset.photolysis_rate_names)} reactions)...\")\n",
    "%timeit tuvx_subset.run(sza=0.0, earth_sun_distance=1.0)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e1f2a3b4",
   "metadata": {},
   "source": [
    "## 6. Vertical Profiles\n",
    "\n",
    "Let's plot vertical profiles for a few of the subset reactions to confirm the results look physically reasonable."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "f2a3b4c5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABKUAAAJOCAYAAABm7rQwAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjgsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvwVt1zgAAAAlwSFlzAAAPYQAAD2EBqD+naQAA5XBJREFUeJzs3QWYE9fXBvB3k3WFXRwWd2txKUVaCi0VqH5tqVIX6kZL3e1fSt3djRpQKC2uxV0Xd1nXyPecO5mQ9WSJTJL39zwh2SRkZ2eSnbPnnntuhN1ut4OIiIiIiIiIiMiPTP78ZkRERERERERERIJJKSIiIiIiIiIi8jsmpYiIiIiIiIiIyO+YlCIiIiIiIiIiIr9jUoqIiIiIiIiIiPyOSSkiIiIiIiIiIvI7JqWIiIiIiIiIiMjvmJQiIiIiIiIiIiK/Y1KKiIiIiIiIiIj8jkkpIgoJTzzxBCIiInD48GEYxfbt29U2ffrppz55/ebNm+Oaa67xyWsTERERBcrMmTNVDPXjjz8a6iD4MvaS15XXJwo3TEoRkcfWrl2LK664Ao0bN0ZMTAwaNWqE0aNHq/sreu7FF1+Mli1bIj4+HnXq1MHAgQPx+++/e5Rs0i/yGh07dsT48eORnZ3t06P39ttv+yyhRERERIEn53mJL2JjY7Fnz55yjw8ePBidO3cud39JSQkmTpyIXr16ISkpCYmJieq23CeP+YLETiaTCfv376/0OZLUOOecczxO9HgS250IfRv0S1RUlIoRr7rqKmzbtg2+NHnyZBVXEpGxRAZ6A4gouPz888+47LLLkJqaiuuuuw4tWrRQFUEfffSRCnK+/fZbnH/++c7n79ixAzk5Obj66qtVgJOfn4+ffvoJ5513Ht577z3ceOONbn3fd955RwV8ubm5mDZtGp599ln8888/mDdvngpqfJWUkiSaUauRNm7cqIJTIiIiOjFFRUV44YUX8MYbb1T73Ly8PJx99tmYNWuWSgBJnCDn46lTp+LOO+9UsdKff/6JhIQErx4Wec0ePXqgQYMGAY3tvOGOO+5QSTxJ4C1btgzvv/+++vlWr16t4kVfJaXeeustwyamPvjgA9hstkBvBpHfMSlFRG7bunUrrrzySjWiNXv2bNStW9f5mARhp556qnp81apV6jlixIgR6uLq9ttvV0HV//73P7eTUhdddJFKEImbb74ZF154oQqiFi5ciH79+oXlUZSRTCIiIjpxJ598skoKjBs3rtqkyD333KMSUpLAkphGd8stt6ikh9x33333qQE1bydVxowZE/DYzhvkdSW2E9deey3atm2rElWfffaZOgbhSKrGiMIRh9iJyG0vv/yyqnSS0SzXoEVIwkgqn2T08KWXXqrydcxmM9LT05GZmVnjvX/aaaep64yMjFL3y2vKiGWtWrWQkpKiAh3ZZlcWiwVPP/00WrVqpRI7Uur+8MMPq1FSndwnJesSdOol5lLCr5MSc5mWKKOKMqWwb9++aoSvKp988ol6neXLl5d77LnnnlP7RZ86sHnzZpV4k9FQmVLQpEkTXHrppcjKyiq1ja5VXDLa+OSTT6JNmzbq/6SlpWHAgAGYPn26B3uWiIgo/EgcYLVaVbVUVXbv3q0qiCQOcU1I6W677TYMGTIEH374oXqu6/n/448/Lnful/sl2VQdqSDatWuXqtAKZGxXUFCA9u3bq4vc1h09ehQNGzZE//791X70VlwnlUNSHS9xkMQ2p59+OrZs2VLu///www9qwDMuLk5tt0xFdJ2OKfGSJAyF6/RBnfyM9957r4pPJTZs164dXnnlFdjt9kq3WWJBeY3XXnut3GPz589Xj33zzTfqa5k1cNddd6nYTV6/Xr16OOOMM1SVWFU9paRKTX4umSKanJyMLl264PXXX3drnxIFCyaliMijXgZyspTRrYpIryh5vKLkjJzspQm5jMjJyXvKlCkqsKgpeR0hiRdXl1xyiTrxP//88+q29IqQRI2r66+/Ho899hi6d++utmXQoEHq+ZL00U2YMEEFQBJ0ffHFF+ryyCOPqMcOHDiggq6//voLt956qwqWCgsL1ZTEX375pdJtlhFBCZa++uqrco/JfZL0kl4OxcXFGD58uKoCGzt2rAqipKJMgp+qEnlSji4/qwTDb775ptrepk2blgp4iIiIqDyZsiZ9jaRaau/evZXuIolfJOkiz62MPCYDYDKdT8gAmUzzkworSSzpSSY5Z8t0ubIV5RWRxJUkMnr27Fntc2WQSmKushfXga2axnYSx0g1kySG9LhIT8bJ60vcJYNs3orrJEkosZVUnkkFlcRG0uvKlXxPifnk+0o8d8MNN6hqehmY0+Omm266SSWBhB7XyUVI4kliOIkJzzzzTFXJL0mp+++/Xx2zykjl2CmnnFJpXCeJpJEjRzqr/KVyTgYcpT2E/DyyL9evX1/p68ugokyrrF27Nl588UW1LyRWlNYVRCHFTkTkhszMTBkqso8cObLK55133nnqednZ2aXuv+mmm9T9cjGZTPaLLrrIfvTo0Wq/7+OPP67+z8aNG+2HDh2yZ2Rk2N977z17TEyMvX79+va8vLxSzxszZkyp/3/++efb09LSnF+vWLFCPe/6668v9bz77rtP3f/PP/847+vUqZN90KBB5bbprrvuUs+dM2eO876cnBx7ixYt7M2bN7dbrVZ1n2yrPO+TTz5xPu+yyy6zN2rUyPkcsWzZslLPW758ufr6hx9+qHLfNGvWzH711Vc7vz7ppJPsZ599dpX/h4iIiI6Tc6+cc5csWWLfunWrPTIy0n7HHXc4H5c4QOKBsjGAnKsro5/X77nnHud9+/bts6emptrPOOMMe1FRkb1bt272pk2b2rOystw6HKeeemqpc35VsYEeb1V20eOLE4ntxo0bp+K52bNnq9eTxydMmFDt9v3777/quR9//LGK6/bu3Wv/888/VfwUERGhjoPr8zp06KD2l+71119X969evVp9XVxcbK9Xr569c+fO9oKCAufz/vjjD/W8xx57zHnfbbfdpu4ra9KkSer+Z555ptT9EqvKNm3ZsqXU/nU9DhKTyv9dv3698z7Zpjp16pR6XkpKivr+VZHny+vr7rzzTntycrLdYrFU+f+Igh0rpYjILVJ9JGTUpyr642VXxpOSZRnxkdG1s846S40ySkWQu2TESsrKZSRTRrtat26tRu1k6pwrGYlyJSN/R44ccW6PXiJfduRLSrZFdVPw9Nfo3bu3GoHTSRN2qWaSxqDr1q2rcvRURmD//fffUqNpMlomo2dCph0KqcQqO/WwKjJlUaYcytQ/IiIi8oxUvkj/JJnKtm/fvhrHQxXFQjIdXyqfJRaS2GTFihVqOp9MyaqOVPssWLDA7al7ffr0Ud+n7EWmo3n6s1T280h1dqdOndRCNlI1LlXn0hPKXdIbS+I66d8lP5dU1EuMWLYSTKrMoqOjnV/rFV36Sn3//fcfDh48qLZBpvfp5DWl2t3duE6qrMpuv8SGUkUl1XGVkQot+b6u1VISv0llmkwhdI3RFi1aVGUVXlnyf2S/sA0DhTompYjILXpAogcwlakswJHAYOjQoSop88cff6hV9M4999wq5+q7khX75KQsSwlLyfiaNWvUHPuyZLqaKyl5FseOHXOuBigr5EhSy5UEi3Lyl8erI8+RJFlZHTp0cD5eGSkdl54LevAivRKk34CUd+v7TBJvkjSTfhTSF0Gm8kkgW1HZvaunnnpKBa7SLFR6DkjZuTQmJSIiIveMHz9eTb2rrLeUO/FQZbGQtAmQZMnixYvVFDN32xhIkkMMGzbMredL7CAxV9lL2bjpRGI7SRRJUk16QMnjet8sd0kbBYnrZCVliVUkWSMJwZrEdaKiuExiT3fjOkmOlT1e7sR1EjtKPPv1118775MYT9ox6H2yhPTkkthVelbJwKYk9fTEWmUk0SYxnQzmSksJSeTpU0KJQgmTUkTkFqnekWRKdUkOeVxOxNWN/El/pSVLlmDTpk1ufX/paSABlYzESYPyylTWx6Bs8suTwMmbZPsuv/xylWSTPlRSMSWBmOtomnj11VfVvpTGq9JIVEbvZERSb5pa2T6SngwSJHbu3FkltaRvllwTERGRe9VSck6urFpKT1RUFQ/pj3Xs2LHU/VK5LZU9QqqqZWDKHVLJI72L9Epqo8R2erJM4hlPq7Rl8EziOumDKbcjIyNPKK4LJBlwlQSTNDeXBN1vv/2mekHJIKhrRZU8R1ZslASYNJiXuK6qKizpISYVdfJ60vNKYkZJUEl1GlEoYVKKiNwmTTplRGzu3LkVPj5nzhw1fU2eVx19xZbqqn+8rVmzZioILBs8SfNyqTKSx6tLXMlzNm7cWO7+DRs2OB+vLniREnhpLiqjaVK+LtVQZUmQJiO2skSz7FtZRebdd9+t8rVlNUApdZfqK2mm2rVrVzUaR0RERJ5VS0lz6bIkKSCJEr1JdkU+//xzlWSRptmupBm4vhiLxFKyqEp1JPki1THeXnXvRGM7SVRJhbbEHN26dVOLyPg7pnONuSqKy+Q+d+M6GSAsWzHmblwnx1liOYnppCm7tF6oqOpLEoBS/TRp0iS1z6WpuyyWUxWpSJNKLGmOLgOP0sJC3l8VrUBIFKyYlCIit8l0MOl9JCdEGe1zJUsBSz8n6fEkz9PJPP+KVoWRE6q8VtlRRF/TV7gpGwjKSivCNehLSEiocLU7eQ0pvZf+DjqZ8y+jqrJCTXU/kySK5CIVTFIxJeX8riOEkrCSYLhsgkpG3IqKiip93bLHRPpcyTTFqv4PERERlSYV2VIt9d5772H//v2lHpPpV5KI+fvvv9VqamXJ4JFMSZNV9WTKle7HH3/Ed999p6YFPvTQQ+rcL8mv6irGpapcYilfJaVqEttJHHfNNdeoip/XX39drX4ng3t33303/E16UElFkex313hHKpBkZbuycZ0oG9tJXCe9TmXlYleyGp8ksiQRWRWJ4aQy6vvvv1f7QmI2ifN08tplE3ayzbL/PInrJA7UX5exHYWSiuskiYgq0KZNG9WEUpbilROuBFzS/0hG0D766CPV1FEqdFyn10mQI0kWmVompd8S3MlIkow+yRQ1SZz400knnaTKniWBJEGJTAeUBJP8XKNGjVJl5DrpvSAB5zPPPKOSOxJASH8ACSbl55QgRabVSXWS/H8Z9ZIkk2u5dlXVUrIcsCg7dU+C2dtvvx0XX3yx6iUgCSoZkZWRWb0ZekUkGSZLBct2yzbJFAEJguW1iIiIyH2PPPKIOvdKtY1MsyqbrJA4RqpepIpJr4iS6Wy//vqrii0kxtFJUumWW25RMYZ+TpYEiEzHkuSOVClVFjtIo253Brz8GdtJXCTTymbMmKH6MEmiRHpESZJN2jPoA4D+EBUVpSraJFEo+12SQ5Igk2SZ7DfXRJneU0tiN6lQl7hKkoNSiSTHRo65/NwSK06bNk0dS1mop6q2Ea5x3cSJE9UxLVthJxVYkqCUfSOvLbGvJDUl4ej6PilLqs8kMSixp/x/6W0l0/9OPvlk5zRSopAQ6OX/iCj4rFq1yn7ZZZfZGzZsaI+KirI3aNBAfa0vz+vqm2++sQ8dOtRev359tcxy7dq11de//vqrW9/r8ccfV0vtyrLBNXmevtRzRkaG876SkhL7k08+aW/RooXa/vT0dLW0cWFhYan/u3//fvvZZ59tT0pKUq8hy0LrZNloWSq4Vq1a9tjYWHvv3r3V8sOu5HvK/5NtKEuWhjabzfa2bduWe2zbtm32MWPG2Fu1aqVeW5aQHjJkiP3vv/8u9byyyxLLUsayHbJNcXFx9vbt29ufffZZtTQxERERlafHCUuWLCn3mJxj5bFOnTqVe6yoqMj+2muv2Xv06GFPSEiwx8fH27t3726fMGFCufPuBRdcoGKJ7du3l7pfYiF5/RdffLHSQ9OzZ0/7rbfe6vahk9hAYpeK/Pvvv+r7/fDDDzWO7ZYuXariubFjx5a632Kx2Hv16mVv1KiR/dixY5VuX1Xb4M7zKoutvvvuO3u3bt3sMTExKm4aPXq0fffu3eW2Uba7bt269oiICPU6upycHPvdd9+ttl9+/jZt2thffvllu81mqzL2ciXvE5PJVO77ynvl/vvvt5900knqfSDvF7n99ttvl3qevK68vu7HH3+0Dxs2zF6vXj17dHS0vWnTpvabbrpJxZBEoSRC/gl0YoyIKNzIyKP0FpCRxUcffTTQm0NEREQGIxU/EivIqsX+rD6impHeWlKpLhVkROQ+9pQiIgoA6TkgPQYqaoRJREREJH2IZPDKtbUAGZO0TJApjTKNj4g8w0opIiI/kn5Rsgy0VEdJkPnzzz9z/xMREREFoTVr1mDp0qWqN5RUwW/btg2xsbGB3iyioMJKKSIiP5Llk++55x7VpFKaVRIRERFRcJIFZaTJuqxIKA3hmZAi8hwrpYiIiIiIiIiIyO9YKUVERERERERERH7HpBQREREREREREfldpP+/pfHYbDbs3bsXSUlJiIiICPTmEBERkZ/Z7Xbk5OSgUaNGMJk4ZneiGFsRERGFN7ubsRWTUoBKSKWnp/vz+BAREZEB7dq1C02aNAn0ZgQ9xlZERETkTmzFpBSgKqT0nZWcnFzpzqKakxUppk2bhmHDhiEqKoq7koifJSJDnZeys7PVAJUeE1Bwx1aMO4ITj1vw4rELXjx2wakkCP6+dje2YlJKliB0TNmToIlJKd99aOLj49X+NeqHhigY8LNE5NvPEqfxh0Zsxd+VwYnHLXjx2AUvHrvgVBJEf19XF1uxaQIREREREREREfkdk1JEREREREREROR3TEoREREREREREZHfsaeUB0sbFxcX+/ZohPic18jISBQWFsJqtQZ6c4gqJXOyzWYz9xARURDHVow7ghOPm/cwniGiYMGklBskYMrIyFDBE9WM3W5HgwYN1Co8bCJLRlerVi31fuV7lYgoOGMrxh3BicfNuxjPEFEwYFLKjZPjvn37VOWELGdoMnHGY01I0Jmbm4vExETuQzL05z0/Px8HDx5UXzds2DDQm0REFHL8EVsx7ghOPG7ewXiGiIIJk1LVsFgs6o/URo0aqSUX6cRK9GNjY5mUIkOLi4tT15KYqlevHqfyEREFYWzFuCM48bh5D+MZIgoWLPupht7/KDo62h/Hg4gMQP8jSXpbEBGRdzG2IvIPxjNEFAyYlHITe8sQhQ9+3omI+LuWKNgxniGiYMCkFBERERERERER+R2TUiHsmmuuwahRowK9GSGlefPmmDBhQqA3I6RJ77HWrVtj/vz5Pnn9d999F+eee65PXpuIiMjoPv30U7UqG/nHwIED8fXXX3v1Nfv27YuffvrJq69JRBQoTEqFsNdff10FHu44cuQIzjzzTNV0NCYmRq2Gc/vttyM7Oxv+tnbtWlxyySWoW7eu2pa2bdviscceU01RXd10001o1aqVauQozx05ciQ2bNiAcFRQUIDHH39c7SvZZ3Xq1MHFF1+s9qWrDz74AKeeeipq166tLkOHDsXixYsNs3160qhFixbo37+/875nn31WfS29ESoKpLdv365K1PVLUlISOnXqhNtuuw2bN28u9dwxY8Zg2bJlmDNnjo9+WiIiCtXBPjnHvPDCC6XunzRpUrlpUtI367XXXkOXLl3UIi9yzj3rrLMwb968Us/7+eefccYZZ6g4Jjk5Gf369cNff/2FcB34kxUZlyxZUur+u+66C4MHDy5139GjR9X9zZo1U31fJX6V8/vOnTt9vp3uHlvx22+/4cCBA7j00kvdjl/LxjRpaWkYNmwYli9f7nzO+PHj8dBDD6nG8EREwY5JqRCWkpLi9kiYLMcsJ0U5eW7atEkls/7++2/cfPPN8KeFCxeiT58+qlrmzz//VNsiCQnZHgna5H5djx498Mknn2D9+vUqgJPlb+WkrTdQdTcAmjlzJoJZUVGRSi59/PHHeOaZZ9Q+mzx5slrdSPal7FOd/KyXXXYZ/v33XyxYsEAlH2Wf7dmzx+3v98QTT6jA3BfbJ8fwzTffxHXXXVfqNeS4SxLrlltuqfJ7yXtWlhlfuXIlnnvuOfXeOOmkkzBjxgzncyR4vfzyyzFx4kS3fwYiIiIhSYgXX3wRx44dq3SHyLlMkhBPPfUU7rzzTnUukvOvnHMluSJJLN3s2bNVfCPnxaVLl2LIkCGqmtc1AVEdiZHKJm2Cef9KnFEVSUhJpZCc82Uga8uWLfj222/Vda9evbBt2zaf7TtPjq2QWOPaa68ttfK0u/GrHtPIc3Jzc1XiKzMzUz0mt3NycjBlyhS3t52IyLDsZM/KyrLLrpDrsgoKCuzr1q1T18Hm6quvto8cOVLdLiwstI8dO9Zet25de0xMjP2UU06xL168uMr///rrr9ubNGni1vfasmWL/bzzzrPXq1fPnpCQYO/Zs6d9+vTpzsetVqv92LFj6royNpvN3rFjR/V/yz5vxYoV9oiICPsLL7xQ6f9fuXKlOo6yLe5q1qyZ/d9///Xo+c8++6z92muvtScmJtrT09Pt7733nvPxfv362R944IFS/+fgwYP2yMhI+6xZs+yeWr16dbXPkX0i+0b2kSvZh7IvZZ/Kvq2IxWKxJyUl2T/77DO3t+nxxx9X7y13ebJ9S5YssZtMJnt2dnaFr/XJJ5/YU1JSyt2fkZGhjv3y5cvLfY/Bgwer4yY/q06ORXR0tD0/P7/C72Pkz31xcbF90qRJ6pqIvPdZqioWoOCLrdyJOzwl575zzjnH3r59e/v999/vvP+XX35RP6vu22+/VV//9ttv5V7jggsusKelpdlzc3Mr/T5yXnzyySfd3i45Nw4aNMij58u5dOrUqepnkbht+PDh9r1796rH//rrLxUryv5zdccdd9iHDBli99SBAwfUpTpyrpZYVc7Pv//+u/P+O++8s9TPd/PNN6tt3rdvX6n/L+f0xo0b288880yf7TtPjq3EfxL/rFmzpsrXLBu/VhTTzJs3T90nx0wnsegVV1xR5Wv7M55hfBK8eOyCU3EQ/E3gbmzFSinPk3jIL7YE5CLfu6YeeOABNff8s88+U1OXpGfP8OHD1WhTRfbu3atKygcNGuTW68sIzogRI1RFiozuyVRAGelzLaOWcveWLVtW+horVqzAunXrcM8995QaURJS7SLVNt98802F/zcvL0+NOsm0Lxmt8qVXX30VPXv2VD/nrbfeqqp3Nm7cqB4bPXq0Gq1zPVbfffedKiuXaXOekNExGQl85513qnye9CmQUVbZR65kH959991qn0rlUEVkSmRJSQlSU1PhK55sn0ypkyl+Mv3OG+R7yEjmjh071Ai0To6fVGotWrTIK9+HiIiMF1sVFFu9HlvJ9DKpxH3jjTewe/fuSs97ci6rqH/hvffeq1omTJ8+vcL/K9OxpALGl+dl/fz/yiuv4IsvvlDVWhKv3Xfffeqx008/XVXau/YskioeiWckzvGUxH/ymocPH672uRLHSWXRI488UuHUNLlP4izZjgYNGpR6TKbDSVwmlUWVxbcnypNjO3fuXNV2oEOHDpW+nrvxq/xswnXGQO/evdmKgIhCQmQgv7mcBF9++WX1x6L8Af7LL784G3PLH8oyX1rKmaUMV6aiSVJCTmzyB75OTjpjx47F77//rv4AvfDCC1UvpcTERJ9sc0GJFR0fC8xc/3VPDUd8tOeHTE54ktiQEmUp99V7C8lJ86OPPsL999/vfK5M7fr1119VDyA54X744YdufQ9JOLgmHZ5++ml1PGU6oPSmEhJgyRz6ysi0LlHZyVvulxO8q7ffflsl3ORnbNeunfqZZHqWL0nyTYIe8eCDD6q+AjIdTr6/9MKSHgeynXoSSgIY2a+eLsvbsGFDVQYu0yolCL7xxhsr3W9S7l8RfV/Kc04++eRyj8v2y+dJPlu+4sn2SfLI9fPtDe3bt3f2aJAATkiQKL9T5PsREYUSxla+j63OP/98dc6SXokSR5Ul57SqYhn9ORWRRJEM9Ek84UsSZ8vUNz0uk1hNpqQJiTlkiprEL/p0ehl0lKljEmd7SmJ3+XklMfXPP/+oHklVkeRY9+7d8dVXX+HKK68s9dihQ4fUdlS1fyXRKFP59HO+N3lybCXGqF+/frmBVk/jV/l5Ja6Wv21cfyaJl3bt2qUSdRV9DyKiYBHQ32Dyi1gSGW+99VaFIzhS0fPoo4+qa6nakWqU8847r9TzZKREmiXLL/M//vhDBWOV/fEerrZu3aqCj1NOOcV5X1RUlDqxyXx2V5Jgkf0tiSn5f1K15A4JoCSIkBOyjK7JiVNe27VSSo5LZSODrjwZtZTjLxVLs2bNUiNXEsQVFhZW+nzpkSXbpl9k+yRR53pfdbp27eq8LYkmGak7ePCg+loaVkpfAAmkREZGhurdVNXIomszy7IXqTCSz4Jsd1XNO2tSRSdBoow2SvJQejhURqqXXPePjBDLz+d6n/7znuj2STK0qm2pCf17l00Kyqhj2eb5RETBjrGVf0hfKak+LxtHnch5WZJATz75JL7//nvUq1ev0udJPOB6DpYYoaJzdVVkcMZ1oFAGwvRYRkjcIr2SpHJeyHn+7LPPrrRXqfSBqiyWkQVOpE/oqlWrVCKmOrIYilQdySI3rpVBrmo6e8Ab+84bMY078ass8CLbI43UpaJcKtUkyeUax0hCSnp3EhEFs4BWSkkyQK/cKUuqGMomMKQBsiRS5ITStGlTFQhMnTpVrdIh03GElFNLJYuMNHm74kLERZnVqFogyPf2NUmwyEWqS6SySap9JDEowUpVJCElx0v2u0wNlBPlRRddVGkwURE5KQs5rt26dSv3uNyvP8f1fSKXNm3aqKlucuKWJItUJlVERgH18nQhTSklsJSG2+6ShJ4rCbhcS8wl0LjjjjvUe1ECTFmdRS6VqSyg1R+T15MpgvKer4jsk8peQ7+/7H6T4yRJKWmi6Zpkq4h8tmRqpWvTTmmMLvtN5xokncj2SSC6evVqeJP+PaQ03pVUWUoSkYgolDC2cpkGl52DpOSkKqtIahpbDRw4ULVBGDduXLnFP2pyXpZBouuvvx4//PBDtdXLEt+6npdl4Fam2rkOEFU3/a+iWMY12SINwyVpJdslMYjEVlWt6CyVVq4rzJUlsZcMekqFtjtker9U+UtFkSs5b0tirKr9Kz+LxKK+2HeexjSVNcR3J36VJFTHjh1VZVlFyUCJYxISEpxT+4iIglVAk1KeysrKUica/RezVKDIbT0hJeRELsGH9IqR8mpvk+9fkyl0gSRBhZQEy1K1snSukMopSebJVLPK6IkWd0Zg5LUlKNP3uVROyXQpT0gpvCTDpFpLAhvXIFJGiCSB8vzzz1f6/yWYkktV2ysjj66jj5GRkWjcuHGlwUtNyHQ7qQqThKkkpa666iq3ppeVtX//fjVNUF5L+lhVRvaV9F6QfeQ6hVKOn+xLCWhc73/ppZfUiobSc8H1s1MZCXZc948Ea9nZ2W7vM0+2T5KREoTKcfR0umNF5HtIEk0SUq6JTqkClBHJipKfVNUOtQLrftVudzgXMJf+o4bIl3YdzcfK3ZmolxSL3i18228nnIRqbCW//y3RZvW6vpraJIM7ErvI9Kuy5z1Z5VVaS5TtPSTnc0kySCW0TvpljhkzRiWApBqpOhK7uJ6DJa4pe672BhkUk2RNkyZN1D6satskASOXisg0x//++09VXpVNxlVGKoRkUFQqsFxnSch2SFWRbJcMNLr2lZLKJEliSbKwssTSie47T46txBgSy0liSpJOnsav0mOqqrYXa9asMWwcU7ByJUr27kVMu/aIadkCc3bPQZ4lD70b9EZqbAXHpjAb2PI3YI4GOpwTiE0mCktbDuZgw/4cpNeOx0npFVfC+kPQZFfkD0gZXZERhOTkZHWf/KIvW94sJxs5EcljlZFf+q6/+OUPbD1RIxdX8rWcKCS4qajhopHpJzk52Up5svSOkkBTKm6kl5dMXZJmkvJzSe+uAwcOqJExCQRkSqTsb5nyJ8+v7meXk7mMNknAIsGllFzL/9H3nVy///77KlEjyaXKSK8rCSYuuOACPPTQQyrYkCBYtr1fv36qAkleT/qMSXm7nPhl1EyajUrljvys0mTdk2Pl6bHVf6bK7pNtkMSUBFMyavZ///d/NXrvyHtbEirSv6Gq/y+NvGW6pQRHclyl6kuOpSTw5PtPmzbN+V6QhJQEh19++aU6rnpZvrtTF/WftaJ94I3tk8b6ktCUaqnOnTs7X0OqI2VEUPozSLNVGW3V33ey3fq2SK8J+ZnkvS3Bmuy/xYsXq+DRtaJNyuWl6b4kqyprpCrbI59/6a1hJPrvqLK/q/zCUoioH6/Vvv99GUCMdxrSE7ljzqaDeHjSWgxpVwfdmnT3+mcpIJ+pAAvl2Eqv+vHkfOXOa7q+XqdOnVSCQs41Qr9fkiYSo1x99dUqNpFeSrI/JGEivTalAkafeiWDVxKLTZgwQcVg+nlZHpdKGnfo39fdn7Oi51d0n7wvJCkkA1kSi0h1VU325cUXX6wukpCq7v+7HjepHJPBK9lHEjvo//eZZ55RPa4kBpTEoMQL0i5BYk95b0ml+onsi6p4cmxlwE0SdTI98JxztESLO/Gr6zZVtV3yuvI6VT3Hn/GM6+/So199hZzffkfavfegdvo1eHHxi9iRswMfDv0Q3etV8Pv72C4VX9jjasPSOjAzUsJZOJ8Hw/24TV29D69M34yLujdGxwad4G3ubltQJKXkh5GTgPxSrW4lMnfIH8MyZ78s+eNY5tiXDcQkMSJ/KHsyFc0o+01WGJOTpZSXS/ApVTvys8jI3o8//qhOUPK47Nv33ntP9ZCSn1Oqh+QEKuXTemBZFdmfUro9YMAAFbhKIkJGhuS19P8vSQVpPFnV60lgIcdBkicyDVO2VUboJLEj26IHvfKzSXNxCeKkAaSc2GXuvSS9ZP6+O9usn6wlgeHJ82U/uj5fkiSyTa73ScN+CaJkmyQR6O7rlyXBhjv/VxKC//vf//Dwww+rppeSqJGpl7IvpRJJfw0JmuSYlG2gKn+USBLQHfr+9+Rncnf7JOCV952sRCPJM538P9eVF3v06KGuJdkk7zl5nwjp5yXkcywjjPKYTFWUBJTr9soI6xVXXFHpzyD7SEZcpUedfIaMyJ3+bN5mshVDHxeWY2cxc8oA+c/qg1I9aVZ9b2QgxdufpXDrMRcusZWsZOeLuEong2aSZBCu98sgm+xXOfdJfCR9laQFhZy3ZLqW/lxpNi6vKc/RF4bRE0Jlp65VRuKSsttV3fPluLs+X855ZX8GSU7K+VYWJJJEUE1jGYnjyr52RVz7I+nHTWKTG264odTPJ7GCVHtLrHjTTTep3wlSiSQVfdKnVhJBnuwLT/adJ8dWP47Se0ymewp34lc9ppH+cJVtlyQv58+fr37eqrY9EPGM/E6tv3sPJKW6Yf0GHJs8Wf0sYuHChdgfWT65nVi4B6fL/ikuxhQv/n4n48eWdOJO5Lht2KPFVvL32eTJ3l8Ayt3YKsJe006BXiZVDK6r75UNmmRkoeyKHR9//LFqhOg6X1t+4covdZmTX1mJeUWjefIHrCxVq48Uup6s5CA1b97c6w2YfU1G7yTpJMv9Bpq8zSTASEpK8sq0LApd0ghVquU2b97sk1U0pQpQAtcNGzZUOgotn3uZfiq/F4z2uZffiXLykYRl2Z4gfqmUelH744KVUuRvPyzd46yUev+K7l7/LEksIH/MynS2srFAsArn2IpxR3AKteMm1YXSV1SmL+otNLxBknXyGZUB5ar4M55x/Z169IknjldKSXuP38+vulLq8CZEvddfq5S6Z7NPt5MMFltSQI/be7MznJVSz5/v/Uopd2MrQ1dK6UGT/HEqowpll5CV6VwyyiAjOHrlhARXMtJSVeNqGdGQS1lyMMseUKmCkZOizGEPluVWJXiU5WhlNEJGkIyw3Xppsb4viSojVXxSyi5T9apqEF9TMnXw888/r7K/g7xH5b1a0e8EowjMtlmPf//ISNkIP39/CmeRjqknpgiTV9/7+mfJqJ91bwuX2IpxR3AKteMmjdU/+ugjNU2v7IIrJ0IWmJHkcXX7KBDxjHwf+T0tzCaz+lpPMEqVZIXbITGFHPcKmvCT/4TTuTCURJ3AcTOZtc+qyaT9nvA2d18zoEkpKU+V6Vw6mQ8uK2LI9C9Z7U1Wb5PeMX/88YcKYPReBvK4NO7u0KGDmn8tZb1S/iyBlr76hy9W3gsW0ktHSoGHDBmiekl5g/RNkCRBRWSURpphEnlL2ZWMvKm6VY2oCiaXfhRW9h0g/7LDEIXdhsfYishYylYqeoMkpIJJkVWrojRHVNLXqiBTuzaXT2wTke9YrVpsZQpwZWpAk1JSyiqJE530MxLSPFAaK0rDQL1ywpWM7A0ePNjZG0YSUdJoUEYDpBGj3nAyXMn+8nZvDOnfUVmjMhmtIaIwIKvtxaUCBUeB3P1AQukKCyJfKnEETpHm4J/S40uMrYjISIqtxTiQf0DdbpKktQAoZ8dc7bpJ9atCE5H3ZBVof9+nxEWFb1JKEktVtbRyp92VVE1JQ2nyLW/OgyeiIJbcSEtKZe8D6nt/7jlRZSxWbVpPpKPUnCrG2IqIjGR37m7Y7DbER8YjLbaSwayM2dp1i0F+3TaicHcsX0tK1YqPDuh2MLIjIiLPklIiR1u2nMhfLDZtoCrKxEopIqJgsTN7p7pumty04ub1liJg50LtdgttlUIi8o/MfG0F3Frxga2UYlKKiIjcl9RQu5ZKKaIAJKXMIdD8mIgo3JJS6UnpFT9h9xK1ui8S6gF12/l344jCXKZj+l5tJqWIiChoJDp6yOVq/SGI/MWqV0qxpxQRUdDYlbNLXTdNalrN1L2BsuyiH7eMiI45K6U4fY+IiIJFYj3tmkkp8rMSR08pM6fvEREFXVKq0kop16QUEflVlrOnFKfvERFRsEhqoF0zKUUBqpSKZFKKiCg0klLFecDu/7TbLU7185YRhTe73e6cvlcrjpVS5CPXXHMNRo0a5Zf9+8QTT+Dkk09GqPPnPg1nV155JZ577jmvvuall16KV1991auvGZYS6mrXeYcDvSUUpj2luPoekf9IY+pJkyZxl1ONyKp7e/O0hVGaJDUp/4SdCwBbCZCSDtRuwb1M5Ec5RRbngB8rpchnXn/9dXz66aduPffIkSM488wz0ahRI8TExCA9PR233347srOz/X6Ejh49irvuugvNmjVDdHS02qYxY8Zg506tUaLunXfeQdeuXZGcnKwu/fr1w5QpUxCuPvvsM/Tq1Qvx8fFISkrCoEGD8Mcff5R6zsyZMzFy5Eg0bNgQCQkJKpH41VdfGWb7xMqVKzF58mTccccdpZKe7du3V9tcu3ZtDB06FIsWLSoXOOuXlJQUnHLKKfjnn3+cj48fPx7PPvsssrKyfPyThrj4Otp1/pFAbwmFGT1w4vQ9MtrAlJxb5dyTmZlZ6utOnTrBarWWem6tWrXKxWbz58/HiBEj1PktNjYWXbp0wf/+979y//e8885D06ZN1XPkPC4DOHv3hudKqO7Giu4eM19o3ry5+h4LFzpWlnOQ7R48eHCNfp7nn39exVISR9WrV0/9bBs3boRR5ZbkwmKzINIUifrxjp6UrjLmaNfsJ0Xkd5l5WpVUXJQZsVFmBBKXsAlh8oe5BD/uMJlMKlnx22+/YdOmTSpg+vvvv3HzzTfDn+Sk3LdvX/W93333XWzZsgXffvutupaT8LZt25zPbdKkCV544QUsXboU//33H0477TT1M6xdu9bt7yfBiiQ8gt19992Hm266Cf/3f/+HVatWYfHixRgwYIDaH2+++WapwFcSeT/99JN63rXXXourrrqqwuRQZSSQk0DLF9sn3njjDVx88cVITEx03te2bVv1vNWrV2Pu3Lnq+w8bNgyHDh0q9X8/+eQT7Nu3D/PmzUOdOnVwzjnnON8znTt3RqtWrfDll196tO1URkKadl2UrS3jTOQnTEpRsJHzz+eff17lc3755Rc1SCMxzb///osNGzbgzjvvxDPPPKMqfGV6hW7IkCH4/vvvVRJCzuNbt27FRRdd5NE2SZJk+/btCGaexIreVJN9JwnEBx980Gs/z6xZs3DbbbepRNf06dNRUlKi4qG8vDwYUVaRNhDYMKEhzKYK/uhlPymigMksKDZEPynBpFQIcx0dKioqUpUnMqoiJ0hJCCxZssT5XBmdu+WWW9CzZ081SnP66afj1ltvxZw5jhEMN7333nuqykqqYS655BKPq1IeeeQRNeonJ+azzjpLjQgOHDgQf/31F6KiotSJWHfuueeqkcU2bdqopIVUwUgio+yIlC+88sorapQyLS1NbZMEBeLhhx9Gnz59yj3/pJNOwlNPPeXx91m/fn25kdKy5OeVaWkvv/yySv60bt0aHTp0UPtDRt3uuece7Nq1y7l9Tz/9NPr3768SNBL4SoXczz//DF/xZPvkZ/3xxx/VsXV1+eWXq+qoli1bqpFnGUGWKj5JcLmSJGyDBg1UAkoq6QoKClTQppPXlUCPTkBsLcAUqd1mtRQFIinF1ZkoSIwdOxaPP/64isEqIomEG264QVVAvf/++6p6WQZdrr/+elVdLOdDSULp7r77bpW8kDhNzuMPPfSQOsfqMYivHD58GOeff76K7STmkgFMYbPZVDJNzreuli9frgY7d+zY4fH3WrNmjVdjxUC78cYb1TGSCnBv/DxTp05V8b3EQhJbyiCyVFPJAK2xaL+vs4q1vwMaJzYu/5TCLGDfCu12c/aTIvK3Y44m5ylxTEoFHxmxkqZ8gbi4jJZ56oEHHlCjahLkLFu2TCUGhg8frkZnKiInR0lUyOidu2RER4Kn33//XZ00JSiRxJZOKlzMZnOlo0wS3EjCYPTo0Sqx4CouLk69lpygK9pmSWbI/5UAT6bx+ZKMZMropFzL/pSAQC/Fl22XKiB5XCeVW5I8kcSKJ2R09IorrlCVTLJvKvPNN9+oZJxUIpV17733qmBVjn1lJHGYmpoKX/Fk+2Q/yfZIcrQyxcXFKniXSkAJyCoj7xn9+brevXur41PZHwjkBkkIxNXWbhcc4y4jv7E4fg9y+l4I8lVsVZLv09iqOjLwYrFYVAVwRaZNm6baJ8iATVkyiCIDbnIOrYjEQjL9XpJTkrjwpSeffFINNMo5WgYDJdaR7y+Jp8suuwxff/11qefLdskUekmeeULiRhnYqyqBcyKxYiC0aNFCzToYN25chbHcif48+uCvL+O4E5EtVdWVJaV2zAfsNiC1FZBSweNE5FOZ+drfSLXjA9vkXDiGu8ltEuA81ygwO+zhvUB0gsf/TRI1MooliRMZgREffPCBqiD56KOPcP/99zufK8HFr7/+qipMJCD68MMP3f4+hYWFqky9cWPtxCJB2Nlnn62qZKRCS06u7dq1qzR4kqlYMrdfqmgqIvdLokaSX5JcEDKdS5JQ8r0l8SFl8B07doQvSVWZTCWTBJv0OZKfccaMGWq0Ux+5kgDt0UcfdQZnEmRJIrAmzUWl74CMisnxkwCwLJluKVVP0oOgLOlJIP225DkVkSSiVMxJhZuveLJ9Mqoq+1XeL2XJFEOZypCfn6+q1OT9K1P0KiLPkR5S8lquiVX5fpKk2r9/v8fBMrmQpFTeISalyK+sjr/nmJQKQT6IreRsWcsHsZWci1ynl4vKKpqlskgqpaRKWWIEGUxxpZ/7Kot7JMYoe/6WqWASg8h5TqqmPJl+X1MSg0h8KGQRkokTJ6oBHqm0lmSKxHlSrSMVPnqSRc7BnurWrZuavnbdddepnkllq6ZrGit6csx8QfaFtBeQeFD6gJ3oz6OTfS2JT0kASoW4EWUXa0mpRomNqpi6xyopokA4ludISiWwUor8QKp2pBpFTlo6SQzJyU2mh7l67bXXVCWVJKbk/8nUKndJMKInpIQki+SEqTdg7NGjB9atW1fqORVx7Z9QHUlyrVixQjW9lumHV199tfoelZGAQAIT/SJfS4Dlel91UxYl8STJDp0kSA4ePOj8WgI0fdRQfhYZ5ZT7KiNJJ9cm3a4X2afSS+CLL76osiG5J/tMJ5Ve0lNKEpTyM1XFdf9IYlOCT9f7qus95u72STJUGu3Lz16W9NKQYy19sSQQllFb1/0uJGiW7ZFgVqqvJOkqPbTKVk9JME8nQK+UyjfGSDSFBxsbnZMB6Oci10tVA3iSYJGp/i+++KJXzuEykCgVRVJlJbGIVFNX9f/lnO16vhZyzte/ru78L1zPo7LgiAwm6edfmXIoiRM97pGeR/KY9IasiFTLVxbzyEUSYDLQKP+/qmmJnuwzT4+ZN/edqFu3rqqGe+yxx0pVb9f059HJtD6Z7mjktgR6Ukp6SlXZ5JyIAjZ9rxYrpYJQVLw2qhao7+1jUjosFxmdk1LgU089VVX8SOLF1+SkLT2ByibKdHK/BCyuFUdSfaN/LUkvqfqRVQcrq/yRvg2uPZ9kxFGSZK4rvVWXNCtb6SXb5FqSLYkReV1J7kmSRfolSYPvykh1WWVJkpycHFUdJNt0wQUXVPgcKe+XqZES6JStRpJpmNJ7SZ7jSoJGGYGUJKQEtNWRAE4nCUD5+aThuU4C1Mp4sn1S+ST7oqLnSiAsx1ouMjosfS0k6SQl8Tr5eaT3lIxGy/upLL38vaLHyAPRjhHnohzuNvKbEkepVKSpfNKagpwPYis5L2fn5CA5KanCKuNS39sD+rnI1e7duyt9fmRkpOqhKMkWWdXYlX7uk/hGpuGVJfeXrf6W86Rc5P9KMkj6eErPospaF0jyRWIRnZw7ZXqcHuu4M/WvurhHH4yTHldyLQNHkoiriHzfyuI8IbGF7CdJ4lW0bTWJFT09Zt7cdzoZ5H377bfV5UR/HiH7SCrAZs+erfp6GZU+fa9cUirvCHBgtXab/aSIAuKYY/peKpNSQUgqOGowhS6Q9KlTsiKZPmVJRp8kgSNlv5XRAw53++9I9YwkGWSKlJAgSQJBqWZyhzxXql+kIkiagrvOrZegQE7k0gerqnnzss1Vba9U0MjF9Wt5PU+n1lVFggOZMiY/h2z3GWecUeF0NJ1UQ1U2aiYrCkpCUIIgCaoqIkkrKaWXRJw0VS3bkF2CpgsvvLBUwCer0knAJw043eG6fySQkyDb3X3myfbJiKuQajf9tifHWt4zVW2XjCjK8als2h+5SXpAiIpW0iHykZwii7pOjGXngZDji9hKYpgoq/a6VSWl/ECqfmSxD+nN5EpWTZMYRKa/lU1KSTPxzZs3q8VJTiROq2igTWJBT1fRrYr0zJQpatJsW5qzyxS8ysg5XwY+KztHSyWY/Mxl4wVvxoru8ua+k8oqGeSVFZ9lgLSmP4/EhrJvpF2FxHPSs8rIckq0wauGiWWSUjvmatd12wOJlcfIROQ7+7MK1XX95BgEGlffCwOSzJCpbXKilwbk8ge/9DaQihQpKxeS9JD57hIQSGn1n3/+qaZkyZQ/d0++sqqfTJ9buXKlmgIn1UdyotVPsBKsyIjfnj17Kn0NmUonz5dEzpQpU1SVkYwCyQlZEmlvvfWW87lSISOPyfZKbyn5Wk7QVU2V8xfZBimn/uGHH2q8PTIyJqOOsh/K9kJwJaOjsoqeHF8JbGXapSwpLQGiVI3JfTKSqk/Zkx5YcmwkESS9leTiy4agnmyfjBh2795dVVa59kSTfhyS5JSeU/I+GjNmjHofVTY9oDLyvpQ/AugE2R29OPRV+Ij8IKdQKzNPig187wMiT73wwgv4+OOP1TnNNT6TARtpmSCDRNJIXGIaqQKWyqqLLrpIxVF6lbL0kpLKZTkX/vPPP6oyWwYefb3AS3UkTpSkmsSU0qvJNeniCUlWSfW4xKxV8SRWNBI5xlLJXbYxvCc/j0zZ+/LLL9VryMCqHse5VnQZic1uQwQiUC++XsVT91glRRQwB7L1pFQsAo1JqTAKhiQJIQ0W5Y9+aZgoq3lI02691470FhowYIAqB5dlhyWo8KSBplSoyBQzWZlF/vCXHgSuZcpywpT+UlX1CJByb0k+yPx/Wa1Ngi0JyORaKrtatmzpfK70LJCpZ1KJdfrpp6vH5WeSk3qgSSApK+pI4m/UqFE1fh0JSFwruyozYcIEta+lf5U0u5TV6ySgkUbprqONslqgbNPzzz+vKrD0S2VTA73F3e0TshS2a/8s6ZkhSSx5/8p0BZl2KPtWEkzu9nMQ0qNCvp8kZOkE6dM2IngKIf/JdVRKJbFSioKQVD7LRVbjKxsvyICRVJtLywSJaWQq+iOPPKIGt/Qei9I0XVZFlnhHniMJIImzZDq+9GIMNBmAk0HJ888/39m/0VNShT1y5Mhqn+dJrGgkUiUmVWASj9T055GFi2TFPelH6hrHfffddzCqOnF1EGUqM5iwnf2kiAJtvyMp1SAl8EmpCHtNOuuFGOlpIyMX8ku+bG8cOXFkZGSo8lipBAomMoImf9DLiEqgqd4O2dlq/1bZ24HCniQvJeCWAMubo78SyEm5uzSHrY6RP/eS1JXKRkn++noZ8Eq9PxjYuxy47FugnbaiJ5GvnfLCP9iTWYCfb+2P7k0dzfa9+FmqKhag4IutGHcEJx437/JnPOP6O/XQ+PHI+vU3fHGaCRkjuuKbc745/sTcg8ArbbTbD2QA8Sc+1ZJCILYkvx63IosV7R+dCskELX7kdNRL8s3vB3djK2YHQpCMwskUvQULFnhUSUJkBDLCKuX7hw8f9urryi/rN954w6uvGbbyj2jX8ezNRf4h42eHc7W+OXUTA18VQkRE1ZPKh3JT9/QqqfpdmJAiCpDdxwpUQio+2myIuIoNQUKQ9IWSuf1SBix9obxBklvSw6Ai0g/BCH2cKHRIWbq3ybRA8pJ8Rw8yjm6SH6fuFVm0aaN1DBA8ERGReyrtJ9XiVO5CogDZeVRb+b1parxzmnggMSkVgmTVMukb5E1SGlhZL6j69et79XsRkYGVFALFudptJqXITw7nassWJ0SbERfNVR+JiIJFpZVSbHJOFDA7j2i5gvTUeEMcBSalyC2yBC4REfId0yqlaWlsLe4Q8osjjql7dZJYJUVEFEzqxtc9/kX2PuDIFmlrDDQL7KqRROFshyMp1TzNGEkp9pQiIiL3SYNSkVAXMEC5L4UHvZ9UWkJ0oDeFiIg8XH3Paftc7bphVyDuxBesIKKa2Xk0T103TUuAETAp5SYuUkgUPvh5r0LeIe06gU3Oyf/T99LYT4qIKKikxaYd/4JT94gMVSnVzCDT95iUqobZrPWuKC7WAmIiCn16TzYui1tFpVRimR4RRH6olKqTyEopIiKjs8myXg5pcUxKERmJzWbHrmOOpJRBpu+xp1R1OygyEvHx8Th06JD6A9VkYh6vJmw2m0rsFRYWch+SoSukJCF18OBB1KpVy5mUJhe5+7XrxAbcLeQ3RxyVUlx5j4jI+EpsxxdHqh3rmKaXtQc4ug2IMLGfFFEAHcwpQmGJDWZTBBrVijPEsWBSqhqyRGLDhg2RkZGBHTt2+OeohOgf+wUFBYiLizPEspNEVZGEVIMGTLpUKOeAdp3E/UP+cySPPaWIiIJFkUX7nR0bGYsoWRhF7JinXTc8CYhNCeDWEYW3HUe0flKNa8UhymyMghsmpdwQHR2NNm3acArfCSgpKcHs2bMxcOBATokiQ5OKSFZIVSFnn3bNpBT5EXtKEfnXNddcg8zMTEyaNIm7njxWbCuGpKLiI12qMDJma9fNT+UeJTJCP6k0Y0zdE0xKuUmm7cXGxvr2aIQw+SPfYrGofcg+PURBLNdRKZVYP9BbQmHkaJ6j0TlX3yMDJmpmzpyJIUOG4NixY6rSVr5+7bXXsHjxYmRnZ6uBzfvvvx+jR48u9f+OHj2Kp556Cr/88gv27duHOnXq4Mwzz8QTTzyBpk2bOp/3/PPP4+eff8aGDRtUxXn//v3x4osvol27dghHn332Gd58802sXbtWxZfdu3dX+/ecc86p9Ji4at68Oe666y518RV3j62776lgU2wpciSl4suvvMekFFFA7dBX3jNIk3NhjHotIiIKDpy+RwFwzJGUSmWjcwoC8+fPR9euXfHTTz9h1apVuPbaa3HVVVfhjz/+KJW06Nu3L/7++2+8++672LJlC7799lt13atXL2zbts353FmzZuG2227DwoULMX36dFV9PmzYMOTlaX9YuEOSH5IQCXb33XcfbrrpJvzf//2f2reS+BswYABGjhypElW+4Om+8+TYhnKllD59T8naDRzLACLMQNO+gd04ojC33VEp1aJOAoyClVJEROQeWU3H2eiclVLkv1VijuU7klLxXH2PjO/hhx8u9fWdd96JadOmqWonvZrnkUcewd69e1WiQu9hKBU0f/31l6qskiTUlClT1P1Tp04t9Xqffvop6tWrh6VLl6q2CL70yiuv4NVXX1UtLC699FJMmDBBVbzLzzhjxgwsWrSo1PNPOukkXHjhhXjsscc8+j7r169H27Ztq+w7Kkk52ZaJEydi7NixzvufffZZtZDOPffco5JT6enpCCRPjm2oKrFqjc5j9el7epVUo5OB2OQAbhkR7XQkpVgpRUREwafgGGDVkgNMSpG/ZBWUwOZYXbwWk1Khu/JpSb7XLwWWgmqfI9/bH7KyspCamupckVgqZ2Q6X9lFNWR63q233qoSGFJxU9lrCf31fOXff//F1q1b1bVMmZNkmFyEbLtUKcnjOplOJ9VLl19+uUffR47BFVdcoarJZN9U5ptvvkFiYqKqlCrr3nvvVRVkUp0WSCd6bEOtUipOr5TaPke7bj4ggFtFRGK7o9F5c1ZKERFRUCalRHQiEMUee+Qf2YXaiHtclBnRkew6EIokedTn6z4B+d6LLl+E+Cj3+2rIFDxJjLiyWq1V/p/vv/8eS5YswXvvvae+PnTokOoj1KFDhwqfL/dLokYqbXr37l0u6SG9kE455RR07twZvlS7dm01JU76NrVv3x5nn322qo664YYb0KlTJ1UV9fXXX+PRRx9Vz//qq6/Qp08ftG7d2qPvI9VR0lNp8ODBaqrj66+/XuHzNm3ahFatWqkFiMpq1KgRkpOT1XNcNWnSpNxz8/O1KgFfqMmxrcl7yugsNou6jjE7YoXdS7XrdE7dIwp0TJVTqH0+m9R2WYggwDh9j4iI3FOojc4jhqX35D95RdofZwkxDFko8KT59DvvvFPqPpnCJpU+FZEqI0m0fPDBByqR46omVVoy9WvNmjWYO9cxHaoSkiByrSgqKipSyR+ZjqeTKWSnnlr5Smiyva6r0TZs2BCrV692fi3VQB9//LFKSsnPIpVMMoWuMpJ0kv5YVZF+S5Jwu/HGGyt83NN9NmfOHCQlJZXbDl/vO0+209P3VDAlpaLN0UBxHnB4o/ZAo26B3TCiMLcvs1Bd14qPQny0ceIq42wJEREZW1GOds1+EORH+cXaHzeJMcf/OKbQEhcZpyqWvEkqinJyclRCQlZQrup7eyIhIaFcJdDu3bsrfK4kYM4991y1Ep9MTdPVrVtXragmfZQqIvdLEqTs97n99ttVVc3s2bMrrABydd5556mqJd2DDz6Ixo0b44477nDeJ19XpexqybJNrtPrLrvsMvW6y5YtQ0FBAXbt2qUakFfm888/r7RKSY6V9KySbZJ9VhHpOSXJOOlvVbZaSno4yUqH8hxXLVq0KLd6XWRkpM/2XU2OrSfvqWBRYtMqXKNNUcD+1YDdBiQ2AJIbBnrTiMLa3qwCdd0wxThVUoJJKSIick9RtnbNSinyo7xirVIqzkAjeuRd8ke6J1Po3CHJE0ukRb1uVUkpX5k5c6Zqav7iiy+Wq/qR7bnkkktURc5TTz1VqveQJHfefvttDB8+3NkzSqpupLH3L7/8ol5XEi3VkWSca4WQ3JbX83RqXVUkMTZo0CD1c8h2n3HGGaoBe2Wk2XdF5Oc77bTTVCWWJN0q6yslSStpci7TIF0bnQupYpIkmjRZP1Ensu88PbahymqzHq+U2rfyeJNzIgqo/VlapVTDFGO14WBzBiIick+JNrqCaO/+8UhUlQJHpVR8NCulKDjIlD3pvySVNZIk2b9/v7q4Nrd+7rnnVMJCEjkyFUyqjKQCShIW0rD7rbfeKjVl78svv1T9myRBor+eJDkCTabwSWPvH374Qd2uaVLyoYceUvuhbG8lV/369VMrGd5///1qFT5psr5hwwaMHz9e9aGS+wK98p6nxzZUWe1aUipSKqUOO/p81au4zxYR+c/hnCJ1XTcxBkbCpBQREbmnxDHtwsPpLkQnorDE5mx0ThQMZKU6mab2/PPPq+of/XLBBRc4n5OWloaFCxeqfkLSv0gaeEuFjVxLU/SWLVs6nyv9hmTFPemF5Pp63333HQLtoosuwpEjR9TPO2rUqBq/jiRsyvZ+qsiECRNUtZH0r5JG7z179lQJH2mUXrZ6KlA8ObahyibT9VRSKhI44lihMbVVYDeKiHAkT1sZMzWx/IIRgcRaeCIick+JVvLLlffInwpKtBH3WCalKMA+/fTTCu+XZJFrY2t5XmXPdVWnTh01HU0uValJQ/Sy3Nme6p4vCaGypH9SYaHj3OAnY8aMUZeqlD0mrrZv3+7TfefJsXX3PRW0lVIRZuCoIymV5r2po0R0YkmptARjJaVYKUVERO6xOP7wYKUU+VGhMynFkIWIKKgqpRABZO7S7kwN/QoxIqPLzHdUSjEpRUREQcmqncgQaazRFQptRRbtj5uYSE7fIyIKpqSUWQ1m2QFzDJBYeRN8IvKPnEKtT2dSbOnVVQONw45ERORZUkpW0yHyk2JHUio6kiELEVFQJaWK87Q7UppIR/vAbhQRIbdIS0olxhirixMjPCIicg+TUhTApFQMk1JEREGVlDK5JqWIKOByHZVSTEoREVFwsmm9fWDiNCrynxKbvooTR9mJiIKqUqooR7sjuXFgN4iIlPxiLSkVH2OsWJ6VUkRE5JkgXhGIghdnfhARBU9CSpiK87UbSfUDt0FEVK5Pp9FWNGZSioiIPMwKMClFRERE5VlsluN/aBbnajcSG3BXEQWY3W53WTzGWGkgY20NEREZV4TjlMFKKfIn5kCJiIIzKVXkSEqxUooo4Iqtx6sYjbZ4jLG2hoiIDMxRKcWkFAXi3cf5e0REhme1WyuolOL0PaJAK3JUSQlWShERUXDi9D0iorBxzTXXYNSoUYHeDArq6XuOnlIJdQO3QURUajVjEW02Vm2SsbaGiIgMjJVSRBTe9u/fj7Fjx6Jly5aIiYlBeno6zj33XMyYMcP5nObNm2PChAnl/u8TTzyBk08+2fn1Bx98gFNPPRW1a9dWl6FDh2Lx4sXl/t/atWtxySWXoG7duup7tm3bFo899hjy8x1/8IeR7du3q6rJFStWlHts8ODBuOuuu3z2vT/99FPUqlWrwsdkmyZNmlTqvn///RcjRoxAWloa4uPj0bFjR9x7773Ys2eP8zlWqxWvvfYaunTpgtjYWPU+OOusszBv3jyEQqPzCEuhdiM+LXAbRERKiWP6XpQ5wnDV50xKERGRe1gpRURhTBIiPXr0wD///IOXX34Zq1evxtSpUzFkyBDcdtttHr/ezJkzcdlll6nkxYIFC1SCa9iwYaWSFgsXLkSfPn1QXFyMP//8E5s2bcKzzz6rEiRnnHGGut+TyidJjJFG9oXsE1947733VJKxQYMG+Omnn7Bu3Tq8++67yMrKwquvvupsOnzppZfiqaeewp133on169er94S8DyTBVjbJFYzT95QIMxBbcTKPiPxHr5QyWpWUiAz0BhARUbBgpRQRha9bb71VjS5LNVNCQoLz/k6dOmHMmDEev95XX31V6usPP/xQJTCk6uqqq65SSYvrrrsOHTp0wM8//wyTSftDolmzZqpaqlu3bqrK5sEHH4QvvfLKKyqRIgkwSaJIFVhUVBQefvhhta2LFi0q9fyTTjoJF154oarm8oQkZVq3bo1gt3v3btxxxx3qIsfHtYJu4MCByMzMVF9///33+PHHH/Hbb7+pajvd+++/jyNHjuD6669XiUfX91owsNiPT99T4lMBx3uXiAxQKRVpvM+j8baIiIiMyWClvkQUGiT5YsvP9/6loKDa58j3dsfRo0dVVZRURFWUJKhsWpcnZDpeSUkJUlNT1dcyRU0qbO655x5nQso18SOVON988w18Saq4tm7dqq4/++wzVaElFzF69GiVoJPHXacarlq1CpdffrlH30eOwxVXXIGrr74aNtvx6V/B6IcfflAJvAceeKDCx/X3ytdff62Si64JKZ1M85PE1PTp0xHM0/eUuNqB2hQiclFs0c53UayUIiKi4OfeH3FERG79RikowMbuPXyysw5U83i7ZUsRER9f7ets2bJFJU7at2/v1veV6qXx48eXuk8SFdJXqKr/06hRI5VsEjJVT0ilVEXk/rlz58KXpMfRm2++CbPZrH72s88+W1VH3XDDDapCTJJjklx59NFHndVfMt3Q04onvSeTTFuTirQvv/yyXCLOVf/+/cs9XlBQUKpnly/I9LvExMQqn7N582YkJyejYcOGVT5Pjm9Vx1Z/TvAnpbQkKxEZo1IqmkkpIiIKXpy+R/7HFCgZgbsVVbr777+/XL+iiRMnYvbs2RU+/4UXXsC3336regpJw+sT+d46SRDddNNNzq+LiopU8kem4+mmTJmimq1XRhJPkpDSSaJFemnppFrq448/Vkkp2U6p3JLKrspI0mnWrFlVbve2bdtUs2+pmqrMd999Vy6hI9tSlTlz5qjXdU0SyjbLFDrXXlBVvU5SUhKWLVtW7v42bdo4b8truttEuKbH1shYKUVkTBZHFWqk2XgzH9hTioiI3MNG5xRAxguhyFsi4uJUxZI3yRSw7JwcJCclVVlxI9/bHZJ0kETDhg0b3Hp+nTp1ylUL6dPyypIkkSSl/v77b3Tt2tV5v0zt0nstSf+osuR+/TkVOe+881TVkmslVuPGjVWvI518XRXpHeVK9oHr9Dpp1C6vK4kaqVTatWsX/u///q/S1/v8888rXTUwJydH9ayqX78+Lrjggiq3S5qBl92/cdUcy549e5ZatU+ShNJU/sUXX3TeJ9+7KvJeqq4KTI6JVFTt27evymopeZ4cw4ro91d1fIMnKcUm50RGUMzpe0REFDJCcGSXiAJHEh3uTKHziM0Gk8UCU3x8lUkpd0lCafjw4XjrrbdUUqdsXylpXl2TvlIvvfSSWk3vr7/+UkkTVzIVTabMSbNsSda4/hwrV65USaznn3++yqoeubh+LT+HN5uJN2nSBIMGDVJVWZKUksbc9erVq/T5TZs2rbRi6LTTTlNJHKkY80Vzb0lauf7ssi+ys7O93lz9oosuwkMPPaSOrWuj87LvFTmm0nvr999/L9dXShrLp6Wlqf0Z9Emp2JRAbQoRVVQpZTLeMB8bnRMRkZuMdxIjIvIXSUhZrVb07t1brZInvYOkokUqbvr16+fx60mFjkx7k+lvsjLb/v371SU3N9eZrPvoo49Us3NZzU6aiu/cuVM10pYkhnzPu+66C4Em090kkSTbVd0UusrIzyqJnD///LPank1GJ1Vckox6/fXX1eqJMl1xx44dmDdvnppO+fTTT6vnSVLq/PPPV9MU5Thv375dNYmX58iKfLIaY7CtvFdxUoqVUkRGYLEat9G58baIiIgMjpVSRBR+WrZsqaapDRkyRK2O1rlzZ1XJIo2/33nnHY9fT/6P9DWSyhqpENIvrj2fpKH3woULVV8n6YckVT3jxo1TiQxZmS0mJgaBJtsvK8XJtLxRo0bV+HWkEs21siuYSbP2adOmqemBkniSirfrr79eNUC/7777nIm477//Hg8//LBKYrVr107195IElvQWO5F9GUj2sjFCbHKgNoWIKmh0zp5SREQU/D2lOH2PiMKUJI1kNTq5VEYqXiryxBNPqEt1zyurS5cupZpx19Snn356ws+fMGFCuftkKlphYSH8QSrKKmsOLokcT7geC3dI4/qyzet1FW2TrKKor6RYmcjISJWk0hNVoaBcpVQMk1JERmC1OSqlvDCl3duMt0VERGRQnL5HRERElSuXoIsJjeo3omBX4khKmdlTioiIgh+n75Ef322szCMiCuJKqeDuEUYUKqx6o3Oz8QaZWSlFRESeTd8jIiIicgen7xEZqtF5JCuliIiIiIiIKCxEB98KgkShSO8pZWZPKSIiCl6slCIi7+HUTKIw+IxFc/oekRFYnEkpGE5AN2n27Nk499xz0ahRI7Us6qRJk8r9In3sscfUSidxcXFqBYvNmzeXes7Ro0cxevRotcSqrP5x3XXXITc3188/CRFRGDFCkEtEQRtbmc1mdV1cXMyjSORD+fn56joqKipw+5mVUkSGYLMbt9F5ZCC/eV5eHk466SSMGTMGF1xwQbnHX3rpJUycOBGfffYZWrRogUcffRTDhw/HunXrEBsbq54jQdO+ffswffp0lJSU4Nprr8WNN96Ir7/+OgA/ERFRCDPeOYzCCd9/IRNbRUZGIj4+HocOHVJ/LJt8MJXAZrOppFdhYaFPXp98g8fNOyT5LAmpgwcPqsSynggOCCaliAzVU8pswHNiQJNSZ511lrpU9st0woQJGD9+PEaOHKnu+/zzz1G/fn016nfppZdi/fr1mDp1KpYsWYKePXuq57zxxhsYMWIEXnnlFTVKSERE3sZKKSKjCobYSiq4pFIrIyMDO3bsgC/Iz1pQUKCqweT7UXDgcfMuSUg1aNAAgRMBRGrJbiIySKVUhPGORECTUlWRQGX//v2qrFyXkpKCPn36YMGCBSpwkmv5ZasHTUKeLyNiixYtwvnnn1/haxcVFamLLjs7W13LaKBcyPv0/cr9SxS8nyWTzQ6zYyTbyt+V5CfyflPXVptX3/dlP0vhcH4yUmwliaLmzZurx3zR98ZisWD+/Pno37+/qsyi4MDj5h3y+ZL3vVRIyT71B/1zbrG6fL/IKJT46ftTzYXTeTCcj1ux47Mo4zT+Otbufh/DnqUlaBIyeudKvtYfk+t69eqVelx+AaempjqfU5Hnn38eTz75ZLn7p02bpsrJyXdkKgARBednqc3+jegIYNeuXVgxebLfvz+Fp23bpczchG3btmHy5C0++yzpvVdCWTjGVtJji4IPj1vwWrxosfMPTIvNhMmMF4IG/04L7eO2bo+USJmxd/duTJ68E/7gbmxl2KSUL40bNw733HNPqdG89PR0DBs2TDX1JN9kSeUDc8YZZwS22SJRkAvkZ8k0bxOwD0hv0gSNRozw6/em8LVq6kb8u28HWrZsiRHD2/rss6RX9lBoxFaMO4ITj1vwH7vefXpj2ddvqfsiY+LU1F8yNn7uwuO4ZczcBuzcgmZN0zFiRCe/bKO7sZVhk1L6/OcDBw6ovgM6+frkk092Pkca+LmSElVZNaaq+dMxMTHqUpYcTCZMfIv7mCiIP0uONWRlGo+JyWXyE71Jtcls8sl7Xv8shcP5Pxxjq0B/f6oZHrfgFWl2+fPSzM9fMOHnLsSPW4TWTCoy0uy386K738d4rdcdZEUYCX5mzJhRKtMm/Qz69eunvpbrzMxMLF261Pmcf/75R/WfkP4IRERERMTYiogCwMyEMJFR2Gx6o3PjdToPaKVUbm4utmzZUqoB54oVK1TfgqZNm+Kuu+7CM888gzZt2jiXLZZVX0aNGqWe36FDB5x55pm44YYb8O6776oStttvv1016uTKe0REvsLV94iMirEVERmGiUkpIqOwOcJ3k/FyUoFNSv33338YMmSI82u9F8HVV1+NTz/9FA888ADy8vJw4403qoqoAQMGqGWKY2OPLy361VdfqUTU6aefrkr8L7zwQkycODEgPw8RUWgz4FmMiEphbEVEhmEybKcYorBjc6x0azJgViqgvykGDx5c5TLAspzpU089pS6Vkaqqr7/+2kdbSERERBQ8GFsRkWFw+h6RASulImA0hu0pRUREBsXZe+TPtxvfb0REwSnCHOgtICIHvRjIgIVSTEoREZGbDDiyQkRERAbFSikiw7AZePoeK6WIiIiIiIjIy39pslKKyChsnL5HREREREREYYONzokMw+rIShmwUIqVUkRE5Ck2+SEiIqLq/tLk6ntExuspFQGj4fQ9IiJyk/FOYhQ+Ivj+IyIKLmx0TmS4IeUIJqWIiIiIiIgo5LGnFJHxGp1HwHBYKUVERERERERe/kuTjc6JjNboPMKAledMShERkWccIy1EREREleL0PSLDhe8RxstJMSlFRERERERE3v5Lk/UPRMZrdA7D4W8KIiIiIiIi8i5WShEZsFIqAkbDpBQRERERERF5VwT/1CQyCrtz/T3j4W8KIiJyjwFHVoiIiMigmJQiMlyjc5MB43kmpYiIiIiIiMi7mJQiMgw7G50TERERERFR2GBSishw0/ciYDyslCIiIiIiIiLvYlKKyDjs2pUBZ+8xKUVEREREREReZsS/fonClN1xHWHAWilWShEREREREZF3sVKKyHAijJeTYlKKiIiIiIiIvM2Af/0ShSm73uncgFgpRUREHjLuSY2IiIgMgpVSRIZhh3ExKUVEREREREShP0+IKMxFGPBzyaQUEREREREReZnx/vglCld2A5dKMSlFRERERERE3mXAigwiMh4mpYiIiMiwDDywR0REREQniEkpIiIiIiIiIqIQFwHjYVKKiIiIiIiIvIvT94jIDUxKERERERER0Qmzl5p0bcSaDCIyGialiIiIiIiIiIjI75iUIiIiIiIiIiIiv2NSioiIiIiIiIiI/I5JKSIiIiIiIiIi8jsmpYiIiMjwuIgTEVGwYaNzIqoek1JEREREREREROR3TEoREREREREREZHfMSlFRERERERERER+x6QUERF5iD0iiIiIiOECEZ04JqWIiIiIiIiIiMjvmJQiIiIiIiIiL2NlNRFVj0kpIiIiIiIiIiLyOyaliIiIiIiIiIjI75iUIiIiIiIiIi/j9D0iqh6TUkRERGRYdnugt4CIiNxlB39pE5FnmJQiIiIiIiIi72KhFBG5gUkpIiIiIiIiIiLyOyaliIiIiIiIyMtYKkVE1WNSioiI3MPmPkREROQuxg1EhmOH8TApRUREnongyCcRERFVw1bCXURkECZH+G43YLKYSSkiInKT8U5iREREZFBWJqWIjMLkyEpZbcaL55mUIiIiIiIiIu9ipRSRYZgdMx0sTEoREVHQMmC5LxERERlHqalBrJQiMgyzo1LKxqQUEREFP/aUIiIiomrYLNxFREabvmc33iAzp+8RERERERGRd1mLuUeJDDZ9z8ZKKSIiCl6OkRUWShEREVF1rKyUIjLa9D0rK6WIiCj4MStFfNcREVE12OicyDBMjkopqw2Gw+l7RETkHgOOrBAREZFBsdE5kWGYHZkfmwHjeSaliIiIiIiIyLuYlCIyXqNzG5NSREQUtIx3EiMiIiKD4vQ9IsM1OrcyKUVEREHPcVIjIiIiqhQrpYgM1+jczul7REREREREFPJsXH2PyHCNzu3Gm/nAnlJEROQhVkoRERFRNVgpRWS4SikrV98jIqKgZbyBFQoDdr7xiIiC83c2e0oRGS4pZWNPKSIiCnrsKUVERETVsXL6HpFRmDh9j4iIgh9LpYiIiMhNrJQiMgyzo3ETK6U8ZLVa8eijj6JFixaIi4tDq1at8PTTT5fqGC+3H3vsMTRs2FA9Z+jQodi8ebNXDyAREbliTymiYMXYioj89wunhDubyCBMrJSqmRdffBHvvPMO3nzzTaxfv159/dJLL+GNN95wPke+njhxIt59910sWrQICQkJGD58OAoLC712AImISI0CcDcQBTnGVkTkN6yUIjJgo3M7jCYSBjZ//nyMHDkSZ599tvq6efPm+Oabb7B48WJnldSECRMwfvx49Tzx+eefo379+pg0aRIuvfTSgG4/EVFocZzE2FOKAoBvO+9gbEVEfsOeUkTGa3RuN15SyjGz0Jj69++PGTNmYNOmTerrlStXYu7cuTjrrLPU1xkZGdi/f7+asqdLSUlBnz59sGDBgoBtNxFRaOP0PaJgxdiKiPxaKWXAP4CJwnn6nsVqvM+koSulHnroIWRnZ6N9+/Ywm82qD8Kzzz6L0aNHq8clISWkMsqVfK0/VpGioiJ10cn3ECUlJepC3qfvV+5fouD9LJmsFphV2a8NNv6uJD+xWm3q2ma1efV9X/azFC7np3CJrcLtuIYKHrfgP3YWS+kV90oKc4HI2ABtFbmDn7swOW52LZ6yeDmeqoq738fQSanvv/8eX331Fb7++mt06tQJK1aswF133YVGjRrh6quvrvHrPv/883jyySfL3T9t2jTEx8ef4FZTVaZPn84dRBSkn6V2+zajPYCdO3di1eTJfv/+FJ62b5eibhO2bt2KySWbffZZys/PRzgIt9iKcUdw4nELXkuWLEGMy9fTJv8OS2RCALeI3MXPXWgft7UHpVLKjP0HDmCyn+J4d2MrQyel7r//fjWip/eG6tKlC3bs2KECHwmcGjRooO4/cOCAWn1PJ1+ffPLJlb7uuHHjcM8995QazUtPT8ewYcOQnJzs058pXEmWVD4wZ5xxBqKiogK9OURBK5CfJdOsVcB+oGmzZmhy5gi/fm8KX0v/3IDZ+3eiVetWGDG0jc8+S3plT6gLl9iKcUdw4nEL/mPXq1cvrPr6Xef9w4acCiRpv1fImPi5C4/jVrR8L77ZugZpdepixIgeftlGd2MrQyelJLNmMpVueyWl5jabVnrWokULFTxJ3yk9UJIfXFbhu+WWWyp93ZiYGHUpSw4mEya+xX1MFMSfJbPZ0SjRDDOTy+Qnehwg7ztfvOf1z1K4nP/DLbYK9PenmuFxC16RkaX/vIyKsMoBDdj2kPv4uQvt4xYdpX027Yjw23nR3e9j6KTUueeeq/ocNG3aVJWYL1++HP/73/8wZswY9XhERIQqOX/mmWfQpk0bFUg9+uijqgR91KhRgd58IqLQwmalREGPsRUR+VVJIXc4kYFW37M4BqGMxNBJqTfeeEMlmW699VYcPHhQJZtuuukmPPbYY87nPPDAA8jLy8ONN96IzMxMDBgwAFOnTkVsLBvqERH5hGP1DiIKPoytiMivLExKERkpKWUzXk7K2EmppKQkTJgwQV0qI9VSTz31lLoQEZEvGW8JWSLyDGMrIvKlCJQZuLIcX5WTiALH5BhUthpw5kPppgJERETVYqUUERERuYGVUkSGqpSy2piUIiKiYGXAkRUKfXa+74iIgobMYimFSSkiQ4hkUoqIiIKfIynFnlIUAHzbEREZn6nsRBxO3yMyBBOTUkREFPRYsUJEREQeVUqxpxSREZgdn02bAeN59pQiIiIPsacUERERVRQhcPoekRGZHJkf9pQiIqIgZryRFQp9fNcREQUP9pQiMqZIR1aKSSkiIgpeerkvm/tQALA+j4goGCulOH2PyAjMeqUUp+8REVHwY3qAAvG24/uOiMjoWClFZEwmRxzFSikiIgpirJSiALzrOH+PiCioKqVKDSGwUorIEMyO1fdsNuMFVmx0TkRERIbHOikioiBk5fQ9IkNVStmZlCIiomBlwJMYhT47W50TEQXx9D0mpYiMINLM6XtERBT0OH2PAoctpYiIjM9UdiIOk1JEhmBmTykiIgqdSilOpKIAvO2IiMjwWClFZEwmR08pC3tKERFR0GPJCgXibcdkKBFR8GFPKSJDVUrZmJQiIqKgxUopCsTbjnudiCh4BxA4fY/IUKvvWQ1Ygs7V94iIyE3sKUWBwwI9IqIgnL5nLQ7UphBRRUkpVkoREVHQYqUUBfJtR0REhsdKKSJjMjMpRUREwY+VUhS49x3b6xMRBWFSipVSRIZg0ntK2WXAz1gjfpy+R0RE7mGlFAWAweImIiKqStkRBCaliAxVKSWMNoOPSSkiInITK6UocEkp9pQiIjI+U0SZPy8t7ClFZLSklMVmg5EwKUVERO6x6ycwTqQiAzTPJSIiY0/fk5uslCIyXqWUDYbCpBQREbmHJSsUAHa9Qo+IiAyPPaWIjCnSJSllNVhvBCaliIjITfoJjBUr5D8Gi5uIiMijnlIl3F9EBmp0LqwGayrFpBQREbmHlVIUyFQoc6FEREFYKVUUqE0hosoqpZiUIiKi4KR3nOZ4BvnxXae/7VihR0QUfI3OSwoDtSlE5MJkinAO8DEpRUREQT6PiiUr5Me3nSMZ6jLAR0REBlVuAMFSEKhNIaJKqqW4+h4REQX36nucR0V+fd/xbUdEFLQrpdosgNUSqM0hogpW4LNY2VOKiIiCGZNSFIi3HSv0iIgMz1RRy2JWSxEZQpRJ+3xa2FOKiIiIiIiIQr5SSrCvFJEhmM3a59Nqc8x+MIhIT/9DRkYG5syZgx07diA/Px9169ZFt27d0K9fP8TGxvpmK4mIyIC9pYj8+LZzrsMXehhbEVEoVrXaTY4/NUvyA7dBROQUadBKKbeTUl999RVef/11/Pfff6hfvz4aNWqEuLg4HD16FFu3blUJqdGjR+PBBx9Es2bNfLvVREREREGOsRURhfLqe3ZzlHbDwhX4iAzV6NwahEkpqYSKjo7GNddcg59++gnp6emlHi8qKsKCBQvw7bffomfPnnj77bdx8cUX+2qbiYgoEPSSfFZKkV/fd9qVwQb1ThhjKyIK2aSU/vvaFK1ds1KKyBCiIrWgqtgahNP3XnjhBQwfPrzSx2NiYjB48GB1efbZZ7F9+3ZvbiMRERmBc/QzxLIDFBRTQUItF8rYiohCnd1s1m5YigK9KUQEINqsxfLFliBMSlWVkCorLS1NXYiIKESTUnZjncgotDkqzUOupxRjKyIK+el7Jsf0vZKCwG0QETnFRGqJ4qJgTEpV5ODBg+piK9O5vWvXrt7YLiIiMhompSgATBGhWSlVEcZWRBRaSSnHn5rsKUVkCNGR2uezqMSKoE5KLV26FFdffTXWr18PuyNClKU/5bZcW63G+gGJiMhb2FOKAtnKLHSzUoytiCgksdE5kaHEOJJSQdlTytWYMWPQtm1bfPTRR2oVPklEERFRGGBPKQqgEM5JMbYiopDrA1i6Uoo9pYiMIDbKMX2vJMiTUtu2bVMr8LVu3do3W0RERMbE6XsUiLeds6dU6GJsRUQhPX2PPaWIDFUpVWgx1uy247813HT66adj5cqVvtkaIiIKguyAsUZXKNSFfk8pxlZEFEqVUnqtFHtKERlLjKNSqjDYK6U+/PBD1VNqzZo16Ny5M6KiHKsqOJx33nne3D4iIjIM9pSiALzrHG+7UO4WwNiKiEKG6+9qM6fvERlJrN5TKthX31uwYAHmzZuHKVOmlHuMjc6JiEJYKGcFiAKIsRURhQqT60ScCCaliIwkJsoxfa8kyKfvjR07FldccQX27dsHm81W6sKV94iIwkEIz6MiCgDGVkQUKlwXwbKbtalCsLLROZERxEQ6Gp0brFLK46TUkSNHcPfdd6uV94iIKAyFcnMfMpxweLsxtiKikOop5fi9bY9wJKW4+h6RIcSGSqXUBRdcgH///dc3W0NERMblHP0MgywBGYbNpr3fzKbQnT7K2IqIQrFSij2liIwlxqCVUh73lGrbti3GjRuHuXPnokuXLuUand9xxx3e3D4iIjKacChdIcOwOd5vodzSjLEVEYUiu4nT94iMJMbR6LzIYg3+1fcSExMxa9YsdSmbGWdSiogoVIVwVoAMy1EoBVMIZ6UYWxFRSHJO3ysO9JYQEVyTUkFeKZWRkeGbLSEiIiIqw2rTAidzCCelGFsRUShx/rY2Of7UZKNzIkOIjXJM3wv2nlJr1qyp9LFJkyad6PYQEZFhcdoe+Z/FUSoVaQ7dpBRjKyIKLY7f187peyUB3Roi0sREGbNSyuOk1PDhwysc0fvpp58wevRob20XEREZtZdUCFeskPFYrI6kVAg3OmdsRUShKMKZlOL0PSIjiDFoo3OPk1LXX389hg4div379zvv++6773DVVVfh008/9fb2ERGR4YRucoCMx6JP3zN5HLIEDcZWRBRatMEEuz59z1IU2M0hotI9pQw2fc/jnlJPPvkkjh49qhJTs2fPxtSpU1Uw9cUXX+DCCy/0zVYSEZGBKqVCNzlAxlPsqJSKdgRSoYixFRGFogh9MMFmCfSmEBGMWynlcVJKvPHGG2qqXt++fbFnzx588803GDlypPe3joiIjMPuGFXRy/GJ/KDYsWyxProXqhhbEVGoiLCXWX2PPaWIDCHGoD2l3EpK/fbbb+Xuu+CCCzBnzhxcdtlliIiIcD7nvPPO8/5WEhFR4NkdJzBWSpEf6YFTqFVKMbYiopDnXH2PPaWIjCDWUSlVGIzT90aNGlXpYx9//LG6CElOWa3G+gGJiMhLbNbSI59EflDsSEqFWqUUYysiCnmcvkdkyEqpwmBMStkcTUaJiCiM6T0hzDWa+U1UIwWOwCkuKrSSoYytiCgU2fX+k6qw2hEvsKcUkSHEOAb49AE/owitYUciIvJ9pZQeZBL5QWGxIykVHVpJKSKiUGSz25w9pSL0HpTsKUVkCDEGbXTuVlLq22+/dfsFd+3ahXnz5p3INhERkRHZSrRrJqXIj/JDtFKKsRURhSIbjv+x60xKsVKKyFCVUhabHRarLbiSUu+88w46dOiAl156CevXry/3eFZWFiZPnozLL78c3bt3x5EjR3yxrUREFEj6SCeTUuRH+SFaKcXYiohCkh2I0G/rPSiZlCIyVE8po1VLuTUHY9asWWqVGFmueNy4cUhISED9+vURGxuLY8eOYf/+/ahTpw6uueYarFmzRj1GREQhRl89xxwd6C2hMFFitTn7HiTGhNa0UcZWRBSKrPbjDZRNeg9Kffo/EQVUtPl4Ukriq4QYGILbEd55552nLocPH8bcuXOxY8cOFBQUqGRUt27d1MWkr7BAREShWykVaZAzGIW8vCJHc30ACSGWlBKMrYgo1NilVErvKcVKKSJDiTSbEGmKUNP3gq5SypUkoapaxpiIiEKUtUi7ZqUU+UmuIyklPRCiXEb3Qg1jKyIKpUopffqeiavvERlOTKQJlmIrCh09O40gdCM8IiLyLgun75F/ZRdoSamk2CjueiKiYFt9z+zoKWV33EFEARfjWDimONganRMREcFSqO2EyFjuDPKL7EJtymhKXOhN3SMiCkVWl/5RpgjHn5p24/zxSxTuYh0r8LFSioiIgo/FMX2PPaXIT7IKtKRUchwrpYiIgmb6nl4pZdIrpYwzTYgo3MU4KqWM1FPK8JVSe/bswRVXXIG0tDTExcWhS5cu+O+//5yP2+12PPbYY2jYsKF6fOjQodi8eXNAt5mIKLQrpdjonPyclOL0Pa9ibEVEPp2+p3/hbHTOpBSRkXpKBX2l1FNPPYX8/Pxy98tKfPKYNx07dgynnHIKoqKiMGXKFKxbtw6vvvoqateu7XzOSy+9hIkTJ+Ldd9/FokWLkJCQgOHDh6Ow0PHHExEReQen75GfZeZrfcxSE6JDet8ztiKiUGGxWY6vvmfS01PsKUVkuJ5SliCulHryySeRm5tb7n5JVMlj3vTiiy8iPT0dn3zyCXr37o0WLVpg2LBhaNWqlbNKasKECRg/fjxGjhyJrl274vPPP8fevXsxadIkr24LEVHYKynQdkFUfNjvCvKPI3laUqp2fGgnpRhbEVEoJaWclVL6LTY6JzJcpVSRgZJSHncOlURQRMTxXzW6lStXIjU1Fd7022+/qaqniy++GLNmzULjxo1x66234oYbblCPZ2RkYP/+/WrKni4lJQV9+vTBggULcOmll1b4ukVFReqiy87OVtclJSXqQt6n71fuX6Lg/SxFWgpVeFkSESkb4PfvT+HnSI52rq4VZ/b6e77sZymQ5yfGVt5nhONKnuNxC/5jVyADWI7CKKsjGWWHHRZ+Fg2Ln7vwOm7RZi2Xk19Y7PNzpLuv73ZSSqbMSTJKLm3bti2VmLJarap66uabb4Y3bdu2De+88w7uuecePPzww1iyZAnuuOMOREdH4+qrr1YJKVG/fv1S/0++1h+ryPPPP19hVde0adMQH88KAF+aPn26T1+fKFz4/bNkt+E8R6XUjFnzURS11r/fn8LS+gwZzTNhz7aNmJy3waefpYpaE/gaYyvfY9wRnHjcgtfceXOR7Li9fsMGDIiWEMKGyZMnB3jLqDr83IXHccs8osVWS5avRNTeFfAld2Mrt5NSMk1ORvLGjBmjEjpSkaSTJFHz5s3Rr18/eJPNZkPPnj3x3HPPqa+7deuGNWvWqP5RkpSqqXHjxqlEl2ullEwTlKmBycn6r1HydpZUPjBnnHGG6hFGREH2WSrKQcQKbcTz9BEjOYWP/OL9HQuAYzk4rX9PnNaurk8/S3rVtD8xtvIdxh3Bicct+I9d997dsfVD7b4OnToCm6W3lAkjRowI9CZSJfi5C6/j9lfOSqw5dgDtOnTCiL5NfbqN7sZWbiel9CSQ9HXq37+/X/4YkhX1OnbsWOq+Dh064KefflK3GzRooK4PHDignquTr08++eRKXzcmJkZdypKfiQkT3+I+JgrSz1KBY8pzhAlRcclABdO4ibztYI7WU6px7QSfvd/1z1Igzv+MrXyPcUdw4nELXjaTDRGO6XuRZu1PTYkY+DeW8fFzFx7HLTZa+1xa7L7/XLr7+h73lBo0aJCqYNq0aRMOHjyobrsaOHAgvEVW3tu4cWOp++T7NmvWzJkgk8TUjBkznEkoycbJKny33HKL17aDiCjsFTsWuIhJYkKK/KLEasPhXC0ZWj85NqT3OmMrIgoVJdaS443OOYBFZDgxkdrqe0UlQdzofOHChbj88suxY8cONZ3PlfSZkv5S3nL33XerqiyZvnfJJZdg8eLFeP/999VF/3533XUXnnnmGbRp00YlqR599FE0atQIo0aN8tp2EBGFvUJH+W0MpziTfxzILlQLNkWbTUhLCO3V9xhbEVGoKLZpFa4Ki6qJDLv6XqHFe3kbvyelpJm59Hn6888/1ZS5ilbi85ZevXrhl19+UT2gnnrqKZV0kv4Lo0ePdj7ngQceQF5eHm688UZkZmZiwIABmDp1KmJjQ3tUlYjIrwqOaddxtbjjyS92HtWaYzapHQeTKbT/smFsRUShoshaBFPpugUgQqvMIKLAi4kyBX+l1ObNm/Hjjz+idevW8IdzzjlHXSojSTFJWMmFiIh8pDBTu45lUor8Y/dRbbXH9NTQXxWXsRURhYoiSxHgSEo5axcitD+CichA0/csxklKefwbok+fPtiyZYtvtoaIiIyJlVIUoEqp9NS4kN/3jK2IKFQUWgthKtPihb2liAw4fa8kyKbvrVq1ynl77NixuPfee7F//3506dKlXEf1rl27en8riYgosPIOadfxdXgkyC+2HtKa6zdPSwjJPc7YiohCNSkVo+ek9Eopk8eTc4jIx0mpYqtxKqXc+g0hK9vJNDnXxuZjxoxx3tYf83ajcyIiMlhSKrFeoLeEwsTmg1pSqk39JIQixlZEFIoKSgpQy5mU0ufxcfoekVHERAXp6nsZGRm+3xIiIjKu3IPadULdQG8JhYESqw3bD+ep223qJSIUMbYiolBUYC1wFkhF6M2lWClFZLhKqaJgW32vWbNmvt8SIiIyLlZKkR9lHM6DxWZHQrQZDVNCczVdxlZEFIryS/KdBVKwO/7oNXH1PSLj9ZSywSg8nuD722+/VXi/TN2LjY1Vq/K1aNHCG9tGRERGkb1Xu05qFOgtoTCwZk+Wuu7YKFnFF6GOsRURhYq8kjyYnH/rOrJT5pgAbhERVbT6XtD1lHI1atSocv2lyvaVGjBgACZNmoTatWt7c1uJiCgQbFYgZ592O6UxjwH53Jo92eq6U6OUsNjbjK2IKFTkWfKcy7tH6CVTZjY6JzKKmCjjTd/zuOvc9OnT0atXL3WdlZWlLnJbljP+448/MHv2bBw5cgT33Xefb7aYiIj830/KZgEizEBife598lulVJfG4ZGUYmxFRKEitzj3eKWUPn3PHB3ITSKiinpKBfP0vTvvvBPvv/8++vfv77zv9NNPV1P3brzxRqxduxYTJkwotTofEREFsazd2nVSQ/aFIL80OV/tSEp1bRIeSSnGVkQUKnJLco8vugeLdsMcFdBtIqLy0/cKg7lSauvWrUhOTi53v9y3bds2dbtNmzY4fPiwd7aQiIgC66j2ux2p7BdIvrdqdxYKSqxITYhGq7qhufJeWYytiChUZBVnwaR3eZEqaxEZF8hNIqIKKqWKLbbgTUr16NED999/Pw4dOuS8T24/8MADalqf2Lx5M9LT0727pUREFBhHt2rXqS15BMjnFm47oq57N0+FyRT6Tc4FYysiChXZRdnO6XsRdj0pxUbnREYR6+wpFcTT9z766COMHDkSTZo0cSaedu3ahZYtW+LXX39VX+fm5mL8+PHe31oiIgpgpRSTUuS/pFTflqlhs7sZWxFRKLDarVqjc71SypmUig3kZhFRBdP3grqnVLt27bBu3TpMmzYNmzZtct53xhlnwGQyOVeRISKiEHFwg3Zdp02gt4RCXG6RBYu2HVW3T2ldB+GCsRURhYI8e5661pNSEbYi7UYUk1JERpu+V2ixwm63IyIi8FXpNVqfU5JPZ555proQEVEIs5YAhxxJqfqdA701FOJmbjyIYqsNLeokoHW98OgnpWNsRUTBLteWq66j7PJHrw2wFWsPRIfX73OiYKiUstsBi82OKHOQJKUmTpyoVtaTFfbkdlXuuOMOb20bEREF2uFNgK0EiEkGajUN9NZQiPtr7QF1PaxTfUOM3PkSYysiCjW5di0pFelISjkrpaITArthROQU4+gppfeVijJ73GY8MEmp1157DaNHj1ZJKbldGQkgmZQiIgoh+1dr1/U7yS/5QG8NhbCCYiv+3XBQ3R7eqQFCHWMrIgrVSimz3REvWAu166j4AG4VEbmKdklCFZVYkRhTo8lzXuXWFmRkZFR4m4iIQtzuJdp1o26B3hIKcZNX71M9pZrUjsPJTWoh1DG2IqJQk2XPUteRNi0pFWHJ1x6QamsiMgSTKUIlpqRdglyMoMa1WsXFxdi4cSMsFseqCkREFHp2LtKu0/sEeksoxH27ZKe6vrRXugqYwhFjKyIKZlk2LSlltmmdziMsWuNzxDIpRWQk0Y5m50ZZgc/jpFR+fj6uu+46xMfHo1OnTti5Uwsix44dixdeeMEX20hERIFQmA0cXKvdbtqXx4B8ZsvBHCzZfgxmUwQu7pkednuasRURhYJsW7a6Nlkdd1i06XyslCIy5gp8RZYgTUqNGzcOK1euxMyZM1WPKd3QoUPx3XffeXv7iIgoUHYtAuw2oFYzICn0e/xQ4Hw6f7u6HtKuHuonh9/S4YytiCgUZNoy1XWEY0pQRAkrpYiMnZSywgg87mo1adIklXzq27dvqZVxpGpq69at3t4+IiIKlC0ztOuWg3kMyGcOZBfi+yW71e3rT20RlnuasRURBTub3YajtqNqnfkIq/aHbkSxlqRCXO3AbhwRlRITZVbXxcFaKXXo0CHUq1ev3P15eXkhv3wzEVFY2fK3dt16aKC3hELY+7O3qUabvZrXRp8WqQhHjK2IKNgdzD8ICyyIsR+veYgoOqbdiAvP3+1ERhUT7NP3evbsiT///NP5tZ6I+vDDD9GvXz/vbh0REQXGse3Akc1AhBloOYhHgXzicG4Rvlq0Q92+/bQ2YTu4xdiKiILdrtxd6jo9rpHzvogSrccU4tMCtVlEVFWj82Cdvvfcc8/hrLPOwrp169TKe6+//rq6PX/+fMyaNcs3W0lERP61dpJ23bQfEJvCvU8+8eyf61FYYsNJTVIwsE2dsN3LjK2IKNhty9qmrpvHNQaQoW5HRNiACBOn7xEZTLTZFNzT9wYMGIAVK1aohFSXLl0wbdo0NZ1vwYIF6NGjh2+2koiI/MduB1Z+o93uegn3PPnErE2H8MvyPZDiqCdHdg7bKinB2IqIgt2GoxvUdduElqX/0kxqCJg9roMgIh+KNGsxV4nVboj9XKPfEK1atcIHH3zg/a0hIqLA27sMOLQBiIwFOo0K9NZQCMovtuCRX1ar29f0b46T02sh3DG2IqJgtvHYRnXdOqGZuo6IilSDDkhJD/CWEVFZkSatNslqC7KkVHa2Y05wNZKTk09ke4iIKNCWf6lddziXU/fIJ56bvB67jxWgca043DesXdjuZcZWRBQKiq3F2JK1Rd1uHd8MhZKUcvSsQS0mpYiMWyllQ1AlpWrVqlVlab1dlv+MiIDVsQQoEREFoZz9wPKvtNvdrwr01lAI+mLBdny5cKe6/cz5nZEQE77TOhhbEVEoWHVoFSw2CxIiElDHnILdkpQyOyowUlsFevOIqAyTI69jjDopD5JS//77b6kE1IgRI9SKe40bSzM7IiIKCfNeB6xFQHpfoPmpgd4aCjFzNh/CE7+vU7fvH94OQ9rVQzhjbEVEoWDR/kXqumVkS6CoSN02RTgKFRp0CeSmEVEFJJ9jJG4npQYNKr0kuNlsRt++fdGypUszOyIiCl45B4D/PtZuD3pAls0J9BZRCNlyMBe3frVM9S+4oFtj3DqYo+eMrYgoFCzapyWlWkW2gs2RlIqIKNYebNA5kJtGREHA49X3iIgoRM1+CbAUAo17Aq1OC/TWUAjZeigXV360CDmFFvRoVhvPX9glrFfbIyIKFTnFOVh9aLWzUsquJ6XMNiAmGailNT4nIuOww1jCt5EDEREdt3c5sOQj7fbpj7FKirxm3d5slZA6kleM1vUS8d6VPRATaeYeJiIKATN3zYTFbkGL5BZINaXCXlCg7jdJT6mGJzGeIDKwCIRApRRHOYmIQoDNCvxxjzZu0uVioGXp6dpENbVs5zFc+v4ClZDq1CgZ393YF3USY7hDq8DYioiCyd87/lbXp6efrq5telIq0g406RXQbSOiihmspZT7lVIXXHBBqa8LCwtx8803IyEhodT9P//8s/e2joiIfG/pp8DeZVqZ/bBnuMfJa03Nb/piKfKLrejZrDY+vrYXkmOjuHddMLYiomCWX5KPeXvnqdtDmw7Flj1bjldKqaRUzwBvIREFw0CY20mplJSUUl9fccUVvtgeIiLypyNbgWmPardPGw8kNeD+pxNis9nxzqyteHXaRtjswKlt6qgpe/HR7BhQFmMrIgpm/+z6B0XWIqQnpaNNrTbYgi2wZR9Tj0XI9D3pUUlEhmOHsbgdIX7yySe+3RIiIvIvSxHw47VASR7Q/FSg1/U8AnRCsvJLcO8PK/D3+oPq64t7NMEz53dmD6lKMLYiomD206af1PV5rc5zVlzY925U1+akZCCpfkC3j4iqZow6KTY6JyIKX38/AexbCcSlAhe8L11JA71FFMTW7s3CLV8uw86j+YiONOHpkZ3wf72aBnqziIjIB7Znbcd/B/6DKcKEUa1HOe+3HcxQ16a66dzvRAZlN1hTKdbSExGFo/V/AAvf1m6PegdIbhToLaIgZbHa8On87Xj5r40ostjQpHYc3r2iBzo3Lj3tn4iIQsfPW7Q+wqc0OgUNEhqgpKREfW0/vEddmxq0Cuj2EVH1DNJSikkpIqKws3MR8NN12u0+twDtzgz0FlEQr673yC9rsH5ftvp6SLu6eO3/Tkat+OhAbxoREflITnEOftz4o7p9UduLnPfHlhyDLScTQBxMTTpy/xMZlN1RKMWkFBER+d+hjcDXlwCWQqDtmVxtj2okM78YL07diG+X7FSBTUpcFMad1R6X9EyHyWSQYTciIvKJr9d/jZySHLSu1RqD0wc772+QuQzWYpO6ba7XhHufyKCsshKNfE5N2uc10Dh9j4goXGTtAb64ACjMBJr0Ai76BDDzNECe9SD4ZfkePPvnehzJK1b3XdSjiUpIpSXGcFcSEYW4/JJ8fLH+C3X7hi43qJ5SuoZZS5FXrA1MmJOTA7aNROReUirSIAOJ/GuEiCgcZO4CvjgfyN4NpLUBLvsOiI4P9FZRkLDZ7Ji+/gDe+Gcz1uzRpuq1rpeIZ0Z1Rt+WaYHePCIi8pMv13+JrKIsNEtuhuHNhx9/oOAY6uSsR3axdk4wyep7RGRIVsf8PZNB5u8xKUVEFOoObgC+vADI3gMkNwau+AlIYCKB3EtGTVmzXyWjNuzPUffFR5tx+2mtcf2AlmqVPSIiCg97c/fig1UfqNs3n3QzzC6r9ppWfQsTrLCWREldLcy1agVwS4moKsUWm7qOMUgcx6QUEVEo27UY+OpibcpenbbAlb8AKezzQNWXdf+xai/e/GcLNh/MVfclxkTi6v7NcN2AlkhNYCNzIqJw8+LiF1FoLUSP+j1wdouzjz9gt8O07FPYrYCt2DEtKLV24DaUiKpUWGJV1zFRTEoREZEvbZ4OfH8VUJIPNO4JjP4BiE/lPqcqg5TfV+7FOzO3YtvhPHVfUmwkxpzSAtee0pyr6hERhalZu2bhn13/IDIiEuP7jEeE67SfjNmIOLoVhZY47WuzGSb2lCIyrEKLlpSKjTpe7RhIrJQiIgo1Nhsw91Xg3+cAuw1oPRS45HMgOiHQW0YGtXF/Dr5ZvBM/L9uN7EKLuq9WfBSuH9ACV/VvjuRYmY5BREThSHpIPb3waXX7yo5XonXt1qWfsPh9dbUnphuA7TCnpCDCIKt6EVF5hSWcvkdERL6SfxT45SZg8zTt625XAue8BpiZVKAyb5ViC/5YtQ/fLt6JZTsznfc3rhWHK/o2w5X9mqkpe0REFN6rrj4671EcyD+gmptLL6lSDm0ENvyhbu6L7Io0bEdk3bqB2Vgi8mj6XhwrpYiIyKv2LAO+vxrI2glExgIjXgG6X8mdTKWs3ZulqqJ+Xb4XOUUW55LAQzvUx6W903Fqm7owG2SJYCIiCqxvN36Lf3f9iyhTFF4e+DLio8qs3DtvorqytR2BkvVav0EmpYiMrchRKcXpe0RE5L3peks+BKY9AliLgdottOl6DbtyD5OyN7NAraL364o9WLU7y7lXmqXF4/96peOiHk1QLymWe4uIiJw2Ht2IV5a8om7f3eNudEjrUHrvZO0BVn2nhSL974B5sXabSSkiYy9mU2xlUoqIiLzl2A7gt9tVk1Gl/TnAyLeAOC7FHO52HsnHlDX7MHnNfqzcdXx6XpQ5AsM7NcBlvZuiX8s0mFgVRUREZRzKP4Tb/7kdxbZiDGwyEFd0uKL8Ppr9MmArAZoNgL1xT0Rmf6DuZlKKyPhT90QsV98jIqIas9uB/z4Gpj0KFOcCUk4/9Amg942A64o4FFa2HsrF1DX7MXn1Pqzdm+28X94SvZql4szODTDy5EZIS4wJ6HYSEZFx5ZfkY+w/Y7E/bz+aJzfHcwOeK73anji8BVj2uXb7tPHqKuqYNgAS1aiR37eZiGqQlIrk6ntERFQDccWHYf7mYiBjpnZH035adVRaK+7PMGxAu+lArkpCSTJq44Ec52NSANW3ZRrO6tIQwzvV5/Q8IiKqltVmxcNzH8baI2tRK6YW3j79baTEpJR/4r/PAHYr0GY40KwfUFKCyEw9KdWQe5rIoAot2tS96EiTYarluawOEVGwsFlh+u8jDFn/OEy2QiAyDjj9MaDPzQCXXg6rVfMWbjuCWRsPYeamQ9hxJN/5mDQsP6V1HZzVuQHO6FifFVFEROTRQMfL/72MGTtnqMbmrw95HenJ6eWfuGcpsPYXqcPV4hCHqMxj2nVDJqWIjKqgWKuUio00wSiYlCIiCgZ7VwB/3AXz3uWQQltbk94wjXoHqNM60FtGfqqGmrXpIGZtOoQlGcecDSr1ka6BbSQR1VCtoJcSH8VjQkREHp9rJi6fiK/Wf6W+fvqUp9G9fveKF1eZ/IB2u+v/AQ06q5vW7GyYCwrVbU7fIzL+9L3YKGNM3RNMShERGVlhNvDvs8Di9wG7DfaYZKyqez46XvkSTDFcLS1UZeWXYO6WwyoRNXvTYezP1gJ9XZPacRjUtq669G9dB4kxPJ0TEVHNvbfqPXy4+kN1+5E+j+DslmdX/MQVXwJ7/gOiE7Velg4lu3ara3NaGkwJCTwURAZPSsVHMylFRETVNTJf/xsw5UEgZ592X+eLYDn9SWyfvRQdTcY5kdCJs1htWL0nSyWgZm8+hOU7j8FmP/54TKQJ/VqlqSTUwLZ10bJOQvmms0RERDXw6ZpP8daKt9Tt+3reh0vbX1rxEwuOAX87ElGDHwKSj0/TK9m1S11HpVcw3Y+IDKOwRKu2Z6UUERFV7shWYPL9wNYZ2tepLYGzXwVanaYaiVJoJKFkdTzpDSWXJduPIbfIUuo5beolOpNQvVukGip4ICKi0ElIvbr0VXV7bLexuLrT1ZU/+e8ngfwjQN32Wj9LFyW79aRUE99uMBGdkAJHpVSMgeJK1vsTERlFcT4w93/AvNcBazFgjgZOuQs49V4gilP1Qj0JlRwbif6t6mBQOy0R1bhWXMC2l4iIQr+H1Dsr31EXcVPXm3Bj1xsr/w8Zs4Gln2i3ZaDMXLp/YUnGdnUd1bSZD7eaiLyVlIpnUoqIiEpN1ds4GZjyEJC1U7uv1enAiJeBtFbcUUGahFq3T09CHcWSjKPIqSAJ1btFGvq2TEXflmno0DAZZoMszUtERKGdkPrf0v/h07Wfqq/v6HYHbuh6Q+X/oTgP+G2sdrvnGKD5gPJP2e5ISjVv7qOtJiLvNjrn6ntERCSObtP6Rm2epu2PlHTgzOeB9ucA7BkUNOQEv2ZPFpbuOIZFGRUnoZJiI9GHSSgiIgogm92G5xY9h+82fqe+fqDXA7iy45VV/6d/ngGObQeSmwBDn6wwyVXiSEpFt2BSiigYklJxbHRORBTmyk7VM0UBp9yhTdWL5qo1Rrc3swDLdh5TSahlOzOxbm8WSqz2CpJQWhUUK6GIiCjQiqxFGDdnHKbvmI4IROCxfo/horYXVf2fts0EFmpT/HDu60BscrmnWI8cgS07G/aICEQ1beqjrScibygodlRKRbKnFBFRGK+q9zvw18NAltYUFC2HaFP16rQJ9NZRBYosVtUPaplKQB3Dsh2Z2J9dWO55dRJj0L1pLfRqnqpWyuN0PCIiMoqsoizc8c8dWHZwGSJNkXhuwHM4q8VZVf+nvMPAzzdJ8AJ0vxpoM7TCpxVt3qyuS1JTYYpjP0SiYFh9L46VUkREYejwZmDKA8DWf45P1Rv+HNDhXE7VM5AD2YXOBJRUQq3Zm41ii3YC10nvpw4Nk9C9aW30aFZbXTepHYcITrkkIiKD2ZO7B7f8fQsysjKQFJWECUMmoHfD3tUPok26FcjdD9RpB5z5QqVP1ZNSxQ3qe3vTichHjc6NtKozV98jIvLHVL3ZLwHz3wRsJY5V9e4EBtwDRMdz/wdQTmEJ1uzJxqrdmVi1JwsrdmZiT2ZBueelJkSrKqhuTbUE1EnpKYiP5imUiIiMbe3htbj9n9txuOAw6sfXxztD30Gb2m5UZi96F9j8F2COAS76uMp4pXDTJnVd1KCBNzediHzZU4pJqZp54YUXMG7cONx5552YMGGCtlMLC3Hvvffi22+/RVFREYYPH463334b9eszU09EBrBxKjD5/uOr6rUZrjUy56p6AZlDv3ZvFlbtzsLqPVlYuTsT2w7llXueLIDXtn6SswKqe7PaaJ4WzyooCkmMrYhC15SMKXh03qOql1Tb2m3x9ulvo36CG38jbZ8HTBuv3R72NNCgc5VPL9qwUbtmUoooaHpKxXH6nueWLFmC9957D127di11/913340///wTP/zwA1JSUnD77bfjggsuwLx587x02IiIaiBzFzD1IWDDH8en6p31EtB+BHenn/pAbdiXo6qfVksV1O4sbDqQA1vpXuRK41px6NI4BV3TU3BSk1o4Kb0WEmNYBUWhj7EVUeiusPfWirfw/qr31dcDmwzEi6e+iMToxOr/c9Zu4PurAJsF6HwR0PvGKp9uLy5G0UZHUqpxY+/8AETkM/mslKqZ3NxcjB49Gh988AGeeeYZ5/1ZWVn46KOP8PXXX+O0005T933yySfo0KEDFi5ciL59+3rp0BERuclaoq1SM/N5oCQfMEUC/W4HBj3AVfV8xGK1YdOBXKzek4mVUgW1Owsb9meXWw1P1E2KwUlNUtClcS2VhJJklDQoJwo3jK2IQlN+ST4envswZuycob6+ttO1uLP7nTCb3OgfU1IAfHcFkH8YaNAFOO+NanteFm3dCntJCUxJSarROREZWwErpWrmtttuw9lnn42hQ4eWSkotXboUJSUl6n5d+/bt0bRpUyxYsKDSpJRM85OLLjs7W13La8mFvE/fr9y/FNL2r0bk72MRcXCN+tKW3hfWM18G6nXQHvfC75dw/ywVWWzYfCAXa/dlqxXx5Hrj/lx1f1m146PQuVGySjx1aZyMzo2TUT8pptw0vHDdl+Gu7Gcp3N4HoR5bhetxDXY8bife0Pye2fdgc+ZmRJmi8GjvR3FOy3Ngs9rUpUp2O8y/3Q7T3uWwx6XCcuFnQERUtbFL3qpV6jq6fXuVwOJnLvjwcxdexy2/WHt+lMn350h3X9/w8xOkV9SyZctUiXlZ+/fvR3R0NGrVqlXqfuknJY9V5vnnn8eTTz5Z7v5p06YhPp5Nh31p+vTpPn19okCIsFvQ5sAfaLfvV0TAiiJzItY1vhQ7UwcA/2UAkIt3hcNnSQZy9uQDu/MisCs3Ql3vK5BpCeVHbWPNdqQn2NE0EUhPtKNpgh2pMRZERBQAxQdQnAEs8/5hoBCgf5by8/MRLsIptgqH35WhiMfNc1tKtuC7/O9QYC9AYkQiLo+/HKYNJkzeMNmt/99h7w9oe+B32GDGgsY34PB8GWDTBtmqUu/PyZDfFnsdn3Meu+DFYxcex23fQamajMC6VSsQtWc5fMnd2MrQSaldu3appuayo2NjY732utIs/Z577ik1mpeeno5hw4YhOTnZa9+HSmdJ5TieccYZiIqK4q6h0HFwPSJ/vw0R+7WRQlu7c2A662V0TqiLqtuC1kyofpZkFbx1+3LURVVA7c3GtsN5FfaAqhUXhY6NktCpYTI6NZJLEprWjodJOpQT1fCzpFf2hLpwia1C9XdlqONx85zdbsfn6z/H5ys/V72kOqd1xsunvqxW2nOXaeknMC//Xd22nTMBvU+6zO3/u/Ojj1EMoN155+KwxcLPXBDi5y68jtu7GQuAnBwM6NsLp7ap49NtdDe2MnRSSkrIDx48iO7duzvvs1qtmD17Nt5880389ddfKC4uRmZmZqkRvQMHDqBBFas/xMTEqEtZcjAZuPgW9zGFDKsFmP86MPMFwFoMxNYCzn4Vps4XwlRN/4Vw/ywdyyvGmr1ZWLMnW12v3ZOF7UcqHkmRfk/a1LsUdGqUoqbgSWPyslPwiE70sxSsnydPhVtsFejvTzXD4+Z+/6gnFjyBKdunqK9HtR6F8X3HI8bsQa/EjVOAvx7Ubg9+GJE9r3L7v1pz81C8ZYu6ndCtm6yewGMXxPi5C4/jVuRoeZEYF+Pz86O7r2/opNTpp5+O1atXl7rv2muvVb0NHnzwQTUCJz/ojBkzcOGFF6rHN27ciJ07d6Jfv34B2moiCnm5h4CfxgAZs7Wv254FnDsBSKr8D7ZwdTC78HgCak+WqoDak1lQ4XMl2SSVT5KAkuRT50YpqJfsvUoOImJsRRQqdmXvwl0z78KmY5sQGRGJB3s/iP9r93+eDdpsnwv8cC1gtwHdrtQWZfFA4ZrVgM2GyEYNEVm3ruc/BBH5Xb7e6DzKjcUP/MTQSamkpCR07lx6AkxCQgLS0tKc91933XWqXDw1NVWVh48dO1YlpLjyHhH5xK4l2lLJOXuBqATg7FcAKXMP88odmT4gySZJPq1VSagsrNmbjUM5xxsfu2qeFu+ofNISUHI7NSHa79tNFG4YWxEFv7l75uKB2Q8gpzgHqbGp+N/g/6FH/R6evciuxcBXlwCWAqDNcOCc1zyOZfL/W6qu47sdr7wkImMrKHEkpaKZlPKa1157DSaTSVVKyaovw4cPx9tvv+29b0BEJOx2YMmHwNRxgK0EqNMWuOQLoF77sNs/NpsdO47mOxJPMv1Om4aXmV9+hQ1p89SqbqJj+p1WBdWxUTKSYzmdhsioGFsRGXcA6MPVH+KN5W/ADju61umqElL1E9zvH6XsXQ58eSFQkge0HAxc8jlg9vy8nL/0P3Ud39PDhBgRBUyBo1Iqnkmpmps5c2apr6VJ51tvvaUuREQ+UVII/H4HsOo77euOI4GRbwExSSG/w602OzIO52L1niys2q1Nv1u3Nxu5RZZyz40yR6Bt/SQ17a6To/qpQ8MkxEcbuiiXKOwxtiIyvtziXIyfNx4zds5QX1/Y5kI83OdhRJs9rDLevwb44nygKBto2h+49GsgyvOp8vaSEhSsWKlux3VnUoooGFisNhRbtZ5SnL5HRBQsinKAby4Dts8BIszAGU8B/W4Lyel6UgElK96t3pOJ1bu1HlBSAaXPPXcVE2lCh4bJzt5PUgHVpn4iYiKNUwpMREQUCrZlbcNd/96FjKwMRJmiVDLqorYXef5Cu5cCX14AFGYCjXsCo78HohNqtE2F69bBXlAAU0oKYtq0hsVaPlYgImNO3ROcvkdEFAzyj2rl7XuXAdGJwKVfaWXuIZSAksSTVEGtVlVQWcirIAElIyn61LsuqgdUClrVTUCk2RSQbSciIgoXUhn1yNxHkFeSh3rx9fDa4NfQtW5Xz19Impp//X9AcS7QpBcw+ocTqvjOW7RYXcf36okIk0mW8azxaxGRf5NS0l5DBpiNgnMqiIgqkr1XK28/tAGISwWu+Alo3D1oE1AZRxwJqN1ZWLUnq9IpeJKAkp5PknxSlyaSgEqEWc5eRERE5BdWmxVvrXgLH6z+QH0tjcxfGfQK6sTV8fzFNk8HvrsCsBQCLQYCl34DxCSe0PblL1yorhP69D2h1yGiQPSTivRspU4fY1KKiKisrD3AJ2cCmTuBpEbAlb8ETUNzaYK6L6sQK3dlYsXuTHUtK+JVlICKjTKpvk969VNXJqCIiIgCLqsoCw/OeRDz9sxTX1/R4Qrc0/MeNXXPY6t/BH65WVukpe2ZwMWf1aiHlCt7cTHyly9XtxP69jmh1yIi/9FbcsRGGavdBpNSRESuCrOAry7WElK1WwBX/QrUbmbYfZSVX4KVjuTTyt1Z6vahnKIKE1AdGzoqoJrUUtecgkdERGQsG49uVP2jdufuRqw5Fo/3fxzntDynZqsGz3kF+OcZ7etOFwAXvF+jVfbKyl+xQvWTMqelIbp16xN+PSLyb1LKSCvvCSaliIh01hLg+6uAg2uBxPrA1b8BtZoaZv8UllixalcmZu2LwIwfVmP13mxkHM4r9zyZate+QRJOSq+Fk5vUQtf0FLSum8geUERERAY2JWMKHp//OAosBWic2BgThkxA+9QaVGpbioE/7gZWfKl93e92baEWk3f+EM2bN19dJ/Tvb6gpQETk7vQ9JqWIiIxHRhR/vxPYNhOISgAu/z6gCSnpA7XlUC5W7NKroDKxYV8OLDa7pJ0A7HM+t3lavEpAndSklrqWpuRGK8slIiKiillsFry29DV8vu5z9XX/Rv3x0sCXkBKT4vkuK8gEvr8SyJgNRJiAs14Cet/g1V2fN9+RlDqlv1dfl4h8K79Ya+dhtL8TWClFRCTmvwGs+AqIMAOXfAY0Otmv+yWroATLdx7Dsp2Z6nrFzkzkVNAHKi0hGg2iCzG0Wxt0b56Gro1TUDshmseQiIgoCB0rPIb7Z92PRfsXqa+v63wdxnYbC3NNqpoObQS+HQ0c2aytGnzRJ0DbYV7dXsuxYyhcs8ZZKUVEwbf6XjwrpYiIDGbvCmDGU9rts14E2pzh8yqorYdysXSHJKG0RNSWg7nlnicnDOn9dLJUQTkudePNmDJlCkYMaYWoqBPvC0FERESB6x915793Yk/uHsRFxuGZU57BsOY1TCKtnQT8ehtQnKst0nL5d0DDrt7eZK1Kym5HTNu2iKpXz+uvT0S+w55SRERGVJwH/HSdtipNh3OBXtd7/VtkF5aoyic9AbVi5zFkF5avgpJpeN2b1ka3ZrXRvWkttKufVK4PVElJide3j4iIiPxrasZUPDrvURRaC9EksQkmnjYRbWq38fyFrBZgxpPA/Ina181P1SqkEuvCF/Jmz1HXiQNP9cnrE5Hvk1Jx0caaMGesrSEi8re/HgaObNFGFc+dCJxgw0673Y7tR/KxZPtRNQ1PqqE2H8xVLatcxUWZ0bVJCro3q40ekohqWgtpiTEn9rMQERGRoVltVkxcPhEfr/n4xPtH5R4CfrwW2K4litB/LHD6E4DZN3/i2W025M6bp24nDGBSiijYFDh6SsWzpxQRkUFIU/OlnwKIAM5/F4hP9fglLFYb1u/LweLtR/Hf9qNYsv0YDucWlXteemqcSj5JEkqqoWR1vLJVUERERBS6copz8ODsBzFnj5ZEurbztbiz25016x+1ZQYw6RYg94C2QMuot4BO58OXCtevh/XwYUTExyO+ezeffi8i8mWlFBudExEFnrUEmPyAdlum7LUc5PaqFTIVT5JPUg0lU/L0X/C6aLNJVUH1aK4loORSN4lVUEREROFqR/YOjP1nLDKyMhBjjsFT/Z/CiJYjPH8hS5HWB3PBm9rXddsDF38G1GsPX8ubPVtdJ/Tvh4hoLrJCFGzyHX+zsNE5EZERLP4AOLwRiE8DTnuk0qcdyS3CfzuOYUnGUSzZcQxr92TBYis9Fy8pNhI9m9VGrxap6NU8VTUnN9pSq0RERBQY8/fOx32z7lOVUvXi62HikInoVKeT5y90aBPw0xhg/2rt6143AMOeBqLi4A+5s7SkVOLAgX75fkTkXQVMShERGUTuQWDm89rt0x8D4mo7H9qTWYBF245gccZRNSVv26G8cv+9YUqsSj71aq4lotrWS4LJdGK9qIiIiCi0SJ/Jr9Z/hZf/exk2uw1d63bFhMETUDfewybkNhuw9BNg2nigJB+ISwVGvgW0r0GlVQ1Zjh1DwcqV6jaTUkTBKb+Ejc6JiIxh7mtAUTbsDU/G9vQLsHjJTizKOIpF246qpFRZbesnomfzVPRunoqezWujca04RJxgQ3QiIiIKXSXWEjy76Fn8tPkn9fXIViPxaL9H1dQ9jxzdBvx2x/Fm5i0GAee/ByQ3hD/lzZ0nWTbEtG2LqAYN/Pq9icjLjc7ZU4qIKHAjltsytqLZ4o/U0qNjD56HP/7nCPIczKYIdG6cgj6OqXgyLa92AvsmEBERkXsyCzNx98y78d+B/xCBCNzb815c1fEqzwa0bFZg4TvAP88AlgIgKh447VGgz82Ayf8LpeTOmqWuEwe514OTiIwnn9P3iIj8y2qzY/2+bDUVb1HGEdWc/Laij3BdZBH+s7XFH4XtVFPyk9IlCZWG3i1S1ep4iTG+WUqZiIiIQtvWzK24fcbt2J27GwlRCXhp4EsY2MTDHkwH1wO/3g7s+U/7usVA4NyJQGoLBILdYkHuHG0QL3Ewk1JEQb/6XpSxet/yLy8iChkWqw3r9mWraXgLpS/U9qPIKdTKVEVdZGJ0zN/q9s6ud+Lb7v1wcnotNiUnIiKiEzZvzzzV0Dy3JBeNExvjzdPeROvard1/geI8YM7/gHmvA7YSICZZa2Te/WoggG0DpJeULSsLppQUxJ10UsC2g4i81ejcWGkgY20NEZGHSai1e7NVAkp6QskKeTlFx5NQQqqepA+UVEKNzPwUsStKgCa9cMGFowMa4BEREVHo+GbDN3hh8QuqoXn3et3x2pDXkBqb6t5/ttuBdb8Cfz0CZO/W7mt7JnD2/4CUxgi03JmOqXsDBiAikn8+EgWrAmejc/9PAa4Kf6sQUVAlodY4klBy+W/7MeSWSUIlxUaqhuR9W6ahT8tUdGyYjEizCbAUAa99oz2p321MSBEREdGJxyY2C15e8jK+3vC1+vq8Vufh8X6PI9rsZj/KgxuAKQ8AGVriBylNgTOfA9qfY5hYJXfmTHWdOHhwoDeFiLyRlIoyVhrIWFtDRFQmCbV6TxYWOqbj/bf9KPIcZae6ZElCtUhD35ZaIqpDw2TVrLycNT8DeYeA5MZaoEdERER0AnKLc3H/7Psxd89c9fWd3e/EdZ2vc6+heWE2MOtFYNG7gM0CyKp8A+4CTrkLiI43zHEp3r0HRZs3A2YzEk8dEOjNIaITUKj3lOLqe0REVTcmn7/1MBZsPaIalJdNQqXERamG5JKAkkRU+waVJKHKWvy+dt3rOsAcxUNARERENXYg7wBunXErNh3bhFhzLJ479Tmc0eyM6v+jtQRY+ikw8wUg/7B2X7sRwPDnAtbIvCq5s7Qqqfhu3WCuVSvQm0NEXqmUYqNzIiLFZrNj08EclYCav/UIFm07gmyXxuR6EqqPMwmVhvYNkmByJwnl6vBmYO8yIMIMdLuKe5+IiIhqbPOxzbjl71twIP8A0mLT8Obpb6Jznc7V941a/zvw9xPA0a3afamtgDNfANoOM+zRcPaT4qp7REE/+G+x2dXtmEj2lCKiMGW327HtcJ5KQC2Uy7YjOJJXXK4xuSSh+rVKU5cODZI9T0KVtep77br16UBi3RN7LSIiIgpbS/YvwZ3/3Imckhw0T26Od4a+gyZJTar+TzsXAdPGA7sXa1/H1wEGPwT0uMbQ1du2/HzkL1qkbicOGRLozSGiE1BitTlvRzEpRUThZNfRfEcl1GGVjDqYU1TqcSkfldXxJAHVv1UddG7kaEzuLTIyudqRlOpyifdel4iIiMLKlIwpeGTuIyixlaBbvW6YOGQiasVWMaXt0Ebgn6e1CikRFQ/0ux3oPxaITYbR5S1cCHtxMaLS0xHdsmWgN4eIvJWUMhtjEQUdG50TkVcdzClUSSi5zNt6GLuOFpR6PDrShO5Na6kElCSiTmpSS93nM3uWAce2A1EJQPsRvvs+REREFLK+Wv8VXlj8grotvaOeG/AcYiNjK37ywfXArJeAtb/I6BgQYQK6XQEMfhhIbohgkfuvY9W9QYPca95ORIZVYtWm7okoE6fvEVEIySooUb2g5juqoTYdyC31uDQhPzm9Fvq1lEqoNHRvVhux/myut/FP7brNUCA6wX/fl4iIiEKi9cCbK97E+6u0BVMua38ZHuz1IMymCmKZA+u0FfXW/aolo4Ss+HvaeKBeBwTbz507y9FPatCgQG8OEXmpUkr+Njvh1ihexkopIvJIQbEVS7YfVUmoBVsPY/WeLDh65jl1bJiMU1pr0/F6tUhVfaICZsPk40EhERERkZusNiueXfQsftj0g/r69pNvx41dbyxfNXRgrUsyyqHDucCgB4EGXYJyfxetXw/LwYOIiI9HfO9egd4cIjpBxRabIafuCSaliKjaX2Ard2di/hZtOt7yncdKlX+KlnUTVBXUKa3qoE/LNKQmRBtjrx7dBhxar62618aNZZqJiIiIVFVBCR6a8xCm7ZiGCERgfN/xuKTdJaV7Vu5aDMyfCGz44/j9HUcCAx8AGlSzGp/B6VVSCf36wRQTE+jNIaITpK+8F+XN3r1ewqQUEZVbLnTd3mxnY3KpisovtpZ6TqOUWPRvXUcloqQvVMOUOGPuxc1/a9fN+gNxtQO9NURERBQECiwFuHvm3Zi3Zx6iTFF44dQXMKz5MO1BqwVY/xuw4C1gz3+O/xEBdBqlJaPqd0QoyJmp95MaGOhNISIvTt9jUoqIDNkzYOuhXMzbovWEWrjtqOoT5Uoqn7TV8bRqqGZp8cHR8HLrP9p166GB3hIiIiIKArnFubj9n9ux9MBSxEXGYcKQCejfqD9QmAUs+wJY9C6QtUt7sjkG6HqJtqJevfYIFZajR1G4arW6zX5SRKGhmNP3iMhIdh3Nd66OJ9VQh3KKSj0uPaD6tEh1VkO1q59kuIZ41bIUA9vnaLdbDQn01hAREZHBZRZm4pa/b8GaI2uQGJWIt4e+jW5RqcDUh4FlnwPFOdoT4+sAva7XLol1EWpyZ89W0xNjOnRAVP36gd4cIvLi9L1Ig628Jzh9jygMHMwuxAJZIU+qobYdxq6jBaUej4k0oWfz2qoxuSShujROQaQB5xt7ZPcSoDhXCxzrB2eTUSIiIvKPIwVHcMP0G7D52GbUiqmF99peg44zXgI2TQXs2rQX1GkH9LtNq46KMmjrAi84vuoep+4Rhd70vQgYDZNSRCEoM78YCyUJtVW7bDmYW+pxWQr0pCYpOKV1HTUtr3vT2oiNqmBp42C2fa523WIgYMARASIiIjKGwwWHcf1f12Nr1lbUNcfhgwNH0WrDPcef0HKINkWv1WkhH1PYS0qQN3eeus2pe0Sho8Qxfc+IhQdMShGFgNwiC5ZkHHU2J1+3L1stCqOT9k8dGyarKiiphurVIlVN0QtpOxxJqeYDAr0lREREZFCHc/fjusmjsa3gIOpZrPho11Y0t1iAuFTg5MuB7lcDddsiXOQvXw5bTg7MtWsjrmvXQG8OEXlJiWP6XjSTUkTkDYUlVizbccxRCXUYK3dnqVXzXLWpl+hYHa8O+rZMRa346PDZ+dJPatcS7XazUwK9NURERGQ0mTtxcMkHuG7XL9hujkB9iwUf7zuIpk36AT2uAdqfA0TFItzoU/cSTh2ACHOIVdEThTELp+8R0YnOAV61OwsLth5Wq+Qt3XnMuYKCrmlqvCMJlYZ+LdNQLzn8AimnfSsASwEQnwbUbRforSEiIiIjKDgGrJ0ErPoeR3YvxPUN62N7dBQaWG34uOGZSB95K1CnDcKZnpRKGjw40JtCRD7oKcXpe0TkFpvNjvX7s7XG5FsPY3HGUeQVW0s9p35yjEo+yQp5cp2eGs+969rkXKT30eYuEhERUXgqKQQ2/6USUdg8DbAWI8sUgZsa1EeGJKSikvHJyC/QpFZLhLvi3XtQvGWr6puVcAorzYlCSYlVm1XDRudEVCG73Y5th/Mwf4vWE0pWysvMLyn1nNrxUVoVlGOFvJZ1EhDBhEvVSanGPfiOIyIiCjc2G7BzPrDqO2Dtr0BRlvOhvPqdcGvtWGwsOoS02DR8cOanaJLSPKCbaxR5c2ar67hu3WBOSQn05hCRT1bfY6NzInLYk1mAeVsOY4GjL9SB7KJS+yYh2ow+UgnlmJLXoUEyTCZW/bhl91Ltukkvvt+IiIjCgazwsncZsO5XYPVPQPbu448lNwa6XIyiTqMwdvUbWLV/CVJiUvD+sPfRnAkpp9zZc9R14sCBATiARORLFmelFJNSRGHrWF6xqoCa56iGyjicV+rx6EgTejar7WxO3rVJiiF/aRhe3mEga6esOQg06hborSEiIiJfVkTtXgys+w1Y/7vj/O8Qkwx0HAl0/T+16IktAhg36z4s2b8ECVEJeG/oe2hbO3xW1auOrbgYeQsXqtuJA08N9OYQkZcV6z2lDFjkEOJrwhMFTn6xRfWCkgSUJKLW7ctWg3g6sylCJZ4kCXVKqzro3qw2YqO4yskJ27dSu05rBcQmn/jrERERkXFYLdrUPD0Rlbv/+GNR8UCbM4BO5wNtzyq1et6rS17G9B3TEWWKwhunvYFOdToFZvsNqmDpUtgLChBZty5i2rcP9OYQka9W34s0XtEDk1JEXpynu2JXplYJteUIlu865mwop2tbPxH9W9XBKa3roE/LVCTHRnH/+yop1aAr9y0REVEosBQD22drU/M2/AnkHyldEdX2TKDjeUCr04Ho8gu/fLX+K3y+7nN1++lTnkavBpzeX1bunLnqOmHAAPYsJQpBFptj+h4rpYhCqzn51kN5mLP5EOZsPoxF246UWyGvca04nNI6TSWhpC9UvaTjI3bkI/tXadcNmZQiIiIKWgWZwJa/gU1/aavnFR5vVo642kD7s4EOI4GWg4DImEpfZuaumXhx8Yvq9p3d78TZLc/2x9YHnbw5Wj+phAFcdY8oFJU4iiUiDdgehpVSRB72hZq39TDmbDqsklF7swrLrZDXv3UdNR1PklFNU+M52uRvB9Zp1/W7+P1bExER0Qk4uhWtDk6B+cv3gV0LAZvl+GMJ9YAO5wAdzgOaDwDM1Veb78jegXFzxsEOOy5scyGu63wdD08FSg4eRNHmzUBEBBL69+c+Igrp1fciYDRMShFVodhiw7KdxzB3s5aEWrUnq1RfKGlO3rt5Kk5tUwcD2tThCnlGKO8/ulW7Xa9DoLeGiIiIqusPJcmnTVOBjVMRdWQzOrs+Xqcd0O5MrT9Uem/A5H7vzfySfNw9827kluSiW71ueKTvIxworGxfORqcx3bsiMjatfmeJQrlnlJmVkoRGX5K3rbDeZizSZuSJ6vl5ZeZkte+QRIGtK6DU9vWVQmpuGg2JzeMI1u0UVXpL5HcKNBbQ0RERGUVHAO2zNASUZunA4WZzofspkgcjm+L1H6jYe4wAkhtWeN47umFT2Pzsc1Ii03DK4NeUQ3OqWJ5C7SkVEK/vtxFRCGqxNFTKtLEpBSR4RT+f3v3ASdldS5+/JmZ7X2XZYEFlt6rUgRFsFAssUTNTblGLPHm+lc+KtFETaKS3EhiNPHGq/FqFGNiri22WBBEQAUUREGKIL0tsAtsr1Pe/+ecdwuLC+wuM/O+77y/r5/xzM7Mzh7O2XfnzDPPeV5/UAeflmwqkg82F8meIzUt7s9NSzCDUAM662yoLhnUhbKt4k1m23mQTkEHAAAWC4VE9q8xA1GqRtTeVSLGUR/4JeeIDJiuM6ICBZNl+Qcfy0XjLxJffMeDSK9tfU3e2v6W+Dw++f2U30teSl54/i0xSAXwqlas0NdTJk60ujsAIp4pZb/3SGzfgyvtLamWxZuLZfGmIlm+7ZDU+s2DVEnweWVcn2wdhFLb8oZ0zRCvDc9SgFYcbti6lzuQ4QEAwCqVRSLbPjADUdsWtTxbntJ5sHnGvEEXivQY17wtz+8/5R9dWFkoD656UF+fddoszrR3Ev69eyVw4IBIfLyknH76KY8/ALsXOveI3RCUgmsKu63eVSKLNxfpQNTXBytb3N8tM0nOHZwn5w7K0wXKUxI4NBypsZ5UB9P9AQBABwT9ZgaUyoRSl/1rW96vttX3mSzSf6pI//NFsgoilvVz3/L7pMpfJaM7j5Zrh10bkZ8TS6pXrtJt8vDh4k1Otro7ACIkoLJW2b4HRP9MeR+oLXmbiuTDLcVSUdt8BheV+DSmV3ZTIErVifKw3cv5jmw32079rO4JAACxrWSXyPbFZhBq+1KRuvKW93cb1RCEmmpmQ7XhbHmn6uWvX5ZP9n8iib5E+fVZvxZfOwqju1X1Z5/pNmXcOKu7AiCCgk01pciUAiK+LW/hxoPy3oYDsmpnSdPBp+SkJsg5AzvLOYPzZMqAzpKZQsHLmA1KkSkFAED4C5Tv+MgMRG1bLFKyo+X9qjaUyoJSQah+54mkRbeO06GaQ/KH1X/Q1289/Vbpndk7qj/fqapXr9ZtyrixVncFQFS271HoHAh7mrbaiqeCUAs2HpD1+1p+SjekW4ZMG5KnM6JG9sgSnw0jwwiT+mqRqmLzelYvhhUAgFMRqBfZu9IMQKlAVOEXIkZzDU7x+MwMKB2IOl+k2+jm2lAW+J8v/kdv2xvWaZj8+5B/t6wfThI4fFj8u3frk8Mkjx5tdXcARBCZUkCYD6gvdpfIgoaMqF2Hq5vuUzGnsb1zZMawrjJ9aBfpmZPC2LtF2Z7muhXJWVb3BgAAZzEMkaKNItuXmIGoXctE/M1rrKYTifQ9V6TfuSK9zhJJyhA72Hxksz7jnvLTcT8Vr8d+mQB2VLPWrP2V2L+f+NLTre4OgAgKNOwgsmOSBtWc4ZhA1Kc7Dsu/1u7X2/MOVdY13ZcQ55Wz++fqQNT5Q/KkU1qipX2FRUobglIRKp4KAEDMKdtr1oPasdQMRlUebHl/ameRvueYgSjVZnYXO2bNP/TZQxIyQjK913Q5vQtnkGurmi/W6DZp1KgIzhAAOwg11pTi7HtA+xYZn+8ulX+tLZS31+2X4ormQFR6UpycPzhPpg/rKlMGdpbUROKrrle22xyCzB6uHwoAAFpVfURkx4cNQailzWetbRSXLNLrTDMTSgWi8oaKeO2ddfR50ee6uHm8N15uH3O71d1xlJp163SbTFAKcM3Z97w2PLkX7+Rhu0DUxv3lOiNKBaP2ldY03ZeZHC8XDu8qF43oJhP6dtIZUkCT8v1mm5HPoAAAoNRXiexaIbJjiRmEOqCCEM0ngRG1zS3/NJE+U8xMqJ5niMQnOWrsnl73tG4v73+59Ejng6n2rLlrN27U15OHD4/Y/ACwh2BDSUC27wHHsbWoUgeh/vVloWwvrmq6PTXBJ9OGdpFLR+fLpP6dCUTh+CoKzTadoBQAwKWCfpF9q80AlNqOt3eVSMjf8jGdBzcEoaaYdaEcXIdR1ZL6aN9HuobUtcOutbo7juLfu1dC5eXiiY+XxH79rO4OgAgLqbqBKihFphTQrKzaL2+s3ScvfbanxVnzVAaU2pp3yah8OXdQniQnWHcmFzgxU6qb1T0BACCKxcm/MgNQ6qKKk9dXtnxMZs/mIFSfySLpXWNmduZtmKdbVUuqIIOaku1Ru8HMkkocOFA8CQkRmR8A9gtKeey3e4/te4h+gbXl2w7JS6v2yLvrD0hdwMwjjPN65OwBuToQpTKj0pPimRq0T2WR2abFzmIbAIBvKNvXHIRSl6qG179GyTlm8EkHoaaI5PS157uQU1RWVyYLdi7Q168dTpZUe9Vu3qTbxCGDwz43AOwn1LBzm5pScK0D5bWyYK9HHnrkY9lT0lwnanDXdPm3sT3l8tO6S04qn9LgFFQVm21aZ4YRABA7akpFdn7cHIQ6vKX14uT6LHlTRLqMsH1x8nB4Z8c74g/5ZXDOYBnWaZjV3XGc+q1bdZs0YIDVXQEQxbPvUVMKruIPhmTRV0V6e96SzUUSMtQ2vBpJS4zTGVHfG9dTRvbIFE8MfnqHKFNnk6g+1Hz6agAAnCpQL7Ln0+YgVOHnIkZDhdqm4uSnNwShVHHy8SJxieI2r299XbeX9bvM6q44Ut0WMyiVSFAKcNX2Pa/Xfu+9bX32vblz58qrr74qmzZtkuTkZDnzzDPld7/7nQwaNKjpMbW1tfKTn/xEXnjhBamrq5MZM2bI448/Ll26dLG0725WXuuXF1fukXnLdkhhWW3T7f3SDfmPacPlktE9JCXB1r96cJraUpFQwLyekmt1bwDAtlhb2VTJLpGt75uXHR9+sy5UpwHNQajekxxdnDwctpVuk42HN0qcN04u7nux1d1xnFBdndTv3q2vJ/Tvb3V3AESzppTYj60jA0uXLpWbb75Zxo0bJ4FAQO655x6ZPn26bNy4UVJTU/Vjbr/9dnn77bfl5ZdflszMTLnlllvkiiuukGXLllndfdfZW1It85btlBdX7ZHKOjNA0Ck1Qa4a20OuGNVNNq1aKhed1l3i4239awcnqikx24R0kTi2gQLA8bC2sgl/jVmUfEtDIOrYLXkq67fvuc1b8jJ7WNVTW1q8Z7FuJ3abKNlJ2VZ3x3H8e/boLHNvWprEdSbDHHCDEDWlOmb+/Pktvn722WclLy9PVq9eLZMnT5aysjJ5+umn5R//+Iecd955+jHz5s2TIUOGyCeffCITJkwIw/ThZNbsKZW/fLRdFy4PNvy2D8hLkx+d3UcuG91dkuJ94vf7xSynCESo3oaSlMnwAsAJsLayiPqE+vA2ka0LzSCUqhEVaM4mF49PpOcZIgOmivSf6pq6UB314d4PdTulxxSru+JITVlSBQWU0QDcwjAbO1bOcVTKigpCKTk5ObpVwSkV7Jg6dWrTYwYPHiwFBQWyYsUKglIRZBiGLN5cJH9esk1W7WzIUhGRSf1z5Yaz+8g5AzvzIofoqW34HXT5dgYAaC/WVhGuDbVjqcjmd81AVOmulvdndBfpf75I/2lmNhQfrLRJaW2prC1eq69P7jE5AhMX++p3mUGp+F4FVncFQJQYDVEpG5aUck5QKhQKyW233SZnnXWWDB8+XN924MABSUhIkKyslm9EVT0pdd/xqNpT6tKovLxctyrApS44sS92l8qDC76Wz3aZ2SnxPo98a0RXue7M3jKkW7q+TW23PFrjuDK+iARP1RH9xyyUmC7BGD+GOZaAyBxLbnx9iuW1lWXzGqgVz/bF4t30L/F8PV88deY4KIY3XoyCiWL0O09C/aaK5A5q+ZG1C38HOzJvy/Yuk5ARkv5Z/SU3MdeVx+6pqtu5U7dx3buHbfzc/LfU6Zg7d8xbqGFHUzAYjNpx2taf45iglKottX79evn444/DUuRzzpw537h9wYIFkpKScsrPH6sO1oi8tdsrXx4x08njPYac3c2QKV1DkpW4R3Z8oS4nfo6FCxdGp7NwlV6Hlsto9TtaWisr33lH3IBjCQjvsVRdXe26IXXD2ioafyt9wTrJq/hS8ktWSdfyNRIXat6WVxuXKfuzxsjBjFFyKG2IBH1JIkdE5Mh2EVEXtHfe3q15V7edqjvJOy55zQ+3/C/XSpqIbDp8RMrCPIasT5yLuYvteTtS4tNlzlev/lwCOxv28kVYW9dWjghKqeLlb731lnz44YfSo0dzoceuXbtKfX29lJaWtvhE7+DBg/q+47n77rtl9uzZLT7N69mzpy6inpGREcF/iTMdLK+VRxdvl1e+3KdrRqmUvytP7y6zzu0n3TKT2hwlVQfMtGnTJD4+PuJ9hrt4P90pskekS8++ctFFF0ks41gCInMsNWb2uEWsr60i/reyrkI8WxeaGVFb3xdPoKbpLiM9X0KDLxFj8LfE12O89PD6hDLl4Zu3fy78p0ixyIWnXygX9Y3t1/xI2T3vWakXkdFTp0rq5LPD8pysT5yLuXPHvP2tcKXsqCiV004/XS4Y1iUqfWzr2irO7nWLZs2aJa+99posWbJE+vTp0+L+MWPG6AlYtGiRXHnllfq2zZs3y+7du2XixInHfd7ExER9OZZ6LgImzfzBkK4Z9fiSrVLrD+nbpg7pIj+7YJAM6GJu02svxhgR4Tej8N7kTPG6JOjJsQSE91hyy+u/29ZWYf35oaDItg9EVj8rsmWhSLB5u6JkFYgMvUxkyGXi6T5GfBQpj8i8BUNB2VRinjpnVN4o1xy34RZs2Iqb1KN72MfQ6mMeHcfcxfa8eT3mbiefzxe1Y7StPyfO7mnl6sx6b7zxhqSnpzfVMsjMzJTk5GTd3nDDDfqTOVX8XH0SpxZaatHEmfdOzaYD5fKTl9bKhkIzujmmV7bcdeFgGdfbLDIP2Ep9pdkmpFrdEwCwNdZWHVBZLPLF30RWzxMpNQtEazn9zEDU0EtFuo225ymNYsyeij1SE6iRJF+S9MlsGVBF2xj19RIsNevCxuXlMWyAS3gaXqJC6mywNmProNSf//xn3Z5zzjktbp83b55ce+21+vof//hH8Xq9+tM8VWBzxowZ8vjjj1vS31gQCIbkyY+2yyMLt0h9MCRZKfEy59JhcumofM6mB/vyN2ybiKcmHACcCGurNlKL9l3LRD57RmTjmyKhhmKt6gx5o//dvHQZRiAqynZXmEHBgowC8XlVfRS0V6Ck4YzFPp/4MjMZQMAlvA1RKRvGpOwdlFIp5ieTlJQkjz32mL7g1GwrrtTZUWv2mJ+eTB2SJw9cMULy0ttWNwqwTKChqGx8MpMAACfA2uokakpF1r5gBqMObW6+vftYkbHXiwz7tkgCH4BYZVf5Lt32yuhlWR+cLnhEVdoX8WVni4dtpoBreM3de2RKwb7eXFsod768VuoCIUlPjJP7Lh2mi5l7SEWHg2pKkSkFAOiQukqR5X8SWf7oUa8pqSIjv2MGo7qNYmBtgKDUqQs0BKXisrPD8GwAnMIjZErBxp+YPr5km/z+PfPTwEn9c+XBq0ZKfhYZJ3CQQEOx2bhvFtkFAOCExcu/+LvI4t+IVB40b8sbagaiRn5XJImzMttJYWWhbrundbe6K44VKivTre+os2sCiH0eakrBrmfX++Xr6+WFVXv01zdM6iP3XDREfF4KdcKhQSkfQSkAQBttfV9kwS9FijaaX2f3EZk2R2TIpdSKsqnimmLd5qVQoLujguUVuvVmEHAF3MRLTSnYTUWtX/7f85/LR1sOiYpB3X/pMLlmYm+ruwV0TLDebOMSGEEAwIkVfSXy3s9Fti0yv07KEpnyU5FxN/I6YnNF1UW67ZLSxequOFawwjyzti893equAIgiL5lSsJNaf1CunbdKVu8qkeR4n/zPD06T84fw4g4HI1MKANAWX74k8sYtIsE6EW+8yPj/EJl8h0hKDuNnc/6gX47UmvWQcpNzre6OY4UqKnXrJSgFuDJTKmTD0+/Z+ux7CL9gyJDbXlijA1IZSXHy/I8myIgenA4WMZIp5SNTCgDQilDIrBv10UPm1/2niVz0oEhOX4bLIcrqy5qK9WYlUg+po0JVVbr1pqWGbW4AOKemlGG/mBRBKbcVNf/1Wxtl/oYDkuDzypPXjCUghdgQCpitjzg7AOAY9VUir/1Y5Kt/mV9Pul3kvHubz48NRyitLdVtRmKG+Lw+q7vjWKFq8+yS3pQUq7sCIKo8+v82jEkRlHKTpz/eIc8u36mvP/xvo2RC305WdwkIDzKlAACtUWfUe+kHIvvXmtm0l/xJZPT3GSsHKq0zg1KZCWT4nwqCUoA7eakpBat9vrtEHnjnK3395xcNkUtG5VvdJSB8gn6zVfVBAABQnwkbQfG9eoMZkErJFfne8yIFExgbhyqvNwt0ZyYSlApLUCqZTCnATTxs34OVquoCMvvFNRIyRC4bnS8/OrsPE4LYEgqaLdv3AAANBh14XbwHPhFJSBe5/j2R3P6MjYNV+c1aSOlqPtFhRk2Nbr0pyYwi4CKehu17dsRmehf4zTtfyc7D1dItM0l+ddlw8TSGSYFYEWrMlKKmFABAxLPjQxl44E1zKC55hIBUDKj0m2eNS42nQPepCNXW6taTlBSWeQHgLIbYD0GpGLdkc5H849Pd+vrD3xklmclsb0IMYvseAKBR9RHxvfGf4hFDQqOvFhlxFWMTQ5lSafFpVnfF0UK1DZlSSWRKAW7isXFeCkGpGFYXCMp9b27Q1687q7ec2T/X6i4BkT37HplSAIBVfxFPVZFUJHaT4PQHGI8YUVlPplQ4GDVmppQ3KTEszwfAYQz75UoRlIphf12+U3YdrpbO6Ynyk+mDrO4OEDnUlAIAKIE6UUEpZXPXy0TiKeYcK6oDZoHuFOb0lITq6nTrSSZTCnAlj/1SpghKxahDlXXy6KKt+vqdMwZJWiK1dhDDqCkFAFA2vCZSeVCMtK5SmDWeMYkh1f6GoFQcgcawFDpPJFMKcBPDfglSTQhKxaj/fn+LVNQFZHj3DLnq9B5WdweIzvY9j4+RBgA3+/R/dRMa+yMx2NIdk5lSyXFk+IQlU4pC54CrGA0lzu2XJ0VQKiYVltbIC6vM4ua/uHioeL12/NUDwhj2bwxK+SjkDwCuVVkkUvi5vhoa9QOre4MwY/veqTP8fpGAuWbysn0PcGWmlJfte4iG/126TfxBQ87okyMT+nZi0BHbjFDzdT4VBwD32vGh2XYZIZKWZ3VvEGY1fnPbGdv3Tj1LSiFTCnCXUFNQSmyH7Xsxpqi8Vv5v1R59/dbzB1jdHSDygv7m6wSlAMC9diw1275TrO4JIqAmYAal2L536vWkVKFjT0JCeCYGgCMYDalSZEoh4p7+eIfUB0Iyple2TOxHlhRcoHHrnkJQCgDca+cys+0z2eqeIAIISp26UENQSp15z2PDLTwAIifYuH/Phoc+mVIxpLo+IP+30qwlddOUfrzYwF1n3lN8fOoHAK5UXyVyZJt5vfsYq3uDCJ59LzU+lfE9xaCUN4UzGAJuE2zYvxfvs19UiqBUDPnn5/ukvDYgvTulyHmDqaUAN27f4+x7AOBKRZvMNjVPJDXX6t4gAqoCVbolKNVxoSozsEdQCnCfQNAMSvm89gsB2a9H6JBQyJB5y3bo69ee2Zsz7sE9gvXNWVKkogOAOxVtNNu8IVb3BBGqhdKYKZUST5ZPR4WqCUoBbs+UirNhpXOCUjFixfbDsr24StIT4+SqsT2t7g4Q/aCUN55RBwC3OrTZbAlKxWw9KUPMN1Scfa/jQtVmthmZUoD7+EPmGct9BKUQKS82nHHvstPyJS0xjoGGewQaglJx1JMCANcq2Wm22X2s7gkioKK+Qrc+j4+z752CUEWlbr3paeGZGACOUR8wg1KJcfbLS7Jfj9BuZdV+mb/hgL7+3bEFjCDcJVhntr5Eq3sCALBKyS6zze7FHMSgSn9lUz0pzhrXcaHKhuBeWnqYZgaA04JSCQSlEAlvrN2nf8mGdMuQ4d0zGGS4C5lSAIDSxqBUb8YihoNS6QkEU05FsDFTKo1MKcBt6oNkSiGCVu8q0e23Rnbj0yO4T8jfXOgcAOBOnoazr9aWW90TREAwFGzavodTGMfyMnMcM/gQG3CbkGHW5bNjtinb92JAYWmNbgtyOBsJXCgUMFsvtdQAwLX6nWu2Wxda3RNEQOMZ91TBc3RcsLRUt76sLIYRcJk4r7fFWfjshKBUDCgsrdVtflay1V0Boo+gFACg/zRzDLYsYCxiUOMZ96r85tnj0DHBEoJSgFv5Gs66FwgSlEKYhUKGHCxvDEolMb5wn4bTm4qHGDsAuFb/qWa7f61IhXnyF8ReplR1oFpCRsPrPjqeKZVNphTgNnGNQanG9042wrs4h6uqD0igIQUvO4WaOnAhw6wzwfY9AHCxtM4i3Uab17cttro3CLPkuObdAPXBesa3gwKHD+k2rlMnxhBwmThfY1CKTCmEWU29+YZcBT4TbXh6RyDiGoqfkikFAC7X/3yz3bbI6p4gzBJ9iU3XawPmDgG0j2EYEihuCErl5TF8gFtrSgUJSiHMqhqCUqkJcbaspA9EXGMaP9v3AMDd+jUGpT5ofm1ATIjzxkm8N15frw0SlOrw1j2/ecZiMqUA94lj+x4ipdZvBqUS4zlFLtwelCIoCwCu1nO8SEK6SPVhkQPrrO4Nwiwpzqydyhn4OiZw8KBufTk54kmg5Afg2kLnITKlEKFfrpBhv18uICrIlAIAKL54kV4T9VXv7uWMSYxpzJTyh8xsH7SPf+9ecxy7d2foANgKRYgcLsFnTmF9gDR1uB2ZUgDger0nma8Iu5a5fihicQufEggFrO6KI9Xv3qPbhJ49re4KAAsYDa3Hhu+ZCEo5XEJDcXOCUnAvsgQBAA16NQSl9qygrlSMZkoRlOoY/14zKBVPUApw9Vsmj/1iUgSlnM7b8FsVCJEpBZdi+x4AoFG3USLxqeKpLZP02n2MSwzxecz6qSGK2HdI/c6duk0oIFMKcKNgQ7mfhuo/tkKmVIwUOk+m0DncKlBvtnHNp4sGALiUL04kf7S+mlVtvglHbAga5pqXs013TO3mr3WbOHBgGGcFgFPUN5T7SYyz3wnSCEo5XE1jUCrB3GcPuE6wzmwJSgEAlPzTdJNVvYPxiCGNGVKNGVNou8ChQxI8fFjv20ns35+hA1yoLmDGDRIbyv/Yif16hA4GpZhKuFRjppSP0xsDAJqDUtnV2xmOGAxKeT2sedurdtNm3Sb06iXe5OSwzw0A+6trzJSKt9/fUPv1CB1Kw2s8Cx/gOkGCUgCAo3QdoZv02kKRhhoacL76htf7xoLnaLuatWt0mzR0KMMGuFRVnZnMkmTDsj9EMhwuGDIXW3FephIuD0rFkSkFAFApUr3FEI/EhWpFqooZkhhR6a/UbVp8mtVdcZya1at1mzx2jNVdAWCR8lq/bjOT7RfYJ5LhcIGGoJTPjmX0gWgImn9g2b4HAGiqMZjZQ1/1lFBXKlaypPwh8/U+NSHV6u44iuH3S/UXZqZUytixVncHgEUnR6tv2GFFUAphFwg2FH0kKAW3FzonnR8A0MDI7mNeISgVU1lSSkpciqV9cZqa9evFqKkRb2YmRc4BlyqvMYP6KmSQasMTpJEp5XD7y2p1m5eeaHVXAGvUV5stn5wCABqlddGNp/oQYxIDDtWY85iZmClxXvu9obKzysVLdJs6caJ4KPcBuFJxpfkhfk5qonhtmMxCUMrhdh6q0m3vXFKZ4VJ1FWabSI0JAIDJSMoyr9SUMSQxoKi6SLddUsxgI9quYtEi3aZPncqwAS5VVF5n60QWglIOt/NwQ1CqE6nMcKn6xqBUhtU9AQDYRVKm2daWWN0ThDEolZeSx3i2Q92OHVK/bZtIXJykTT6bsQNcqqjC3F3VJYOgFMLMMAzZWFiur/fJJUsELlVrHgOSmG51TwAAdpFkflDhacymhaPtr9qvWzKl2qf83Xd1mzp+nPgy+PAOcKv9TSV/ksSOyJRysPX7yqWwrFaS430yple21d0BrFHVUC8kJZcZAACYQgGz9dnv1Ndov60lW3XbN7Mvw9dGRjAopS+/oq9nXn454wa42O7DZg3eApvuriIo5WDvrjc/NTpnUGdJTvBZ3R3AGlVmSr+kdWYGAACmgFk/w/DZc6sC2ufrkq91OyB7AEPXRlUffyyB/fvFl5kp6TNmMG6Ai+1sKvljzzrUBKUcvHVv/voD+voFw7ta3R3AGqFQc6ZUKnUmAAANAvVmG2fPrQpou2p/teyp2KOvE5Rqu5IXX2rKkvImEpwF3GxXQ6ZULzKlEE7vrDsg2w9V6a175w3mzThcqqpY5aeLeLwiqWzfAwA0qDlitpyZ1fE2HN4ghhiSm5yrLzi52s2bpfKDD/T1rO/+G0MGuNiRqno5XGV+UNM7l0wphEldICi/nf+Vvv7jKX0lPYl6CXCp0t1mm55P3RAAQBPPEbMGkZHTj1FxuBWFK3Q7vut4q7viGMV/elS3GRddJIl9qcMFuNmGwjLd9u6UImmJcWJHbN9zoL8u3yl7jtToUzr+x2ReaOBipbvMNruX1T0BANiI59AW80qn/lZ3Bafo0/2f6nZi/kTGsg1q1q2TykWLRLxeyb3lFsYMcLkNheaZyofm2/cMnASlHGZ7caU8usj89O+O6YMkJcGe0U4gqkGprAIGHABgqikVT8NJMIxOFMZ2srK6Mll/eL2+PqHbBKu744ias0UPPayvZ156qST27WN1lwDYJCg1LD9T7IqglIMUldfKNc+slIq6gJxWkCVXnN7D6i4B1jq8zWyzWXQBABpsfV83lYldRRLTGRYHe2v7WxIyQjI4Z7B0TeXEPidT+uKLUv3pp+JJSpLcW26OyhwBsHeg+rOdZo3FUT2yxK4ISjlEea1fZs5bJXtLavR+0KeuGSs+r8fqbgHWKt5stp0HMhMAANNXb+qmMGssI+LwN1OvfP2Kvn7FgCus7o7t1e/dKwcf/L2+njf7dknowYfXgNvtPlIt+8tqJd7nkTG9ssWuCEo5QK0/KD9+brV8tb9cctMS5bnrz9At4GqGIdJYMySXoBQAQL0zrxbZslAPRWHWOIbEwdYWr5WtpVslyZckF/e92Oru2JoRCsn+e34uRnW1pIwdK9lXX211lwDYwCfbD+t2dM8sSU7wiV1RkMjmVLrdz/75pWwrrtLV8p+9bpwUdEqxuluA9Sr2i9SViXi8IpxdCQCgrHlexF8tRmaBlCX3ZkwcnCX1l3V/0den954uGQn2LdBrh7E6OPe3Ur1ypXhSUqTb3AfE4yXvAIDIh18f0sNwRp9Oth4OglI2VVHrlwfnb5a/fWIWclaZUY9+/zQZ3t2+BcqAqNq7ymzzhonEJzH4AOB2ZftE3p+jr4bO+E+RYsocONWC3Qtk6d6lEueNk+uGXWd1d2xe2PwhKfnb3/TX3e6/TxJ69rS6WwBsoKzaLwu/Oqivzxhm75p8BKVsaNFXB+UXr6/X+z+V747tKfdcNEQyU+Kt7hpgH3tWmm1PtmcAgOupLd1v3S5SXyHSY7yExtwgMv891w+LE1WFquSJz57Q128ccaP0z+5vdZdsq/hPf5IjTz+jr3e9/359xj0AUN5cu0/qAyEZ0i1Dhne3d7YpQSmb8AdDsmLbYXlh1W55Z90BfVtBTor89ooRcmb/XKu7B9g4KHWG1T0BAFjti7+JbHlPxJcgcumjIl771s7A8QVDQXmr5i0p8ZdI/6z+OiiFbzLq66X4fx6Tw08+qb/ucs89kv297zJUAJqyKF/6bK++/p0xPcTjsXfmcMxsOH7sscekd+/ekpSUJGeccYasXNnwhtXGauqD8t6GAzL7xTUy5tcL5ZpnVuqAlDqp3o8n95X3bptMQAo43haNws/N6wSlAMC9a6v6KpF/3Sby5izz68l3iuQNtrpX6ICS2hKZtWSWrPOvE6/HK78681cS72OXwLEqly2T7Zd/uykglXfnHZJzzQ/5nQPQ5B8rd8u6fWWS4PPK5ad1F7uLiUypF198UWbPni1PPPGEXjQ98sgjMmPGDNm8ebPk5eWJnZTV+GXxpiKZv/6ALPm6SGr9oab7ctMSZNrQrnL1hAIZlk/tKOC4Vv6vSCgg0muSSE4fBgoA3Li22rda5J83ihzZZn498RaRSbOt7hU6YF3xOpm9dLYcqDog8RIvcybOkRGdRzCWR6nfu0+KfvdbqVj4vv7al5MjeT+9U7Iuv5xxAtBk/b4ymfPmRn39J9MHSk5qgthdTASl/vCHP8iNN94o111nFkJUC6i3335bnnnmGbnrrrus7p4UV9TJwo0HZf6GA7Ji2yHxB42m+7pnJcsFw7vqy+kF2eJTaVIAjq+uUuSzZ83rZ97CSAGA29ZWwYDIx38UWfpb8wOK9HyRb/9ZpO851vYLHdtisvkl+e2q30ogFJCC9AK5VC6VC3pfwGg2CNXWyuG/PC2Hn3pKjLo6EZ9Psv/9B9L5llvEl2HvOjEAop8Ac9Pzq6U+GJKpQ7rIf0zu64gpcHxQqr6+XlavXi133313021er1emTp0qK1asaPV76urq9KVReXm5bv1+v76Ey+HKOrnlhbWyeneprr/ZqH/nVJk+tItMH5onQ7ulN+3xDAUDEgpKTGoc13COL9zJu/o58dWViZHTVwJ9zlO/VOImHEtAZI4lXp+csbaSsr3ie/3H4t37qf4yNOQyCV74kEhydovXA+bV/moCNfKblb+Rd3a+o78+r+d58vMxP5cVS1ZwPDao/uQTKbr/fgnsK9RfJ48bJ7l33yWJAwaI2msRstEaiGPOuZi72Jg3wzDkjpfWyp4jNdIjK0l+++2hEggEbNHHmA9KHTp0SILBoHTp0qXF7errTZs2tfo9c+fOlTlzzFMGH23BggWSkpIStr6FDJGvC31iGB4pSDVkZKeQjMwxpEtymUh9mexa87XsWiOusnDhQqu7AIebuPXvojaOfJlytux8d764FccSEN5jqbq6miF1wNoqqf6InLt/vXi8SfJlz5myN/FMkcWtB8oU/lbaV71RL6srVotXvDIjaYacWX6mDkgpzJspeft26bmvUPyZmVJ88cVSOXKEyJYt5sWmmDvnYu6cPW8ldSIrt/rE5xH5Xs9KWbbY+vfdbV1bOT4o1RHqkz9VJ+HoT/N69uwp06dPl4wwp8HmDjkiBTnJkp+VLG6moqTqgJk2bZrEx1O0EqcgNEMCX78rQ/ueK0MTUl03lBxLQGSOpcbMHth/beXZUSBGdm8ZmVUgI4/zGP5WOsPp5adLcW2xjMkbo79m3r6pon9/SZ08WYaEMbgbCcydczF3sTNvF0yrkzV7y+T8wfao/djWtZXjg1K5ubni8/nk4MGDLW5XX3ft2rXV70lMTNSXY6nJDHfA5OxBLT9ldLtIjDHcJl5kxLfF7TiWgPAeS7w2OWdtJQPPb/NDmVt769epn6j/jsW8Ncu55BJxEubOuZg7589b1+x4uSA7Teyira//XnG4hIQEGTNmjCxatKjptlAopL+eOHGipX0DAABwGtZWAAAgWhyfKaWodPGZM2fK2LFjZfz48fq0xVVVVU1njAEAAABrKwAAYC8xEZT67ne/K8XFxXLvvffKgQMHZPTo0TJ//vxvFOgEAAAAaysAAGAPMRGUUm655RZ9AQAAAGsrAABgf46vKQUAAAAAAADnISgFAAAAAACAqCMoBQAAAAAAgKgjKAUAAAAAAICoIygFAAAAAACAqCMoBQAAAAAAgKgjKAUAAAAAAICoIygFAAAAAACAqCMoBQAAAAAAgKgjKAUAAAAAAICoIygFAAAAAACAqCMoBQAAAAAAgKgjKAUAAAAAAICoIygFAAAAAACAqCMoBQAAAAAAgKiLi/6PtB/DMHRbXl5udVdilt/vl+rqaj3G8fHxVncHcCyOJSAyx1LjGqBxTQBnr634W+lMzJtzMXfOxdw5k98B76/burYiKCUiFRUVejB69uwZjbkBAAA2XhNkZmZa3Q3HY20FAADasrbyGHwkKKFQSAoLCyU9PV08Hk/T4IwbN05WrVol4RKO5+voc7T3+9rz+LY8VkVJVdBvz549kpGRIW4T7t8lp/WJY4ljya6/S+HAsRQbr0tqOaQWTfn5+eL1Ut0gUmurts7R8e5v6+12WXfY4fUv3MfOiR7T3vvsOm+t9c2K57LD3HHMRXbuTnXe2jNHdj/u+HsZe38v27q2IlNKFdbyeqVHjx7fGByfzxfWCQ7H83X0Odr7fe15fHseqx5n9UFjhXD/LjmtTxxLHEt2/V0KB46l2HldIkMq8murts7R8e5v7+1Wrzvs8PoX7mPnRI9p7312nbdwz1001u+RmjuOucjO3anOW0fmyK7HHX8vY/PvZVvWVnwUeAI333yz7Z6vo8/R3u9rz+PDPU6xyI5jFM0+cSxZM06xyI5jxLHE6xLCf9wc7/723m41O7z+hXtNd6LHtPc+u85buPsWjfV7pOaOYy6yc3eq83ai+5k7/l465e8l2/cQFSq9UEVJy8rKLI/kAk7GsQRwLIG/lbGK1zjnYu6ci7lzpvIYen9NphSiIjExUe677z7dAuBYAqzG61JsY36diXlzLubOuZg7Z0qMoffXZEoBAAAAAAAg6siUAgAAAAAAQNQRlAIAAAAAAEDUEZQCAAAAAABA1BGUAgAAAAAAQNQRlIItVVdXS69eveSOO+6wuiuAI5WWlsrYsWNl9OjRMnz4cHnqqaes7hLgSHv27JFzzjlHhg4dKiNHjpSXX37Z6i4hzHbs2CHnnnuunuMRI0ZIVVUVY+wQvXv31seleq1TcwjnYK3vPKwtnWuPzdcynH0PtvTzn/9ctm7dKj179pSHHnrI6u4AjhMMBqWurk5SUlL0GywVmPrss8+kU6dOVncNcJT9+/fLwYMH9ZveAwcOyJgxY+Trr7+W1NRUq7uGMJkyZYr813/9l5x99tly5MgRycjIkLi4OMbXIUGp9evXS1pamtVdQTux1nce1pbOtd/maxkypWA7W7ZskU2bNsmFF15odVcAx/L5fDogpajglGEY+gKgfbp166YXcUrXrl0lNzdXBy4QGzZs2CDx8fE6IKXk5OQQkAIijLW+M7G2dK5uNl/LEJRCu3z44YdyySWXSH5+vng8Hnn99de/8ZjHHntMf3KVlJQkZ5xxhqxcubJdP0Nt2Zs7dy4zg5gWjWNJpVmPGjVKevToIXfeead+AQJiTTSOpUarV6/WnxSrLF7ExvyqN8cqy0b9jNNPP10eeOCBMP8L3Csax6Z6XpXpNm7cOHn++efD2Hv3Yq3vXKwtnetDl69lCEqhXdQ2IPUmVx0UrXnxxRdl9uzZct9998nnn3+uHztjxgwpKipqekxjjZtjL4WFhfLGG2/IwIED9QWIZZE+lpSsrCxZu3atrpfyj3/8Q6ftArEmGseSoj5RvOaaa+TJJ5+Myr8L0ZnfQCAgH330kTz++OOyYsUKWbhwob7AGcfmxx9/rN9gvfnmmzqg+OWXXzJ1Np831vqRw9rSuarcvpYxgA5Svz6vvfZai9vGjx9v3HzzzU1fB4NBIz8/35g7d26bnvOuu+4yevToYfTq1cvo1KmTkZGRYcyZM4c5QkyLxLF0rJtuusl4+eWXT7mvgBuPpdraWuPss882nnvuubD2F9bP7/Lly43p06c3ff3ggw/qC5z3OnfHHXcY8+bNO+W+ohlrfedibelc4sK1DJlSCJv6+nr9adXUqVObbvN6vfpr9eljW6hte+rsADt37tQFzm+88Ua59957mSW4SjiOJZUVVVFRoa+XlZXptOBBgwZFrM9ArB5Lan147bXXynnnnSc//OEPI9hbWDG/atuX+qS5pKREQqGQ/ls5ZMgQJsMBc6cyCxpf5yorK+WDDz6QYcOGRazPYK3vZKwtnaveBWsZglIIm0OHDun9qV26dGlxu/paVfkHEL1jadeuXbpwr0rvVe2sWbP0qc4BNwnHsbRs2TKdNq/qO6jUeHVZt25dhHqMaM+vOsue2vY1efJkfZrsAQMGyLe+9S0mwgFzpz58mTRpkn6dmzBhgt6SooKMiBzW+s7F2tK5DrlgLcP5bmFbKpoLoGPGjx8va9asYfiAU6Te9KoMGsQudbZfzvjrPH379tV1E+FcrPWdhbWlc02y+VqGTCmEjTqzlzpV6LHFlNXX6tSTADiWgGjidSm2Mb/Oxdw5E/PmXMydc+W64D02QSmETUJCgowZM0YWLVrUdJuKyKqvJ06cyEgDHEtAVPG6FNuYX+di7pyJeXMu5s65ElzwHpvte2gXVUhy69atTV+rU82rLUI5OTlSUFCgT1U5c+ZMGTt2rE7xfOSRR3Qhyuuuu46RBjiWgLDjdSm2Mb/Oxdw5E/PmXMydc1W6/T221af/g7MsXrxYn6by2MvMmTObHvPoo48aBQUFRkJCgj595SeffGJpnwE74lgCOJbA38pYxuucMzFvzsXcOddil7/H9qj/WR0YAwAAAAAAgLtQUwoAAAAAAABRR1AKAAAAAAAAUUdQCgAAAAAAAFFHUAoAAAAAAABRR1AKAAAAAAAAUUdQCgAAAAAAAFFHUAoAAAAAAABRR1AKAAAAAAAAUUdQCgAAAAAAAFFHUAoAAAAAAABRR1AKAGCpb3/725KdnS1XXXUVMwEAAMDaCi5CUAoAYKlbb71VnnvuOWYBAACAtRVchqAUYCPnnHOO3HbbbY7/2Vb+OxB5hw8flry8PNm5c2fYfl/S09Nbve973/uePPzww2H5OQAA92FtBSdgbQU3IygFRNG1114rHo9HXxISEqR///7yq1/9SgKBQNh+hh0CQq+++qr8+te/Dss4xcfHS58+feSnP/2p1NbWWjYWBw4ckFmzZknfvn0lMTFRevbsKZdccoksWrRIoilS89ue5/3Nb34jl112mfTu3Vsi7Re/+IX+eWVlZRH/WQAA52Ft1f5xYm3VEmsrwFpxFv98wHUuuOACmTdvntTV1ck777wjN998s14c3H333RIrcnJywjZOfr9fVq9eLTNnztQLqd/97ncSbSoj6KyzzpKsrCz5/e9/LyNGjND9eu+99/T8bdq0Sdyiurpann76af1vb6vRo0e3GnhdsGCB5Ofnn/B7hw8fLv369ZO///3veqwBADgWa6u2YW1lT6yt4HoGgKiZOXOmcdlll7W4bdq0acaECRP09SlTphizZs0y7rzzTiM7O9vo0qWLcd9997V4fG1trX5M586djcTEROOss84yVq5c2fT8ItLismPHjpN+X+PPvvXWW/X1v/71r0ZOTo7+nqOpvl999dX6+ssvv2wMHz7cSEpK0o89//zzjcrKym8818ke29ZxuuKKK4zTTjutxW3vvvuu/ndkZmbq57344ouNrVu3nnAsgsGg8cADDxi9e/fW/Rk5cqTu34lceOGFRvfu3Vvtc0lJSZvm5uhxPtEcn2isjvdvOtE4tOXnnuj35liqf+rf19rt7ZnjYy1evNi48sorW71vzpw5xqRJk9r8XAAA92BtxdqKtVXrWFvBKdi+B1gsOTlZ6uvrm77+61//KqmpqfLpp5/Kgw8+qLf3LVy4sOl+tY3tn//8p37c559/rrcAzpgxQ44cOSL//d//LRMnTpQbb7xR9u/fry9qm9nJvu9Y3/nOdyQYDMqbb77ZdFtRUZG8/fbbcv311+vn/f73v6+vf/XVV7JkyRK54oorVJD7G8/Vnscez/r162X58uV6y+PRqqqqZPbs2fLZZ5/pbXRer1efyS0UCh13LObOnauLaj/xxBOyYcMGuf322+Xqq6+WpUuXtvqz1fjMnz9fZ+moeTmWyp5qy9wc7XhzfLKxOt6/6UTj0Jafe6Lfm2N99NFHMmbMmLDP8YmMHz9eVq5cqbMLAQA4GdZWJ8fairUVayvYhtVRMcCtn+aFQiFj4cKFOqPmjjvuaPqk59iMkHHjxhk/+9nP9HWVeRIfH288//zzTffX19cb+fn5xoMPPthqllJHv++mm27SGUKNHn74YaNv376636tXr9bZNDt37mz133n0c53ssccbJ5/PZ6SmpurxUd/v9XqNV1555YTfV1xcrB+7bt26Vv9NKpMpJSXFWL58eYvvu+GGG4zvf//7rT7np59+qp/z1VdfPeHPbssYn2yO2zJWrc3vycbhZD+3rc+rqN/f66+/vsVtHZnjo6msqtzcXCM5OVlnpB07P2vXrj2l5wcAxC7WVm0fJ9ZWrWNtBViLmlJAlL311luSlpamaxKpTJYf/OAHcv/99zfdP3LkyBaP79atm85SUrZt26a/T9U3aqTqUalMEpWhcjwd+T6VNTNu3DjZt2+fdO/eXZ599tmmIpmjRo2S888/X9dWUplA06dPl6uuukqys7O/8TzteezRzj33XPnzn/+ss4D++Mc/SlxcnFx55ZUtHrNlyxa59957debPoUOHmjKDdu/erWsRHWvr1q163/60adNa3K4y1U477bRW+9HWbJ/2jPHx5rijY9XWcTjR71Zb1dTUSFJSUovbOtrvRu+///5JP/FW1NwBAHAs1lasrVhbsbaCc7F9D4gyFWxZs2aNDiSoN/iNW6qODmQcTQWBjt2GFQ0qSKOCDWqrmyo0rra6qaCU4vP59Lavd999V4YOHSqPPvqoDBo0SHbs2PGN52nPY4+mxkRtf1N9eOaZZ3TARRXYPpo6+53aGvfUU0/p+9VFOXo75NEqKyt1q7YhqjlovGzcuFFeeeWVVr9nwIABeg7CWcz8eHPc0bFq6ziE43crNzdXSkpKWtzW0X63VeP2x86dO4fl+QAAsYW1FWsr1lbtw9oKdkJQCoiyxmBLQUGBzv5pD3UWMlVXadmyZU23qeycVatW6WCAou5X9aDa+32t+dGPfqQzpNRZ8KZOndqizpAKaKisoDlz5sgXX3yhn/+1115r9Xna89jWqBpJ99xzj/ziF7/QgTzl8OHDsnnzZn2bytIZMmTIN4Ilx46F+rcmJibqDCI1B0dfjldDSZ1JUGX/PPbYYzpr61ilpaWnNMbtHatj/01tGYe2aO335njBShXEa2+/T4Wqe9GjRw8dEAMA4FisrVhbnQhrK9ZWsDe27wEOW3TddNNNcuedd+pgiQpsqYLValvTDTfcoB/Tu3dvnSmzc+dOvU1QPa4t39catbXwjjvu0Bk4KmOqkXp+VVBbbdPKy8vTXxcXF+uAyLHa89gTUcXXVf9VcEj1SW0N69Spkzz55JN6G5oKNN11110tvqe1sVDfq4qbqwyhSZMmSVlZmQ4kZWRkyMyZM1v92epnqoCL2oqnioOrFPFAIKCzg9QWQ7U9r6Nj3N6xau3fdLJxaIvWnlcFA4+lAnR33323Dnw1bs8L1xwfjyqurp4bAIBwY23F2oq1FWAxi2taAYbbT1t8skKL6vHq+xrV1NQYs2bN0oWhVRHws846y1i5cmXT/Zs3bzYmTJigi0arQ3zHjh1t+r7jFXn84Q9/aOTk5Ogi4Y02btxozJgxw+jcubN+roEDBxqPPvpoq891sse2Z5zmzp2rn0cVFVdUofghQ4bo5x05cqSxZMkS/W9+7bXXjjsWqlD7I488YgwaNEgXJlfPp/q3dOnSE/apsLDQuPnmm41evXoZCQkJuiD3pZdeqk+329a5Odkct2WsWvs3nWwcTvZzj/e8xzN+/HjjiSeeaPq6I3PcVmpMMzMzjRUrVoTl+QAAsYW1FWsr1lbtw9oKduNR/7M6MAbAvtSWsGHDhsmf/vQnq7sCm1A1uVRGmNpW11o2VTipTDS1DXDBggUR/TkAAEQLaysci7UV3IztewBapbZnLVmyRF8ef/xxRglNLr74Yl2oX52Z8Xi1uMJZuFQVTgcAwOlYW+F4WFvBzciUAtAqtb9eLZ5++ctf6jpMAAAA6DjWVgDwTQSlAAAAAAAAEHWRLQYCAAAAAAAAtIKgFAAAAAAAAKKOoBQAAAAAAACijqAUAAAAAAAAoo6gFAAAAAAAAKKOoBQAAAAAAACijqAUAAAAAAAAoo6gFAAAAAAAAKKOoBQAAAAAAACijqAUAAAAAAAAoo6gFAAAAAAAAKKOoBQAAAAAAACijqAUAAAAAAAAJNr+P74g6ojF8GDkAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 1200x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "height = results[\"vertical_edge\"].values\n",
    "\n",
    "fig, axes = plt.subplots(1, 2, figsize=(12, 6))\n",
    "\n",
    "# O3 photolysis channels\n",
    "axes[0].semilogx(\n",
    "    results[\"photolysis_rate_constants\"].sel(reaction=\"jo3_a\").values, height,\n",
    "    label=\"jo3_a: O3 + hv → O2 + O(1D)\"\n",
    ")\n",
    "axes[0].semilogx(\n",
    "    results[\"photolysis_rate_constants\"].sel(reaction=\"jo3_b\").values, height,\n",
    "    label=\"jo3_b: O3 + hv → O2 + O(3P)\"\n",
    ")\n",
    "axes[0].set_xlabel(\"Photolysis Rate Constant (s$^{-1}$)\")\n",
    "axes[0].set_ylabel(\"Height (km)\")\n",
    "axes[0].set_title(\"O3 Photolysis\")\n",
    "axes[0].legend()\n",
    "axes[0].grid(True)\n",
    "\n",
    "# NOx and HOx channels\n",
    "for name, label in [\n",
    "    (\"jno2\",    \"NO2 + hv → NO + O(3P)\"),\n",
    "    (\"jhno3\",   \"HNO3 + hv → OH + NO2\"),\n",
    "    (\"jh2o2\",   \"H2O2 + hv → OH + OH\"),\n",
    "    (\"jch2o_a\", \"CH2O + hv → H + HCO\"),\n",
    "]:\n",
    "    axes[1].semilogx(\n",
    "        results[\"photolysis_rate_constants\"].sel(reaction=name).values, height,\n",
    "        label=label\n",
    "    )\n",
    "axes[1].set_xlabel(\"Photolysis Rate Constant (s$^{-1}$)\")\n",
    "axes[1].set_ylabel(\"Height (km)\")\n",
    "axes[1].set_title(\"NOx / HOx Photolysis\")\n",
    "axes[1].legend()\n",
    "axes[1].grid(True)\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a3b4c5d6",
   "metadata": {},
   "source": [
    "## 7. Summary\n",
    "\n",
    "To run TUV-x with a reduced set of photolysis reactions:\n",
    "\n",
    "1. Load the JSON configuration with `json.load(open(vTS1.config_file_path()))`\n",
    "2. Filter `config[\"photolysis\"][\"reactions\"]` to only the reactions your mechanism needs\n",
    "3. Change to the config directory (`os.chdir`) so relative data file paths resolve correctly\n",
    "4. Pass the modified config as a JSON string via `TUVX(..., config_string=json.dumps(config))`\n",
    "\n",
    "The radiative transfer calculation is shared across all photolysis reactions and is not affected by the reaction list. Only the per-reaction cross section integration step is reduced, so savings scale with the fraction of reactions removed.\n",
    "\n",
    "The results for any given reaction are numerically identical to the full configuration run, since the radiation field is unchanged."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "musica",
   "language": "python",
   "name": "musica"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.14.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}