{
 "nbformat": 4,
 "nbformat_minor": 5,
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "name": "python",
   "version": "3.11.0"
  }
 },
 "cells": [
  {
   "cell_type": "markdown",
   "id": "intro",
   "metadata": {},
   "source": [
    "# Meteorites vs UFOs: Detection Bias Study\n",
    "\n",
    "1,279 records across 3 datasets exploring observation bias in sky-watching phenomena:\n",
    "- `temporal_comparison`: Year-by-year meteorite falls vs UFO reports (1900\u20132023)\n",
    "- `state_comparison`: 58 US states, UFO-per-meteorite ratio\n",
    "- `meteorite_detail`: 1,097 individual witnessed falls\n",
    "\n",
    "**Dataset**: [github.com/lukeslp/meteorites-ufos-detection-bias](https://github.com/lukeslp/meteorites-ufos-detection-bias)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "imports",
   "metadata": {},
   "outputs": [],
   "source": [
    "import json\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.ticker as mticker\n",
    "import numpy as np\n",
    "\n",
    "plt.style.use('seaborn-v0_8-whitegrid')\n",
    "plt.rcParams['figure.figsize'] = (14, 5)\n",
    "plt.rcParams['font.size'] = 11\n",
    "print('Libraries loaded')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "load-h",
   "metadata": {},
   "source": [
    "## 1. Load Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "load",
   "metadata": {},
   "outputs": [],
   "source": [
    "with open('meteorites_ufos_detection_bias.json') as f:\n",
    "    data = json.load(f)\n",
    "\n",
    "temporal = pd.DataFrame(data['temporal_comparison'])\n",
    "states = pd.DataFrame(data['state_comparison'])\n",
    "meteorites = pd.DataFrame(data['meteorite_detail'])\n",
    "\n",
    "print(f'Temporal records: {len(temporal)} (years {temporal.year.min()}-{temporal.year.max()})')\n",
    "print(f'State records: {len(states)}')\n",
    "print(f'Meteorite detail records: {len(meteorites)}')\n",
    "temporal.head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "temporal-h",
   "metadata": {},
   "source": [
    "## 2. Temporal Trends: UFO Reports vs Meteorite Falls (1900\u20132023)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "temporal",
   "metadata": {},
   "outputs": [],
   "source": [
    "fig, ax1 = plt.subplots(figsize=(14, 6))\n",
    "\n",
    "ax2 = ax1.twinx()\n",
    "ax1.fill_between(temporal['year'], temporal['ufo_sightings'], alpha=0.25, color='#FF6B35', label='UFO sightings')\n",
    "ax1.plot(temporal['year'], temporal['ufo_sightings'], color='#FF6B35', linewidth=1.5, label='UFO sightings')\n",
    "ax2.bar(temporal['year'], temporal['meteorite_falls'], alpha=0.6, color='#2196F3', width=0.8, label='Meteorite falls')\n",
    "\n",
    "ax1.set_xlabel('Year')\n",
    "ax1.set_ylabel('UFO Sightings (NUFORC)', color='#FF6B35')\n",
    "ax2.set_ylabel('Witnessed Meteorite Falls (NASA)', color='#2196F3')\n",
    "ax1.set_title('UFO Reports vs Meteorite Falls (1900\u20132023)', fontsize=14, fontweight='bold', pad=12)\n",
    "ax1.yaxis.set_major_formatter(mticker.FuncFormatter(lambda x, _: f'{x:,.0f}'))\n",
    "\n",
    "# Mark key events\n",
    "ax1.axvline(1947, color='purple', linestyle='--', alpha=0.5, label='Roswell (1947)')\n",
    "ax1.axvline(1993, color='green', linestyle='--', alpha=0.5, label='X-Files debut (1993)')\n",
    "\n",
    "lines1, labels1 = ax1.get_legend_handles_labels()\n",
    "lines2, labels2 = ax2.get_legend_handles_labels()\n",
    "ax1.legend(lines1 + lines2, labels1 + labels2, loc='upper left', fontsize=9)\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "state-h",
   "metadata": {},
   "source": [
    "## 3. State-Level UFO-per-Meteorite Ratio (Top 20 States)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "state-chart",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Filter states with valid ratios and sort\n",
    "states_valid = states.dropna(subset=['ufo_per_meteorite']).sort_values('ufo_per_meteorite', ascending=False)\n",
    "top20 = states_valid.head(20)\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(14, 8))\n",
    "colors = ['#FF4444' if v > 1000 else '#FF9800' if v > 100 else '#4CAF50' for v in top20['ufo_per_meteorite']]\n",
    "bars = ax.barh(top20['state'][::-1], top20['ufo_per_meteorite'][::-1], color=colors[::-1])\n",
    "\n",
    "ax.set_xlabel('UFO Reports per Witnessed Meteorite Fall')\n",
    "ax.set_title('Detection Bias: UFO Reports per Meteorite Fall by State', fontsize=14, fontweight='bold', pad=12)\n",
    "ax.xaxis.set_major_formatter(mticker.FuncFormatter(lambda x, _: f'{x:,.0f}'))\n",
    "\n",
    "for bar, (_, row) in zip(bars[::-1], top20.iterrows()):\n",
    "    ax.text(bar.get_width() + 10, bar.get_y() + bar.get_height()/2,\n",
    "            f'{row[\"ufo_per_meteorite\"]:,.0f}x', va='center', fontsize=9)\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()\n",
    "\n",
    "print('Top 5 most biased states:')\n",
    "print(top20[['state','meteorite_falls','ufo_sightings','ufo_per_meteorite']].head(5).to_string(index=False))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "meteor-h",
   "metadata": {},
   "source": [
    "## 4. Meteorite Fall Locations & Mass Distribution"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "meteor-scatter",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Filter to valid coordinates\n",
    "m_valid = meteorites.dropna(subset=['latitude', 'longitude', 'mass_g'])\n",
    "m_valid = m_valid[(m_valid['latitude'].between(-90, 90)) & (m_valid['longitude'].between(-180, 180))]\n",
    "\n",
    "fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(16, 6))\n",
    "\n",
    "# Scatter map\n",
    "sizes = np.sqrt(m_valid['mass_g'].clip(100, 1e8)) / 10\n",
    "ax1.scatter(m_valid['longitude'], m_valid['latitude'], s=sizes, alpha=0.5, color='#FF6B35', edgecolor='none')\n",
    "ax1.set_xlim(-180, 180)\n",
    "ax1.set_ylim(-90, 90)\n",
    "ax1.axhline(0, color='gray', linewidth=0.5)\n",
    "ax1.axvline(0, color='gray', linewidth=0.5)\n",
    "ax1.set_title(f'Witnessed Meteorite Falls (n={len(m_valid):,})', fontsize=13, fontweight='bold')\n",
    "ax1.set_xlabel('Longitude')\n",
    "ax1.set_ylabel('Latitude')\n",
    "\n",
    "# Mass histogram (log scale)\n",
    "ax2.hist(np.log10(m_valid['mass_g'].clip(lower=1)), bins=40, color='#2196F3', edgecolor='white', linewidth=0.5)\n",
    "ax2.set_title('Meteorite Mass Distribution (log scale)', fontsize=13, fontweight='bold')\n",
    "ax2.set_xlabel('log10(mass in grams)')\n",
    "ax2.set_ylabel('Count')\n",
    "ax2.set_xticks(range(0, 9))\n",
    "ax2.set_xticklabels(['1g', '10g', '100g', '1kg', '10kg', '100kg', '1t', '10t', '100t'])\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bias-h",
   "metadata": {},
   "source": [
    "## 5. Detection Bias Summary Table"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "bias-table",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Top 10 most 'biased' states\n",
    "ranked = states.dropna(subset=['ufo_per_meteorite']).sort_values('ufo_per_meteorite', ascending=False)\n",
    "\n",
    "print('=== Top 10 Most \"Biased\" States (UFO reports per meteorite fall) ===')\n",
    "top10 = ranked.head(10)[['state', 'meteorite_falls', 'ufo_sightings', 'ufo_per_meteorite']].copy()\n",
    "top10.columns = ['State', 'Meteorite Falls', 'UFO Reports', 'UFO per Meteorite']\n",
    "top10['UFO per Meteorite'] = top10['UFO per Meteorite'].apply(lambda x: f'{x:,.0f}x')\n",
    "print(top10.to_string(index=False))\n",
    "\n",
    "print('\\n=== Key Observations ===')\n",
    "total_ufo = states['ufo_sightings'].sum()\n",
    "total_met = states['meteorite_falls'].sum()\n",
    "print(f'Total US UFO reports: {total_ufo:,}')\n",
    "print(f'Total US witnessed meteorite falls: {total_met}')\n",
    "print(f'Overall US ratio: {total_ufo/total_met:.0f} UFO reports per meteorite fall')\n",
    "print(f'\\nPost-1990s UFO surge: {temporal[temporal.year >= 1990][\"ufo_sightings\"].sum():,} reports vs {temporal[temporal.year < 1990][\"ufo_sightings\"].sum():,} before 1990')"
   ]
  }
 ]
}