{ "cells": [ { "cell_type": "code", "execution_count": 12, "metadata": { "scrolled": true }, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Analyse de la courbe de Keeling\n", "\n", "\n", "1. Séparation des phénomènes :\n", " Méthode : Moyenne mobile sur 52 semaines\n", " → Tendance = moyenne mobile\n", " → Oscillation = données brutes - tendance\n", "\n", "2. Caractérisation de l'oscillation périodique :\n", " - Amplitude : 4.76 ppm\n", " - Période : 1.0 an\n", " - Maximum : Avr (concentration maximale)\n", " - Minimum : Oct (concentration minimale)\n", "\n", "3.Modèle de la contribution lente :\n", " Modèle choisi : Polynôme de degré 3\n", " CO₂(t) = -356789.11 + 561.5711·t + -0.2946250·t² + 0.0000515645·t³\n", "\n", " Paramètres estimés :\n", " - a₀ = -356789.11\n", " - a₁ = 561.5711\n", " - a₂ = -0.2946250\n", " - a₃ = 0.0000515645\n", "\n", "4. Extrapolation à 2025 :\n", " - CO₂ en 1958 : 316.25 ppm\n", " - CO₂ prédit en 2025 : 425.11 ppm\n", " - Augmentation : 108.86 ppm (+34.4%)\n" ] } ], "source": [ "import pandas as pd\n", "import numpy as np\n", "import matplotlib.pyplot as plt\n", "from scipy.optimize import curve_fit\n", "\n", "import warnings\n", "warnings.filterwarnings('ignore', category=FutureWarning)\n", "\n", "# Lecture des données\n", "data = pd.read_csv('weekly_in_situ_co2_mlo.csv', skiprows=46, header=None, \n", " names=['date', 'co2'])\n", "data['date'] = pd.to_datetime(data['date'])\n", "data = data.dropna()\n", "data['decimal_year'] = data['date'].dt.year + (data['date'].dt.dayofyear - 1) / 365.25\n", "\n", "# Calcul de la tendance avec moyenne mobile\n", "window = 52 # 52 semaines = 1 an\n", "data['trend'] = data['co2'].rolling(window=window, center=True).mean()\n", "data['seasonal'] = data['co2'] - data['trend']\n", "\n", "# Graphique : superposition des deux phénomènes\n", "plt.figure(figsize=(14, 6))\n", "\n", "plt.plot(data['date'], data['co2'], linewidth=0.5, alpha=0.6, \n", " label='Données brutes (oscillation + tendance)', color='steelblue')\n", "plt.plot(data['date'], data['trend'], linewidth=2.5, \n", " label='Tendance à long terme', color='red')\n", "\n", "plt.xlabel('Année', fontsize=12)\n", "plt.ylabel('CO₂ (ppm)', fontsize=12)\n", "plt.title('Courbe de Keeling : Oscillation saisonnière superposée à la tendance', \n", " fontsize=14, fontweight='bold')\n", "plt.legend(loc='upper left', fontsize=11)\n", "plt.grid(True, alpha=0.3)\n", "plt.tight_layout()\n", "plt.savefig('keeling_superpose.png', dpi=200)\n", "plt.show()\n", "\n", "# ========== 2. SÉPARATION ET ANALYSE ==========\n", "\n", "data_clean = data.dropna()\n", "\n", "# CARACTÉRISATION DE L'OSCILLATION SAISONNIÈRE\n", "seasonal_clean = data_clean['seasonal']\n", "amplitude = (seasonal_clean.max() - seasonal_clean.min()) / 2\n", "periode = 1.0 # an\n", "\n", "# Trouver la phase (moment du max et du min)\n", "idx_max = seasonal_clean.idxmax()\n", "idx_min = seasonal_clean.idxmin()\n", "mois_max = data_clean.loc[idx_max, 'date'].month\n", "mois_min = data_clean.loc[idx_min, 'date'].month\n", "mois_noms = ['', 'Jan', 'Fév', 'Mar', 'Avr', 'Mai', 'Juin', \n", " 'Juil', 'Août', 'Sep', 'Oct', 'Nov', 'Déc']\n", "\n", "# MODÈLE POLYNOMIAL de la tendance lente\n", "def poly3(x, a, b, c, d):\n", " return a + b*x + c*x**2 + d*x**3\n", "\n", "x_data = data_clean['decimal_year'].values\n", "y_data = data_clean['trend'].values\n", "params, _ = curve_fit(poly3, x_data, y_data)\n", "\n", "# EXTRAPOLATION à 2025\n", "co2_2025 = poly3(2025, *params)\n", "co2_debut = data_clean['trend'].iloc[0]\n", "\n", "# Affichage des résultats\n", "print(\"Analyse de la courbe de Keeling\")\n", "print(\"\\n\\n1. Séparation des phénomènes :\")\n", "print(\" Méthode : Moyenne mobile sur 52 semaines\")\n", "print(\" → Tendance = moyenne mobile\")\n", "print(\" → Oscillation = données brutes - tendance\")\n", "\n", "print(\"\\n2. Caractérisation de l'oscillation périodique :\")\n", "print(f\" - Amplitude : {amplitude:.2f} ppm\")\n", "print(f\" - Période : {periode:.1f} an\")\n", "print(f\" - Maximum : {mois_noms[mois_max]} (concentration maximale)\")\n", "print(f\" - Minimum : {mois_noms[mois_min]} (concentration minimale)\")\n", "\n", "print(\"\\n3.Modèle de la contribution lente :\")\n", "print(\" Modèle choisi : Polynôme de degré 3\")\n", "print(f\" CO₂(t) = {params[0]:.2f} + {params[1]:.4f}·t + {params[2]:.7f}·t² + {params[3]:.10f}·t³\")\n", "print(\"\\n Paramètres estimés :\")\n", "print(f\" - a₀ = {params[0]:.2f}\")\n", "print(f\" - a₁ = {params[1]:.4f}\")\n", "print(f\" - a₂ = {params[2]:.7f}\")\n", "print(f\" - a₃ = {params[3]:.10f}\")\n", "\n", "print(\"\\n4. Extrapolation à 2025 :\")\n", "print(f\" - CO₂ en 1958 : {co2_debut:.2f} ppm\")\n", "print(f\" - CO₂ prédit en 2025 : {co2_2025:.2f} ppm\")\n", "print(f\" - Augmentation : {co2_2025 - co2_debut:.2f} ppm (+{(co2_2025/co2_debut - 1)*100:.1f}%)\")\n", "\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.6.4" } }, "nbformat": 4, "nbformat_minor": 2 }