{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Sujet 1 - Concentration de CO2 dans l'atmosphère depuis 1958"
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [],
"source": [
"%matplotlib inline\n",
"import matplotlib.pyplot as plt\n",
"import numpy as np\n",
"import pandas as pd"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Récupération des données et analyse préliminaire"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Les données sur la concentration de CO2 dans l'atmosphère à l'observatoire de Mauna Loa sont disponibles sur le site web de l'[institut Scripps](https://scrippsco2.ucsd.edu/data/atmospheric_co2/primary_mlo_co2_record.html). La série de données s'étend de 1958 à nos jours. Les données considérées sont les relevés avec une granularité hebdomadaire. Elles sont disponibles à l'adresse suivante:"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
"data_url = \"https://scrippsco2.ucsd.edu/assets/data/atmospheric/stations/in_situ_co2/weekly/weekly_in_situ_co2_mlo.csv\""
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Pour nous protéger contre une éventuelle disparition ou modification du serveur de l'institut Scripps, nous faisons une copie locale de ce jeux de données que nous préservons avec notre analyse. Il est inutile et même risquée de télécharger les données à chaque exécution, car dans le cas d'une panne nous pourrions remplacer nos données par un fichier défectueux. Pour cette raison, nous téléchargeons les données seulement si la copie locale n'existe pas."
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [],
"source": [
"data_file = \"CO2_Mauna_Loa_Hebdo.csv\""
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [],
"source": [
"import os\n",
"import urllib.request\n",
"if not os.path.exists(data_file):\n",
" urllib.request.urlretrieve(data_url, data_file)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Le jeu de données est téléchargé le 08/10/2020 à 16h30."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Un examen préliminaire des données permet d'en observer la structure.\n",
"\n",
"On note que les 44 premières lignes décrivent le contexte général. Ellent devront donc être supprimées lors de l'extraction des données.\n",
"\n",
"Les lignes 40 à 44 précisent que les unités de concentration de CO2 sont des micro-moles de CO2 par mole (ppm). Ces valeurs sont considérées à 12h00 le samedi de chaque semaine.\n",
"\n",
"On note également que les dates sont au format \"AAAA-MM-JJ\"."
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"1 \"-------------------------------------------------------------------------------------------\"\n",
"\n",
"2 \" Atmospheric CO2 concentrations (ppm) derived from in situ air measurements \"\n",
"\n",
"3 \" at Mauna Loa, Observatory, Hawaii: Latitude 19.5°N Longitude 155.6°W Elevation 3397m \"\n",
"\n",
"4 \" \"\n",
"\n",
"5 \" Source: R. F. Keeling, S. J. Walker, S. C. Piper and A. F. Bollenbacher \"\n",
"\n",
"6 \" Scripps CO2 Program ( http://scrippsco2.ucsd.edu ) \"\n",
"\n",
"7 \" Scripps Institution of Oceanography (SIO) \"\n",
"\n",
"8 \" University of California \"\n",
"\n",
"9 \" La Jolla, California USA 92093-0244 \"\n",
"\n",
"10 \" \"\n",
"\n",
"11 \" Status of data and correspondence: \"\n",
"\n",
"12 \" \"\n",
"\n",
"13 \" These data are subject to revision based on recalibration of standard gases. Questions \"\n",
"\n",
"14 \" about the data should be directed to Dr. Ralph Keeling (rkeeling@ucsd.edu), Stephen Walker\"\n",
"\n",
"15 \" (sjwalker@ucsd.edu) and Stephen Piper (scpiper@ucsd.edu), Scripps CO2 Program. \"\n",
"\n",
"16 \" \"\n",
"\n",
"17 \" Baseline data in this file through 05-Oct-2020 from archive dated 06-Oct-2020 12:01:19 \"\n",
"\n",
"18 \" \"\n",
"\n",
"19 \"-------------------------------------------------------------------------------------------\"\n",
"\n",
"20 \" \"\n",
"\n",
"21 \" Please cite as: \"\n",
"\n",
"22 \" \"\n",
"\n",
"23 \" C. D. Keeling, S. C. Piper, R. B. Bacastow, M. Wahlen, T. P. Whorf, M. Heimann, and \"\n",
"\n",
"24 \" H. A. Meijer, Exchanges of atmospheric CO2 and 13CO2 with the terrestrial biosphere and \"\n",
"\n",
"25 \" oceans from 1978 to 2000. I. Global aspects, SIO Reference Series, No. 01-06, Scripps \"\n",
"\n",
"26 \" Institution of Oceanography, San Diego, 88 pages, 2001. \"\n",
"\n",
"27 \" \"\n",
"\n",
"28 \" If it is necessary to cite a peer-reviewed article, please cite as: \"\n",
"\n",
"29 \" \"\n",
"\n",
"30 \" C. D. Keeling, S. C. Piper, R. B. Bacastow, M. Wahlen, T. P. Whorf, M. Heimann, and \"\n",
"\n",
"31 \" H. A. Meijer, Atmospheric CO2 and 13CO2 exchange with the terrestrial biosphere and \"\n",
"\n",
"32 \" oceans from 1978 to 2000: observations and carbon cycle implications, pages 83-113, \"\n",
"\n",
"33 \" in \"A History of Atmospheric CO2 and its effects on Plants, Animals, and Ecosystems\", \"\n",
"\n",
"34 \" editors, Ehleringer, J.R., T. E. Cerling, M. D. Dearing, Springer Verlag, \"\n",
"\n",
"35 \" New York, 2005. \"\n",
"\n",
"36 \" \"\n",
"\n",
"37 \"-------------------------------------------------------------------------------------------\"\n",
"\n",
"38 \" \"\n",
"\n",
"39 \" \"\n",
"\n",
"40 \" The data file below contains 2 columns indicaing the date and CO2 \"\n",
"\n",
"41 \" concentrations in micro-mol CO2 per mole (ppm), reported on the 2008A \"\n",
"\n",
"42 \" SIO manometric mole fraction scale. These weekly values have been \"\n",
"\n",
"43 \" adjusted to 12:00 hours at middle day of each weekly period as \"\n",
"\n",
"44 \" indicated by the date in the first column. \"\n",
"\n",
"45 1958-03-29, 316.19\n",
"\n",
"46 1958-04-05, 317.31\n",
"\n",
"47 1958-04-12, 317.69\n",
"\n",
"48 1958-04-19, 317.58\n",
"\n",
"49 1958-04-26, 316.48\n",
"\n",
"50 1958-05-03, 316.95\n",
"\n",
"51 1958-05-17, 317.56\n",
"\n",
"52 1958-05-24, 317.99\n",
"\n",
"53 1958-07-05, 315.85\n",
"\n",
"54 1958-07-12, 315.85\n",
"\n",
"55 1958-07-19, 315.46\n",
"\n",
"56 1958-07-26, 315.59\n",
"\n",
"57 1958-08-02, 315.64\n",
"\n",
"58 1958-08-09, 315.10\n",
"\n",
"59 1958-08-16, 315.09\n",
"\n",
"60 1958-08-30, 314.14\n",
"\n"
]
}
],
"source": [
"Lignes = []\n",
"with open(data_file, \"r\", encoding='utf-8') as entree:\n",
" for ligne in entree:\n",
" Lignes.append(ligne)\n",
"for Cpt in range(60):\n",
" print(Cpt+1, Lignes[Cpt])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"On charge les données à partir de la ligne 45 et on examine leur teneur."
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"df.plot('Date', 'CO2')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Un graphique sur une période plus réduite permet de mieux observer le phénomène saisonnier.\n",
"\n",
"Il semble que le point bas du phénomène saisonnier se situe vers le mois d'octobre et son haut vers le mois de mai."
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
""
]
},
"execution_count": 15,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"
"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"df[200:350].plot('Date', 'CO2')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Elaboration d'un modèle prévisionnel de la tendance de long terme"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Nous nous proposons de tenter d'approcher cette courbe par un polynôme du second degré. L'ajustement du polynôme sera réalisé en minimisant la somme du carré des écarts. Cela devrait placer la courbe résultante comme la position moyenne du phénomène. On note un écart significatif entre la dynamique de la tendance de long terme et celle du phénomène saisonnier. Les deux parties du phénomène devraient donc pouvoir être correctement séparées. La tendance de long terme devrait pouvoir être approchée par un polynôme du second degré."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Toutefois, le data frame contient de nombreuses données manquantes ('NaN'). La présence de ces données manquantes pose problème à l'algorithme d'ajustement du polynôme. Il faut donc limiter l'ajustement à la période de mesure disponible (indice 0 à 3255, cf. ci-dessous). Pour les données manquantes à l'intérieur de cette période, nous utiliserons une interpolation linéaire qui ne devrait trop affecter le résultat final."
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"df[-500:-300].plot('Date', ['CO2', 'Prév_LT'])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Observation du phénomène saisonnier"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Pour séparer le phénomène saisonnier de la tendance de long terme, nous calculons la différence entre les mesures de CO2 et la tendance de long terme précédemment évaluée."
]
},
{
"cell_type": "code",
"execution_count": 30,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
Date
\n",
"
Week
\n",
"
CO2
\n",
"
Prév_LT
\n",
"
Saisonnier
\n",
"
\n",
" \n",
" \n",
"
\n",
"
0
\n",
"
1958-03-29
\n",
"
13
\n",
"
316.19
\n",
"
314.693329
\n",
"
1.496671
\n",
"
\n",
"
\n",
"
1
\n",
"
1958-04-05
\n",
"
14
\n",
"
317.31
\n",
"
314.707831
\n",
"
2.602169
\n",
"
\n",
"
\n",
"
2
\n",
"
1958-04-12
\n",
"
15
\n",
"
317.69
\n",
"
314.722342
\n",
"
2.967658
\n",
"
\n",
"
\n",
"
3
\n",
"
1958-04-19
\n",
"
16
\n",
"
317.58
\n",
"
314.736864
\n",
"
2.843136
\n",
"
\n",
"
\n",
"
4
\n",
"
1958-04-26
\n",
"
17
\n",
"
316.48
\n",
"
314.751394
\n",
"
1.728606
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" Date Week CO2 Prév_LT Saisonnier\n",
"0 1958-03-29 13 316.19 314.693329 1.496671\n",
"1 1958-04-05 14 317.31 314.707831 2.602169\n",
"2 1958-04-12 15 317.69 314.722342 2.967658\n",
"3 1958-04-19 16 317.58 314.736864 2.843136\n",
"4 1958-04-26 17 316.48 314.751394 1.728606"
]
},
"execution_count": 30,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df = df.assign(Saisonnier = df['CO2'] - df['Prév_LT'])\n",
"df.head()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Quelques graphiques pour observer le phénomène saisonnier. Le calcul de la moyenne sur la partie saisonnière devrait être proche de 0, si tout va bien. Ce qui semble être le cas.\n",
"\n",
"On notera toutefois une sorte de pseudo période avec ce qui semble être des pics vers 1960, 1990 et 2020."
]
},
{
"cell_type": "code",
"execution_count": 31,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"-0.01333697107852895"
]
},
"execution_count": 31,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df['Saisonnier'].mean()"
]
},
{
"cell_type": "code",
"execution_count": 32,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
""
]
},
"execution_count": 32,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"
"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"df[-500:-300].plot('Date', ['Saisonnier'])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Evaluation du phénomène saisonnier moyen depuis 1958"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Nous avons observé que le phénomène saisonnier atteint un pic minimum vers le premier octobre et un point haut vers le mois de mai. Nous avons également observé que de nombreuses semaines de relevés sont manquantes.\n",
"\n",
"Une manière d'aborder le phénomène sans trop se préoccuper ni de son phasage dans l'année ni des données manquantes, serait de calculer la valeur moyenne de la partie saisonnière en fonction de la position de la semaine dans l'année. Cela permettrait d'utiliser au mieux les données disponibles. Une moyenne sur de nombreuses annèes de devrait pas être sensiblement affecter par l'absence de quelques données.\n",
"\n",
"Cela permettrait de raffiner les prévisions ultérieures."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Pour cela, on construit un tableau qui va permettre, pour chaque numéro de semaine de l'année, de cumuler la part saisonnière du taux de CO2 ainsi que le nombre de valeurs considérées. Il suffira ensuite de calculer la valeur moyenne observée pour chacune des semaines."
]
},
{
"cell_type": "code",
"execution_count": 34,
"metadata": {},
"outputs": [],
"source": [
"# Parcourir le tableau.\n",
"# Pour chaque date, calculer la position de la semaine dans l'année.\n",
"# Ajouter la valeur du CO2_Saisonnier dans le tableau de résultats et incrémenter le nombre de valeurs\n",
"# disponibles pour ce numéro de semaine.\n",
"\n",
"année = [] # Créer le tableau de résultats.\n",
"for i in range (0, 55):\n",
" # La première information est le numéro de semaine,\n",
" # le second, le cumul de taux, le troisième le nombre de valeurs cumulées.\n",
" # Le quatrième contient la moyenne calculée.\n",
" année.append([i, 0.0, 0, 0.0])\n",
" \n",
"for cpt in range(df.index.min(), df.index.max()):\n",
" # Récupérer les informations.\n",
" sem = df.at[cpt, 'Week']\n",
" taux_w = df.at[cpt, 'Saisonnier']\n",
" if pd.isna(taux_w):\n",
" pass\n",
" else:\n",
" année[sem][1] = année[sem][1] + df.at[cpt, 'Saisonnier']\n",
" année[sem][2] = année[sem][2] + 1\n",
"\n",
"for i in range (0, 55): # Calculer la moyenne pour chaque semaine.\n",
" if année[i][2] != 0:\n",
" année[i][3] = année[i][1] / année[i][2] "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"On affiche le tableau résultant afin de vérifier que toutes les semaines ont été correctement traitées et que le nombre des données est suffisant pour que la moyenne ait un sens (de même que l'approche proposée)."
]
},
{
"cell_type": "code",
"execution_count": 35,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
Num_semaine
\n",
"
Cumul
\n",
"
Nbre_valeurs
\n",
"
Moyenne_semaine
\n",
"
\n",
" \n",
" \n",
"
\n",
"
0
\n",
"
1
\n",
"
-10.589429
\n",
"
62
\n",
"
-0.170797
\n",
"
\n",
"
\n",
"
1
\n",
"
2
\n",
"
-0.588472
\n",
"
62
\n",
"
-0.009491
\n",
"
\n",
"
\n",
"
2
\n",
"
3
\n",
"
10.751143
\n",
"
60
\n",
"
0.179186
\n",
"
\n",
"
\n",
"
3
\n",
"
4
\n",
"
18.266655
\n",
"
59
\n",
"
0.309604
\n",
"
\n",
"
\n",
"
4
\n",
"
5
\n",
"
27.526169
\n",
"
61
\n",
"
0.451249
\n",
"
\n",
"
\n",
"
5
\n",
"
6
\n",
"
32.662548
\n",
"
59
\n",
"
0.553603
\n",
"
\n",
"
\n",
"
6
\n",
"
7
\n",
"
44.786100
\n",
"
59
\n",
"
0.759086
\n",
"
\n",
"
\n",
"
7
\n",
"
8
\n",
"
51.654865
\n",
"
60
\n",
"
0.860914
\n",
"
\n",
"
\n",
"
8
\n",
"
9
\n",
"
60.867316
\n",
"
60
\n",
"
1.014455
\n",
"
\n",
"
\n",
"
9
\n",
"
10
\n",
"
69.049186
\n",
"
60
\n",
"
1.150820
\n",
"
\n",
"
\n",
"
10
\n",
"
11
\n",
"
86.480754
\n",
"
59
\n",
"
1.465775
\n",
"
\n",
"
\n",
"
11
\n",
"
12
\n",
"
101.715843
\n",
"
61
\n",
"
1.667473
\n",
"
\n",
"
\n",
"
12
\n",
"
13
\n",
"
124.307536
\n",
"
61
\n",
"
2.037828
\n",
"
\n",
"
\n",
"
13
\n",
"
14
\n",
"
145.581964
\n",
"
61
\n",
"
2.386590
\n",
"
\n",
"
\n",
"
14
\n",
"
15
\n",
"
153.215802
\n",
"
61
\n",
"
2.511734
\n",
"
\n",
"
\n",
"
15
\n",
"
16
\n",
"
167.599048
\n",
"
61
\n",
"
2.747525
\n",
"
\n",
"
\n",
"
16
\n",
"
17
\n",
"
176.867921
\n",
"
62
\n",
"
2.852708
\n",
"
\n",
"
\n",
"
17
\n",
"
18
\n",
"
178.714452
\n",
"
61
\n",
"
2.929745
\n",
"
\n",
"
\n",
"
18
\n",
"
19
\n",
"
186.536568
\n",
"
61
\n",
"
3.057977
\n",
"
\n",
"
\n",
"
19
\n",
"
20
\n",
"
191.059264
\n",
"
62
\n",
"
3.081601
\n",
"
\n",
"
\n",
"
20
\n",
"
21
\n",
"
188.481844
\n",
"
62
\n",
"
3.040030
\n",
"
\n",
"
\n",
"
21
\n",
"
22
\n",
"
175.280241
\n",
"
61
\n",
"
2.873447
\n",
"
\n",
"
\n",
"
22
\n",
"
23
\n",
"
162.132929
\n",
"
62
\n",
"
2.615047
\n",
"
\n",
"
\n",
"
23
\n",
"
24
\n",
"
138.229765
\n",
"
60
\n",
"
2.303829
\n",
"
\n",
"
\n",
"
24
\n",
"
25
\n",
"
121.184148
\n",
"
61
\n",
"
1.986625
\n",
"
\n",
"
\n",
"
25
\n",
"
26
\n",
"
101.067671
\n",
"
61
\n",
"
1.656847
\n",
"
\n",
"
\n",
"
26
\n",
"
27
\n",
"
80.253080
\n",
"
63
\n",
"
1.273858
\n",
"
\n",
"
\n",
"
27
\n",
"
28
\n",
"
50.572217
\n",
"
62
\n",
"
0.815681
\n",
"
\n",
"
\n",
"
28
\n",
"
29
\n",
"
16.581010
\n",
"
62
\n",
"
0.267436
\n",
"
\n",
"
\n",
"
29
\n",
"
30
\n",
"
-4.190797
\n",
"
62
\n",
"
-0.067594
\n",
"
\n",
"
\n",
"
30
\n",
"
31
\n",
"
-38.626081
\n",
"
62
\n",
"
-0.623001
\n",
"
\n",
"
\n",
"
31
\n",
"
32
\n",
"
-66.919409
\n",
"
62
\n",
"
-1.079345
\n",
"
\n",
"
\n",
"
32
\n",
"
33
\n",
"
-97.759617
\n",
"
62
\n",
"
-1.576768
\n",
"
\n",
"
\n",
"
33
\n",
"
34
\n",
"
-127.481090
\n",
"
61
\n",
"
-2.089854
\n",
"
\n",
"
\n",
"
34
\n",
"
35
\n",
"
-154.977823
\n",
"
62
\n",
"
-2.499642
\n",
"
\n",
"
\n",
"
35
\n",
"
36
\n",
"
-179.235258
\n",
"
62
\n",
"
-2.890891
\n",
"
\n",
"
\n",
"
36
\n",
"
37
\n",
"
-192.193257
\n",
"
62
\n",
"
-3.099891
\n",
"
\n",
"
\n",
"
37
\n",
"
38
\n",
"
-216.123917
\n",
"
62
\n",
"
-3.485870
\n",
"
\n",
"
\n",
"
38
\n",
"
39
\n",
"
-219.345177
\n",
"
62
\n",
"
-3.537825
\n",
"
\n",
"
\n",
"
39
\n",
"
40
\n",
"
-210.551953
\n",
"
60
\n",
"
-3.509199
\n",
"
\n",
"
\n",
"
40
\n",
"
41
\n",
"
-197.019602
\n",
"
59
\n",
"
-3.339315
\n",
"
\n",
"
\n",
"
41
\n",
"
42
\n",
"
-186.854816
\n",
"
60
\n",
"
-3.114247
\n",
"
\n",
"
\n",
"
42
\n",
"
43
\n",
"
-182.706539
\n",
"
61
\n",
"
-2.995189
\n",
"
\n",
"
\n",
"
43
\n",
"
44
\n",
"
-158.255958
\n",
"
60
\n",
"
-2.637599
\n",
"
\n",
"
\n",
"
44
\n",
"
45
\n",
"
-144.295596
\n",
"
62
\n",
"
-2.327348
\n",
"
\n",
"
\n",
"
45
\n",
"
46
\n",
"
-121.739727
\n",
"
62
\n",
"
-1.963544
\n",
"
\n",
"
\n",
"
46
\n",
"
47
\n",
"
-102.243115
\n",
"
61
\n",
"
-1.676117
\n",
"
\n",
"
\n",
"
47
\n",
"
48
\n",
"
-89.919792
\n",
"
62
\n",
"
-1.450319
\n",
"
\n",
"
\n",
"
48
\n",
"
49
\n",
"
-71.575725
\n",
"
62
\n",
"
-1.154447
\n",
"
\n",
"
\n",
"
49
\n",
"
50
\n",
"
-55.622259
\n",
"
62
\n",
"
-0.897133
\n",
"
\n",
"
\n",
"
50
\n",
"
51
\n",
"
-42.949393
\n",
"
62
\n",
"
-0.692732
\n",
"
\n",
"
\n",
"
51
\n",
"
52
\n",
"
-27.680726
\n",
"
61
\n",
"
-0.453782
\n",
"
\n",
"
\n",
"
52
\n",
"
53
\n",
"
-4.525448
\n",
"
11
\n",
"
-0.411404
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" Num_semaine Cumul Nbre_valeurs Moyenne_semaine\n",
"0 1 -10.589429 62 -0.170797\n",
"1 2 -0.588472 62 -0.009491\n",
"2 3 10.751143 60 0.179186\n",
"3 4 18.266655 59 0.309604\n",
"4 5 27.526169 61 0.451249\n",
"5 6 32.662548 59 0.553603\n",
"6 7 44.786100 59 0.759086\n",
"7 8 51.654865 60 0.860914\n",
"8 9 60.867316 60 1.014455\n",
"9 10 69.049186 60 1.150820\n",
"10 11 86.480754 59 1.465775\n",
"11 12 101.715843 61 1.667473\n",
"12 13 124.307536 61 2.037828\n",
"13 14 145.581964 61 2.386590\n",
"14 15 153.215802 61 2.511734\n",
"15 16 167.599048 61 2.747525\n",
"16 17 176.867921 62 2.852708\n",
"17 18 178.714452 61 2.929745\n",
"18 19 186.536568 61 3.057977\n",
"19 20 191.059264 62 3.081601\n",
"20 21 188.481844 62 3.040030\n",
"21 22 175.280241 61 2.873447\n",
"22 23 162.132929 62 2.615047\n",
"23 24 138.229765 60 2.303829\n",
"24 25 121.184148 61 1.986625\n",
"25 26 101.067671 61 1.656847\n",
"26 27 80.253080 63 1.273858\n",
"27 28 50.572217 62 0.815681\n",
"28 29 16.581010 62 0.267436\n",
"29 30 -4.190797 62 -0.067594\n",
"30 31 -38.626081 62 -0.623001\n",
"31 32 -66.919409 62 -1.079345\n",
"32 33 -97.759617 62 -1.576768\n",
"33 34 -127.481090 61 -2.089854\n",
"34 35 -154.977823 62 -2.499642\n",
"35 36 -179.235258 62 -2.890891\n",
"36 37 -192.193257 62 -3.099891\n",
"37 38 -216.123917 62 -3.485870\n",
"38 39 -219.345177 62 -3.537825\n",
"39 40 -210.551953 60 -3.509199\n",
"40 41 -197.019602 59 -3.339315\n",
"41 42 -186.854816 60 -3.114247\n",
"42 43 -182.706539 61 -2.995189\n",
"43 44 -158.255958 60 -2.637599\n",
"44 45 -144.295596 62 -2.327348\n",
"45 46 -121.739727 62 -1.963544\n",
"46 47 -102.243115 61 -1.676117\n",
"47 48 -89.919792 62 -1.450319\n",
"48 49 -71.575725 62 -1.154447\n",
"49 50 -55.622259 62 -0.897133\n",
"50 51 -42.949393 62 -0.692732\n",
"51 52 -27.680726 61 -0.453782\n",
"52 53 -4.525448 11 -0.411404"
]
},
"execution_count": 35,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"sem = pd.DataFrame(année[1:54], columns=['Num_semaine', 'Cumul', 'Nbre_valeurs', 'Moyenne_semaine'])\n",
"sem"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Nous pouvons donc éventuellement envisager de compléter les données manquantes dans les mesures en utilisant ce tableau."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"On peut afficher la courbe résultante sur le cycle annuel."
]
},
{
"cell_type": "code",
"execution_count": 36,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
""
]
},
"execution_count": 36,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
""
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"sem.plot('Num_semaine', 'Moyenne_semaine')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Elaboration d'un modèle prévisionnel du cycle annuel"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Nous allons tenter d'approcher le cycle annuel moyen par un polynôme de degré 7. Le résultat semble correct après quelques essais effectués avec des polynômes de degré inférieur."
]
},
{
"cell_type": "code",
"execution_count": 37,
"metadata": {},
"outputs": [],
"source": [
"p = np.polyfit(sem['Num_semaine'], sem['Moyenne_semaine'], deg=7, full=False)"
]
},
{
"cell_type": "code",
"execution_count": 38,
"metadata": {},
"outputs": [],
"source": [
"def taux_semaine(x):\n",
" return np.polyval(p, x)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Calculer une année de prévision saisonnière."
]
},
{
"cell_type": "code",
"execution_count": 42,
"metadata": {},
"outputs": [],
"source": [
"prév_sem = []\n",
"for cpt in range(sem.index.min()+1, sem.index.max()+2):\n",
" prév_sem.append(taux_semaine(cpt))"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Afficher le plot. Le phénomène saisonnier moyen est affiché avec le modèle de prévision proposé. Le résultat semble correct."
]
},
{
"cell_type": "code",
"execution_count": 43,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
""
]
},
"execution_count": 43,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
""
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"sem = sem.assign(Prév_Sem = prév_sem)\n",
"sem.plot('Num_semaine', ['Moyenne_semaine', 'Prév_Sem'])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Nous disposons donc d'un modèle de prévision du phénomène saisonnier."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Comparaison entre les mesures et les prévisions du modèle"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Nous calculons pour chaque date du data frame \"df\", la part saisonnière du taux de CO2 en fonction du numéro de semaine, en utilisant le data frame \"sem\". Puis nous ajoutons la colonne correspondante au data frame \"df\"."
]
},
{
"cell_type": "code",
"execution_count": 44,
"metadata": {},
"outputs": [],
"source": [
"prév_sem = []\n",
"for cpt in range(df.index.min(), df.index.max()+1):\n",
" w = df.at[cpt, 'Week']\n",
" prév_sem.append(sem.at[w-1, 'Moyenne_semaine'])\n",
"df = df.assign(Prév_Sem = prév_sem)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Nous pouvons maintenant calculer la somme des contributions saisonnière et long terme et ajouter la colonne correspondante au data frame \"df\"."
]
},
{
"cell_type": "code",
"execution_count": 45,
"metadata": {},
"outputs": [],
"source": [
"df = df.assign(Prév_total = df['Prév_LT'] + df['Prév_Sem'])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Nous pouvons maintenant afficher quelques graphiques pour comparer un peu les données mesurées avec les données issues de la prévision."
]
},
{
"cell_type": "code",
"execution_count": 46,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
""
]
},
"execution_count": 46,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
""
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"df.plot('Date', ['CO2', 'Prév_total'])"
]
},
{
"cell_type": "code",
"execution_count": 47,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
""
]
},
"execution_count": 47,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
""
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"df[-500:-300].plot('Date', ['CO2', 'Prév_total'])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Nous calculons l'écart entre les mesures et les prévisions lorsque cela a un sens (hors valeurs 'NaN')."
]
},
{
"cell_type": "code",
"execution_count": 48,
"metadata": {},
"outputs": [],
"source": [
"Tab_écart_prév = []\n",
"for cpt in range(df.index.min(), df.index.max()):\n",
" # Récupérer les informations.\n",
" Ecart_prév = df.at[cpt, 'CO2'] - df.at[cpt, 'Prév_total']\n",
" if pd.isna(Ecart_prév):\n",
" pass\n",
" else:\n",
" Tab_écart_prév.append(Ecart_prév)\n",
"df_écart_prév = pd.DataFrame(Tab_écart_prév)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Nous affichons un histogramme afin d'évaluer l'ampleur des erreurs. L'hitogramme ne semble pas être tout à fait une gaussienne. On note cependant que la majorité des erreurs sont proches de 0. La gamme des écarts de prévision vont de -2ppm à +3ppm environ. Les erreurs les plus importantes sont en faible nombre. Pour une valeur de taux de CO2 qui est autour de 300ppm au moins, ces erreurs ne représentent que 1% environ."
]
},
{
"cell_type": "code",
"execution_count": 49,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"array([[]],\n",
" dtype=object)"
]
},
"execution_count": 49,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
""
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"df_écart_prév.hist()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Prévisions pour l'année 2025"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Nous pouvons maintenant établir notre prévision pour l'année 2025. Commençons par afficher le graphique correspondant."
]
},
{
"cell_type": "code",
"execution_count": 50,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
""
]
},
"execution_count": 50,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
""
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"df[-52:].plot('Date', ['Prév_LT', 'Prév_total'])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"On observe un taux moyen annuel (lié à la tendance long terme, choisie au début du mois de juillet), de l'ordre de 425.5ppm. Avec un minimum annuel de 422ppm et un maximum de l'ordre de 428ppm."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Recherchons les valeurs numériques plus précisément."
]
},
{
"cell_type": "code",
"execution_count": 51,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Le taux moyen de CO2 prévu pour l'année 2025 est de 425.21ppm, avec un minimum de 422.28ppm et un maximum de 427.99ppm.\n"
]
}
],
"source": [
"moy2025 = round(df[-52:]['Prév_total'].mean(), 2)\n",
"min2025 = round(df[-52:]['Prév_total'].min(), 2)\n",
"max2025 = round(df[-52:]['Prév_total'].max(), 2)\n",
"print(f\"Le taux moyen de CO2 prévu pour l'année 2025 est de {moy2025}ppm, avec un minimum de {min2025}ppm et un maximum de {max2025}ppm.\" )"
]
},
{
"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
}