{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 高温热浪与银发群体健康风险 -- 探索性数据分析\n",
    "焦作市 . 郑州市 | 2010-2024 年气象数据"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from pathlib import Path\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "sns.set_style(\"whitegrid\")\n",
    "plt.rcParams[\"font.sans-serif\"] = [\"SimHei\", \"Microsoft YaHei\", \"DejaVu Sans\"]\n",
    "plt.rcParams[\"axes.unicode_minus\"] = False\n",
    "\n",
    "from src.utils.config import DATA_PROCESSED, DATA_EXTERNAL, OUTPUT_FIGURES, CITIES\n",
    "\n",
    "# 尝试加载数据\n",
    "try:\n",
    "    df_jz = pd.read_csv(DATA_PROCESSED / \"jiaozuo_processed.csv\", parse_dates=[\"time\"])\n",
    "    df_zz = pd.read_csv(DATA_PROCESSED / \"zhengzhou_processed.csv\", parse_dates=[\"time\"])\n",
    "    df_combined = pd.read_csv(DATA_PROCESSED / \"combined_processed.csv\", parse_dates=[\"time\"])\n",
    "    print(f\"焦作: {df_jz.shape[0]} 天, 郑州: {df_zz.shape[0]} 天\")\n",
    "    data_loaded = True\n",
    "except FileNotFoundError:\n",
    "    print(\"处理后的数据不存在，请先运行 preprocess.py\")\n",
    "    print(\"将使用模拟数据演示分析框架\")\n",
    "    data_loaded = False\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "if data_loaded:\n",
    "    fig, axes = plt.subplots(1, 2, figsize=(14, 5))\n",
    "    for ax, (df, name) in zip(axes, [(df_jz, \"焦作\"), (df_zz, \"郑州\")]):\n",
    "        annual = df.groupby(df[\"time\"].dt.year)[\"temp_mean\"].agg([\"mean\", \"max\", \"min\"])\n",
    "        annual.plot(ax=ax, color=[\"#ff9800\", \"#f44336\", \"#5b9bd5\"])\n",
    "        ax.set_title(f\"{name} - 年均气温趋势\", fontsize=14)\n",
    "        ax.set_ylabel(\"温度 (C)\")\n",
    "        ax.set_xlabel(\"年份\")\n",
    "        ax.legend([\"平均\", \"最高\", \"最低\"])\n",
    "    fig.tight_layout()\n",
    "    plt.savefig(OUTPUT_FIGURES / \"annual_temp_trend.png\", dpi=150, bbox_inches=\"tight\")\n",
    "    plt.show()\n",
    "else:\n",
    "    print(\"需要数据文件\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "if data_loaded:\n",
    "    fig, axes = plt.subplots(1, 2, figsize=(12, 5))\n",
    "    labels = [\"低\", \"中\", \"高\", \"严重\"]\n",
    "    colors = [\"#00e676\", \"#ffeb3b\", \"#ff9800\", \"#f44336\"]\n",
    "    for ax, (df, name) in zip(axes, [(df_jz, \"焦作\"), (df_zz, \"郑州\")]):\n",
    "        counts = df[\"risk_label\"].value_counts().sort_index()\n",
    "        values = [counts.get(i, 0) for i in range(4)]\n",
    "        ax.bar(labels, values, color=colors)\n",
    "        ax.set_title(f\"{name} - 风险等级分布\", fontsize=14)\n",
    "        for i, v in enumerate(values):\n",
    "            ax.text(i, v + max(values)*0.01, str(v), ha='center')\n",
    "    fig.tight_layout()\n",
    "    plt.savefig(OUTPUT_FIGURES / \"risk_distribution.png\", dpi=150, bbox_inches=\"tight\")\n",
    "    plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "if data_loaded:\n",
    "    for df, name in [(df_jz, \"焦作\"), (df_zz, \"郑州\")]:\n",
    "        annual_hw = df.groupby(df[\"time\"].dt.year)[\"heatwave\"].sum()\n",
    "        print(f\"\\n{name} 热浪天数统计:\")\n",
    "        print(annual_hw.describe())\n",
    "        print(f\"  年均热浪天数: {annual_hw.mean():.1f} 天\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "try:\n",
    "    er = pd.read_csv(DATA_EXTERNAL / \"exposure_response.csv\")\n",
    "    fig, ax = plt.subplots(figsize=(8, 5))\n",
    "    ax.plot(er[\"percentile\"], er[\"rr\"], \"o-\", color=\"#f44336\", linewidth=2, markersize=8)\n",
    "    ax.axhline(y=1.0, color=\"gray\", linestyle=\"--\", alpha=0.7)\n",
    "    ax.set_xlabel(\"温度百分位数 (%)\", fontsize=12)\n",
    "    ax.set_ylabel(\"相对风险 (RR)\", fontsize=12)\n",
    "    ax.set_title(\"温度-老年人死亡率暴露反应曲线\\n(来源: Chen et al. 2018, Lancet Planet Health)\", fontsize=13)\n",
    "    ax.fill_between(er[\"percentile\"], 1.0, er[\"rr\"], alpha=0.2, color=\"#f44336\")\n",
    "    plt.tight_layout()\n",
    "    plt.savefig(OUTPUT_FIGURES / \"exposure_response.png\", dpi=150, bbox_inches=\"tight\")\n",
    "    plt.show()\n",
    "except Exception as e:\n",
    "    print(f\"无法加载暴露反应数据: {e}\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "if data_loaded:\n",
    "    fig, axes = plt.subplots(1, 2, figsize=(14, 5))\n",
    "    for ax, (df, name) in zip(axes, [(df_jz, \"焦作\"), (df_zz, \"郑州\")]):\n",
    "        monthly = df.groupby(df[\"time\"].dt.month)[\"temp_mean\"].agg([\"mean\", \"std\"])\n",
    "        ax.fill_between(monthly.index, monthly[\"mean\"]-monthly[\"std\"],\n",
    "                        monthly[\"mean\"]+monthly[\"std\"], alpha=0.3, color=\"#ff9800\")\n",
    "        ax.plot(monthly.index, monthly[\"mean\"], \"o-\", color=\"#f44336\", linewidth=2)\n",
    "        ax.set_title(f\"{name} - 月均气温模式\", fontsize=14)\n",
    "        ax.set_xlabel(\"月份\")\n",
    "        ax.set_ylabel(\"温度 (C)\")\n",
    "        ax.set_xticks(range(1, 13))\n",
    "    fig.tight_layout()\n",
    "    plt.savefig(OUTPUT_FIGURES / \"monthly_temp_pattern.png\", dpi=150, bbox_inches=\"tight\")\n",
    "    plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## EDA 小结\n",
    "\n",
    "1. 郑州和焦作两市气温趋势高度一致，均呈缓慢上升趋势\n",
    "2. 夏季（6-8月）是高温热浪高发期，7月风险最高\n",
    "3. 风险等级分布呈长尾特征：低风险占多数，严重风险为稀有事件\n",
    "4. 温度-死亡率暴露反应曲线呈 J 型，高温端风险显著上升\n",
    "5. 两市老龄化率均在 11-13%，郑州老年人口绝对数量更大\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": ".venv",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "name": "python",
   "version": "3.13.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}