first commit

module3/exo3/exercice_fr.ipynb

 {
- "cells": [],
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# CO₂ concentration in the atmosphere since 1958\n",
+    "\n",
+    "This computational document analyzes the evolution of atmospheric carbon dioxide concentration measured at Mauna Loa Observatory since 1958.\n",
+    "\n",
+    "The data come from NOAA Global Monitoring Laboratory. The objective is to visualize the long-term trend, the seasonal cycle, and the annual increase in CO₂ concentration."
+   ]
+  },
+  {
+   "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",
+    "from scipy.stats import linregress"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## 1. Loading the data\n",
+    "\n",
+    "We use the monthly mean CO₂ concentration data from Mauna Loa Observatory.\n",
+    "The concentration is expressed in parts per million, abbreviated ppm."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "url = \"https://gml.noaa.gov/webdata/ccgg/trends/co2/co2_mm_mlo.txt\"\n",
+    "\n",
+    "columns = [\n",
+    "    \"year\",\n",
+    "    \"month\",\n",
+    "    \"decimal_date\",\n",
+    "    \"average\",\n",
+    "    \"deseasonalized\",\n",
+    "    \"days\",\n",
+    "    \"stdev\",\n",
+    "    \"uncertainty\"\n",
+    "]\n",
+    "\n",
+    "co2 = pd.read_csv(\n",
+    "    url,\n",
+    "    comment=\"#\",\n",
+    "    sep=r\"\\s+\",\n",
+    "    names=columns\n",
+    ")\n",
+    "\n",
+    "co2.head()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## 2. Cleaning the data\n",
+    "\n",
+    "Missing monthly values are coded as negative values in the NOAA file. We replace them by missing values and remove them from the analysis."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "co2[\"average\"] = co2[\"average\"].replace(-99.99, np.nan)\n",
+    "co2[\"deseasonalized\"] = co2[\"deseasonalized\"].replace(-99.99, np.nan)\n",
+    "\n",
+    "co2 = co2.dropna(subset=[\"average\"])\n",
+    "\n",
+    "co2[\"date\"] = pd.to_datetime(\n",
+    "    co2[\"year\"].astype(str) + \"-\" + co2[\"month\"].astype(str) + \"-15\"\n",
+    ")\n",
+    "\n",
+    "co2.head()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## 3. Descriptive statistics\n",
+    "\n",
+    "We compute basic descriptive statistics for monthly atmospheric CO₂ concentration."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "stats = co2[\"average\"].describe()\n",
+    "stats"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "print(\"Minimum CO2 concentration:\", round(co2[\"average\"].min(), 2), \"ppm\")\n",
+    "print(\"Maximum CO2 concentration:\", round(co2[\"average\"].max(), 2), \"ppm\")\n",
+    "print(\"Mean CO2 concentration:\", round(co2[\"average\"].mean(), 2), \"ppm\")\n",
+    "print(\"Median CO2 concentration:\", round(co2[\"average\"].median(), 2), \"ppm\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## 4. Evolution of monthly CO₂ concentration\n",
+    "\n",
+    "The following graph shows the monthly average CO₂ concentration since 1958."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "plt.figure(figsize=(10, 5))\n",
+    "plt.plot(co2[\"date\"], co2[\"average\"])\n",
+    "plt.xlabel(\"Year\")\n",
+    "plt.ylabel(\"CO₂ concentration (ppm)\")\n",
+    "plt.title(\"Monthly CO₂ concentration at Mauna Loa since 1958\")\n",
+    "plt.grid(True)\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## 5. Seasonal variability\n",
+    "\n",
+    "The monthly data show a seasonal cycle. To see this more clearly, we compute the average CO₂ concentration for each month of the year."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "monthly_cycle = co2.groupby(\"month\")[\"average\"].mean()\n",
+    "\n",
+    "plt.figure(figsize=(8, 5))\n",
+    "plt.plot(monthly_cycle.index, monthly_cycle.values, marker=\"o\")\n",
+    "plt.xlabel(\"Month\")\n",
+    "plt.ylabel(\"Average CO₂ concentration (ppm)\")\n",
+    "plt.title(\"Average seasonal cycle of CO₂ concentration\")\n",
+    "plt.xticks(range(1, 13))\n",
+    "plt.grid(True)\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## 6. Annual mean CO₂ concentration\n",
+    "\n",
+    "We compute the annual average concentration to remove most of the seasonal variation and focus on the long-term trend."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "annual = co2.groupby(\"year\")[\"average\"].mean().reset_index()\n",
+    "annual.head()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "plt.figure(figsize=(10, 5))\n",
+    "plt.plot(annual[\"year\"], annual[\"average\"], marker=\"o\")\n",
+    "plt.xlabel(\"Year\")\n",
+    "plt.ylabel(\"Annual average CO₂ concentration (ppm)\")\n",
+    "plt.title(\"Annual average CO₂ concentration at Mauna Loa\")\n",
+    "plt.grid(True)\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## 7. Linear trend\n",
+    "\n",
+    "We estimate a simple linear trend for annual average CO₂ concentration."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "slope, intercept, r_value, p_value, std_err = linregress(\n",
+    "    annual[\"year\"],\n",
+    "    annual[\"average\"]\n",
+    ")\n",
+    "\n",
+    "annual[\"trend\"] = intercept + slope * annual[\"year\"]\n",
+    "\n",
+    "print(\"Estimated annual increase:\", round(slope, 3), \"ppm per year\")\n",
+    "print(\"R-squared:\", round(r_value**2, 4))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "plt.figure(figsize=(10, 5))\n",
+    "plt.plot(annual[\"year\"], annual[\"average\"], marker=\"o\", label=\"Annual mean\")\n",
+    "plt.plot(annual[\"year\"], annual[\"trend\"], label=\"Linear trend\")\n",
+    "plt.xlabel(\"Year\")\n",
+    "plt.ylabel(\"CO₂ concentration (ppm)\")\n",
+    "plt.title(\"Linear trend of annual CO₂ concentration\")\n",
+    "plt.legend()\n",
+    "plt.grid(True)\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## 8. Annual growth rate\n",
+    "\n",
+    "We compute the year-to-year difference in annual mean CO₂ concentration."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "annual[\"growth\"] = annual[\"average\"].diff()\n",
+    "\n",
+    "plt.figure(figsize=(10, 5))\n",
+    "plt.bar(annual[\"year\"], annual[\"growth\"])\n",
+    "plt.xlabel(\"Year\")\n",
+    "plt.ylabel(\"Annual increase (ppm)\")\n",
+    "plt.title(\"Year-to-year increase in atmospheric CO₂\")\n",
+    "plt.grid(True)\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "largest_growth = annual.loc[annual[\"growth\"].idxmax()]\n",
+    "smallest_growth = annual.loc[annual[\"growth\"].idxmin()]\n",
+    "\n",
+    "print(\"Largest annual increase:\")\n",
+    "print(largest_growth[[\"year\", \"growth\"]])\n",
+    "\n",
+    "print(\"\\nSmallest annual increase:\")\n",
+    "print(smallest_growth[[\"year\", \"growth\"]])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## 9. Conclusion\n",
+    "\n",
+    "The analysis shows a clear long-term increase in atmospheric CO₂ concentration at Mauna Loa since 1958. The monthly data also show a seasonal cycle, which is visible when averaging observations by month. The annual averages confirm that CO₂ concentration has increased steadily over the period covered by the dataset.\n",
+    "\n",
+    "This notebook is reproducible because it downloads the data from the original source, documents each transformation step, and produces the tables and figures directly from the code."
+   ]
+  }
+ ],
  "metadata": {
   "kernelspec": {
    "display_name": "Python 3",
@@ -16,10 +318,9 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.6.3"
+   "version": "3.6.4"
   }
  },
  "nbformat": 4,
  "nbformat_minor": 2
 }
-