{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Concentration de CO2 dans l'atmosphère depuis 1958\n", "\n", "Ce document analyse l'évolution de la concentration de CO2 dans l'atmosphère depuis 1958 à aujourd'hui, en se basant sur les données disponibles sur le site du [Scripps CO2 Program](https://scrippsco2.ucsd.edu/data/atmospheric_co2/primary_mlo_co2_record.html). Les données utilisées proviennent de relevés mensuels effectués à la station de Mauna Loa.\n", "\n", "Pour plus d'informations, vous pouvez consulter :\n", "\n", "_C. D. Keeling, S. C. Piper, R. B. Bacastow, M. Wahlen, T. P. Whorf, M. Heimann, and H. A. Meijer, Exchanges of atmospheric CO2 and 13CO2 with the terrestrial biosphere and oceans from 1978 to 2000. I. Global aspects, SIO Reference Series, No. 01-06, Scripps Institution of Oceanography, San Diego, 88 pages, 2001._\n" ] }, { "cell_type": "code", "execution_count": 75, "metadata": {}, "outputs": [], "source": [ " %matplotlib inline\n", "import matplotlib.pyplot as plt\n", "import pandas as pd\n", "import os\n", "import urllib\n", "import numpy as np\n", "from scipy.optimize import curve_fit\n", "import math" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Le fichier `csv` contenant les données est disponible à l'adresse suivante :" ] }, { "cell_type": "code", "execution_count": 76, "metadata": {}, "outputs": [], "source": [ "data_url = \"https://scrippsco2.ucsd.edu/assets/data/atmospheric/stations/in_situ_co2/monthly/monthly_in_situ_co2_mlo.csv\"" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ " Nous conservons les données dans un fichier local, afin que la conservation du fichier sur lequel nous effectuons notre analyse soit robuste aux défaillances du site Sentiweb. À cette fin également, nous ne retéléchargeons pas automatiquement une nouvelle version à chaque exécution, mais seulement si le fichier local est manquant." ] }, { "cell_type": "code", "execution_count": 77, "metadata": {}, "outputs": [], "source": [ "if not os.path.exists(\"monthly_in_situ_co2_mlo.csv\"):\n", " urllib.request.urlretrieve(data_url, \"monthly_in_situ_co2_mlo.csv\")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Comme expliqué dans le fichier `csv`, les données sont organisées en 10 colonnes :\n", "\n", "Numéro de colonne | Libellé | Contenu | Unité ou format (si applicable)\n", ":---------------- | :------ | :------ | :--------------------\n", "1 | Yr | Année du relevé |\n", "2 | Mn | Mois du relevé |\n", "3 | Date | Date du relevé | format Excel\n", "4 | Date | Date du relevé | date décimale\n", "5 | CO2 | Relevé de la concentration en C02 (2012 SIO manometric mole fraction scale), ajusté aux 24h du 15ème de chaque mois | ppm\n", "6 | seasonally adjusted | Relevés orignaux (CO2) ajustés pour retirer le cycle saisonnier régulier | ppm\n", "7 | fit | Relevés originaux (C02) lissés | ppm\n", "8 | seasonally adjusted fit | Relevés lissés ajustés saisonalement | ppm\n", "9 | CO2 filled | Relevés originaux (CO2) complétés par les résultats lissés (fit) | ppm\n", "10 | seasonally adjusted fit filled | Relevés ajustés saisonalement (seasonally adjusted) complétés par les résultats lissés et ajustés (seasonally adjusted fit) | ppm\n", "\n", "Les 55 premières lignes contiennent les explications ci-dessus, donc nous ne les traiterons pas avec les autres données. Nous renommons également les colonnes pour plus de concision et de clarté." ] }, { "cell_type": "code", "execution_count": 78, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
YrMnEDateDDateC02SAFitSAFitFilledSAFitFilled
019581212001958.0411-99.99-99.99-99.99-99.99-99.99-99.99
119582212311958.1260-99.99-99.99-99.99-99.99-99.99-99.99
219583212591958.2027315.71314.44316.20314.91315.71314.44
319584212901958.2877317.45315.16317.30314.99317.45315.16
419585213201958.3699317.51314.70317.88315.07317.51314.70
519586213511958.4548-99.99-99.99317.26315.15317.26315.15
619587213811958.5370315.87315.20315.86315.22315.87315.20
719588214121958.6219314.93316.21313.98315.29314.93316.21
819589214431958.7068313.21316.10312.45315.36313.21316.10
9195810214731958.7890-99.99-99.99312.43315.41312.43315.41
10195811215041958.8740313.33315.21313.61315.46313.33315.21
11195812215341958.9562314.67315.43314.77315.52314.67315.43
1219591215651959.0411315.58315.52315.64315.58315.58315.52
1319592215961959.1260316.49315.84316.29315.64316.49315.84
1419593216241959.2027316.65315.37316.99315.70316.65315.37
1519594216551959.2877317.72315.42318.09315.77317.72315.42
1619595216851959.3699318.29315.48318.67315.85318.29315.48
1719596217161959.4548318.15316.01318.06315.94318.15316.01
1819597217461959.5370316.54315.87316.67316.03316.54315.87
1919598217771959.6219314.80316.08314.81316.13314.80316.08
2019599218081959.7068313.84316.74313.30316.22313.84316.74
21195910218381959.7890313.33316.33313.32316.31313.33316.33
22195911218691959.8740314.81316.69314.54316.40314.81316.69
23195912218991959.9562315.58316.35315.73316.48315.58316.35
2419601219301960.0410316.43316.37316.63316.56316.43316.37
2519602219611960.1257316.98316.33317.30316.64316.98316.33
2619603219901960.2049317.58316.27318.04316.72317.58316.27
2719604220211960.2896319.03316.70319.14316.80319.03316.70
2819605220511960.3716320.03317.21319.69316.87320.03317.21
2919606220821960.4563319.59317.46319.03316.93319.59317.46
.................................
75020207440272020.5383414.42413.64414.77414.03414.42413.64
75120208440582020.6230412.52414.10412.61414.23412.52414.10
75220209440892020.7077411.18414.69410.90414.43411.18414.69
753202010441192020.7896411.12414.73411.02414.62411.12414.73
754202011441502020.8743412.88415.15412.57414.81412.88415.15
755202012441802020.9563413.89414.81414.08414.99413.89414.81
75620211442112021.0411415.15415.08415.25415.16415.15415.08
75720212442422021.1260416.47415.69416.13415.34416.47415.69
75820213442702021.2027417.16415.62417.06415.50417.16415.62
75920214443012021.2877418.24415.46418.47415.67418.24415.46
76020215443312021.3699418.95415.55419.24415.84418.95415.55
76120216443622021.4548418.70416.12418.58416.02418.70416.12
76220217443922021.5370416.65415.84416.97416.20416.65415.84
76320218444232021.6219414.34415.89414.79416.39414.34415.89
76420219444542021.7068412.90416.41413.05416.58412.90416.41
765202110444842021.7890413.55417.17413.15416.76413.55417.17
766202111445152021.8740414.82417.09414.70416.94414.82417.09
767202112445452021.9562416.43417.36416.20417.11416.43417.36
76820221445762022.0411418.01417.94417.35417.26418.01417.94
76920222446072022.1260418.99418.20418.20417.40418.99418.20
77020223446352022.2027418.45416.90419.08417.51418.45416.90
77120224446662022.2877420.02417.23420.44417.63420.02417.23
77220225446962022.3699420.78417.36421.16417.75420.78417.36
77320226447272022.4548420.68418.09420.43417.86420.68418.09
77420227447572022.5370418.66417.85418.75417.98418.66417.85
77520228447882022.6219-99.99-99.99-99.99-99.99-99.99-99.99
77620229448192022.7068-99.99-99.99-99.99-99.99-99.99-99.99
777202210448492022.7890-99.99-99.99-99.99-99.99-99.99-99.99
778202211448802022.8740-99.99-99.99-99.99-99.99-99.99-99.99
779202212449102022.9562-99.99-99.99-99.99-99.99-99.99-99.99
\n", "

780 rows × 10 columns

\n", "
" ], "text/plain": [ " Yr Mn EDate DDate C02 SA Fit SAFit Filled \\\n", "0 1958 1 21200 1958.0411 -99.99 -99.99 -99.99 -99.99 -99.99 \n", "1 1958 2 21231 1958.1260 -99.99 -99.99 -99.99 -99.99 -99.99 \n", "2 1958 3 21259 1958.2027 315.71 314.44 316.20 314.91 315.71 \n", "3 1958 4 21290 1958.2877 317.45 315.16 317.30 314.99 317.45 \n", "4 1958 5 21320 1958.3699 317.51 314.70 317.88 315.07 317.51 \n", "5 1958 6 21351 1958.4548 -99.99 -99.99 317.26 315.15 317.26 \n", "6 1958 7 21381 1958.5370 315.87 315.20 315.86 315.22 315.87 \n", "7 1958 8 21412 1958.6219 314.93 316.21 313.98 315.29 314.93 \n", "8 1958 9 21443 1958.7068 313.21 316.10 312.45 315.36 313.21 \n", "9 1958 10 21473 1958.7890 -99.99 -99.99 312.43 315.41 312.43 \n", "10 1958 11 21504 1958.8740 313.33 315.21 313.61 315.46 313.33 \n", "11 1958 12 21534 1958.9562 314.67 315.43 314.77 315.52 314.67 \n", "12 1959 1 21565 1959.0411 315.58 315.52 315.64 315.58 315.58 \n", "13 1959 2 21596 1959.1260 316.49 315.84 316.29 315.64 316.49 \n", "14 1959 3 21624 1959.2027 316.65 315.37 316.99 315.70 316.65 \n", "15 1959 4 21655 1959.2877 317.72 315.42 318.09 315.77 317.72 \n", "16 1959 5 21685 1959.3699 318.29 315.48 318.67 315.85 318.29 \n", "17 1959 6 21716 1959.4548 318.15 316.01 318.06 315.94 318.15 \n", "18 1959 7 21746 1959.5370 316.54 315.87 316.67 316.03 316.54 \n", "19 1959 8 21777 1959.6219 314.80 316.08 314.81 316.13 314.80 \n", "20 1959 9 21808 1959.7068 313.84 316.74 313.30 316.22 313.84 \n", "21 1959 10 21838 1959.7890 313.33 316.33 313.32 316.31 313.33 \n", "22 1959 11 21869 1959.8740 314.81 316.69 314.54 316.40 314.81 \n", "23 1959 12 21899 1959.9562 315.58 316.35 315.73 316.48 315.58 \n", "24 1960 1 21930 1960.0410 316.43 316.37 316.63 316.56 316.43 \n", "25 1960 2 21961 1960.1257 316.98 316.33 317.30 316.64 316.98 \n", "26 1960 3 21990 1960.2049 317.58 316.27 318.04 316.72 317.58 \n", "27 1960 4 22021 1960.2896 319.03 316.70 319.14 316.80 319.03 \n", "28 1960 5 22051 1960.3716 320.03 317.21 319.69 316.87 320.03 \n", "29 1960 6 22082 1960.4563 319.59 317.46 319.03 316.93 319.59 \n", ".. ... .. ... ... ... ... ... ... ... \n", "750 2020 7 44027 2020.5383 414.42 413.64 414.77 414.03 414.42 \n", "751 2020 8 44058 2020.6230 412.52 414.10 412.61 414.23 412.52 \n", "752 2020 9 44089 2020.7077 411.18 414.69 410.90 414.43 411.18 \n", "753 2020 10 44119 2020.7896 411.12 414.73 411.02 414.62 411.12 \n", "754 2020 11 44150 2020.8743 412.88 415.15 412.57 414.81 412.88 \n", "755 2020 12 44180 2020.9563 413.89 414.81 414.08 414.99 413.89 \n", "756 2021 1 44211 2021.0411 415.15 415.08 415.25 415.16 415.15 \n", "757 2021 2 44242 2021.1260 416.47 415.69 416.13 415.34 416.47 \n", "758 2021 3 44270 2021.2027 417.16 415.62 417.06 415.50 417.16 \n", "759 2021 4 44301 2021.2877 418.24 415.46 418.47 415.67 418.24 \n", "760 2021 5 44331 2021.3699 418.95 415.55 419.24 415.84 418.95 \n", "761 2021 6 44362 2021.4548 418.70 416.12 418.58 416.02 418.70 \n", "762 2021 7 44392 2021.5370 416.65 415.84 416.97 416.20 416.65 \n", "763 2021 8 44423 2021.6219 414.34 415.89 414.79 416.39 414.34 \n", "764 2021 9 44454 2021.7068 412.90 416.41 413.05 416.58 412.90 \n", "765 2021 10 44484 2021.7890 413.55 417.17 413.15 416.76 413.55 \n", "766 2021 11 44515 2021.8740 414.82 417.09 414.70 416.94 414.82 \n", "767 2021 12 44545 2021.9562 416.43 417.36 416.20 417.11 416.43 \n", "768 2022 1 44576 2022.0411 418.01 417.94 417.35 417.26 418.01 \n", "769 2022 2 44607 2022.1260 418.99 418.20 418.20 417.40 418.99 \n", "770 2022 3 44635 2022.2027 418.45 416.90 419.08 417.51 418.45 \n", "771 2022 4 44666 2022.2877 420.02 417.23 420.44 417.63 420.02 \n", "772 2022 5 44696 2022.3699 420.78 417.36 421.16 417.75 420.78 \n", "773 2022 6 44727 2022.4548 420.68 418.09 420.43 417.86 420.68 \n", "774 2022 7 44757 2022.5370 418.66 417.85 418.75 417.98 418.66 \n", "775 2022 8 44788 2022.6219 -99.99 -99.99 -99.99 -99.99 -99.99 \n", "776 2022 9 44819 2022.7068 -99.99 -99.99 -99.99 -99.99 -99.99 \n", "777 2022 10 44849 2022.7890 -99.99 -99.99 -99.99 -99.99 -99.99 \n", "778 2022 11 44880 2022.8740 -99.99 -99.99 -99.99 -99.99 -99.99 \n", "779 2022 12 44910 2022.9562 -99.99 -99.99 -99.99 -99.99 -99.99 \n", "\n", " SAFitFilled \n", "0 -99.99 \n", "1 -99.99 \n", "2 314.44 \n", "3 315.16 \n", "4 314.70 \n", "5 315.15 \n", "6 315.20 \n", "7 316.21 \n", "8 316.10 \n", "9 315.41 \n", "10 315.21 \n", "11 315.43 \n", "12 315.52 \n", "13 315.84 \n", "14 315.37 \n", "15 315.42 \n", "16 315.48 \n", "17 316.01 \n", "18 315.87 \n", "19 316.08 \n", "20 316.74 \n", "21 316.33 \n", "22 316.69 \n", "23 316.35 \n", "24 316.37 \n", "25 316.33 \n", "26 316.27 \n", "27 316.70 \n", "28 317.21 \n", "29 317.46 \n", ".. ... \n", "750 413.64 \n", "751 414.10 \n", "752 414.69 \n", "753 414.73 \n", "754 415.15 \n", "755 414.81 \n", "756 415.08 \n", "757 415.69 \n", "758 415.62 \n", "759 415.46 \n", "760 415.55 \n", "761 416.12 \n", "762 415.84 \n", "763 415.89 \n", "764 416.41 \n", "765 417.17 \n", "766 417.09 \n", "767 417.36 \n", "768 417.94 \n", "769 418.20 \n", "770 416.90 \n", "771 417.23 \n", "772 417.36 \n", "773 418.09 \n", "774 417.85 \n", "775 -99.99 \n", "776 -99.99 \n", "777 -99.99 \n", "778 -99.99 \n", "779 -99.99 \n", "\n", "[780 rows x 10 columns]" ] }, "execution_count": 78, "metadata": {}, "output_type": "execute_result" } ], "source": [ "raw_data = pd.read_csv(\"monthly_in_situ_co2_mlo.csv\", skiprows=56)\n", "c_names = raw_data.columns.values\n", "raw_data = raw_data.rename(columns = {c_names[0]:\"Yr\", c_names[1]:\"Mn\", c_names[2]:\"EDate\", c_names[3]:\"DDate\", c_names[4]:\"C02\", c_names[5]:\"SA\", c_names[6]:\"Fit\", c_names[7]:\"SAFit\", c_names[8]:\"Filled\", c_names[9]:\"SAFitFilled\"})\n", "raw_data" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "On notera que les valeurs manquantes sont notées `-99.99` dans le fichier, et que même les colonnes `Filled` contiennent ces valeurs manquantes pour les données de janvier et février 1958 et les mois les plus récents :" ] }, { "cell_type": "code", "execution_count": 79, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
YrMnEDateDDateC02SAFitSAFitFilledSAFitFilled
019581212001958.0411-99.99-99.99-99.99-99.99-99.99-99.99
119582212311958.1260-99.99-99.99-99.99-99.99-99.99-99.99
77520228447882022.6219-99.99-99.99-99.99-99.99-99.99-99.99
77620229448192022.7068-99.99-99.99-99.99-99.99-99.99-99.99
777202210448492022.7890-99.99-99.99-99.99-99.99-99.99-99.99
778202211448802022.8740-99.99-99.99-99.99-99.99-99.99-99.99
779202212449102022.9562-99.99-99.99-99.99-99.99-99.99-99.99
\n", "
" ], "text/plain": [ " Yr Mn EDate DDate C02 SA Fit SAFit Filled \\\n", "0 1958 1 21200 1958.0411 -99.99 -99.99 -99.99 -99.99 -99.99 \n", "1 1958 2 21231 1958.1260 -99.99 -99.99 -99.99 -99.99 -99.99 \n", "775 2022 8 44788 2022.6219 -99.99 -99.99 -99.99 -99.99 -99.99 \n", "776 2022 9 44819 2022.7068 -99.99 -99.99 -99.99 -99.99 -99.99 \n", "777 2022 10 44849 2022.7890 -99.99 -99.99 -99.99 -99.99 -99.99 \n", "778 2022 11 44880 2022.8740 -99.99 -99.99 -99.99 -99.99 -99.99 \n", "779 2022 12 44910 2022.9562 -99.99 -99.99 -99.99 -99.99 -99.99 \n", "\n", " SAFitFilled \n", "0 -99.99 \n", "1 -99.99 \n", "775 -99.99 \n", "776 -99.99 \n", "777 -99.99 \n", "778 -99.99 \n", "779 -99.99 " ] }, "execution_count": 79, "metadata": {}, "output_type": "execute_result" } ], "source": [ "missing_data = raw_data.iloc[lambda df: [(raw_data.at[row,'SAFitFilled'] == -99.99) for row in df.index]]\n", "missing_data" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "This missing data is removed from the dataframe." ] }, { "cell_type": "code", "execution_count": 80, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
YrMnEDateDDateC02SAFitSAFitFilledSAFitFilled
219583212591958.2027315.71314.44316.20314.91315.71314.44
319584212901958.2877317.45315.16317.30314.99317.45315.16
419585213201958.3699317.51314.70317.88315.07317.51314.70
519586213511958.4548-99.99-99.99317.26315.15317.26315.15
619587213811958.5370315.87315.20315.86315.22315.87315.20
719588214121958.6219314.93316.21313.98315.29314.93316.21
819589214431958.7068313.21316.10312.45315.36313.21316.10
9195810214731958.7890-99.99-99.99312.43315.41312.43315.41
10195811215041958.8740313.33315.21313.61315.46313.33315.21
11195812215341958.9562314.67315.43314.77315.52314.67315.43
1219591215651959.0411315.58315.52315.64315.58315.58315.52
1319592215961959.1260316.49315.84316.29315.64316.49315.84
1419593216241959.2027316.65315.37316.99315.70316.65315.37
1519594216551959.2877317.72315.42318.09315.77317.72315.42
1619595216851959.3699318.29315.48318.67315.85318.29315.48
1719596217161959.4548318.15316.01318.06315.94318.15316.01
1819597217461959.5370316.54315.87316.67316.03316.54315.87
1919598217771959.6219314.80316.08314.81316.13314.80316.08
2019599218081959.7068313.84316.74313.30316.22313.84316.74
21195910218381959.7890313.33316.33313.32316.31313.33316.33
22195911218691959.8740314.81316.69314.54316.40314.81316.69
23195912218991959.9562315.58316.35315.73316.48315.58316.35
2419601219301960.0410316.43316.37316.63316.56316.43316.37
2519602219611960.1257316.98316.33317.30316.64316.98316.33
2619603219901960.2049317.58316.27318.04316.72317.58316.27
2719604220211960.2896319.03316.70319.14316.80319.03316.70
2819605220511960.3716320.03317.21319.69316.87320.03317.21
2919606220821960.4563319.59317.46319.03316.93319.59317.46
3019607221121960.5383318.18317.53317.60316.98318.18317.53
3119608221431960.6230315.90317.22315.67317.02315.90317.22
.................................
74520202438762020.1257414.05413.27413.83413.04414.05413.27
74620203439052020.2049414.45412.88414.82413.23414.45412.88
74720204439362020.2896416.11413.31416.25413.43416.11413.31
74820205439662020.3716417.15413.76417.02413.63417.15413.76
74920206439972020.4563416.29413.74416.36413.83416.29413.74
75020207440272020.5383414.42413.64414.77414.03414.42413.64
75120208440582020.6230412.52414.10412.61414.23412.52414.10
75220209440892020.7077411.18414.69410.90414.43411.18414.69
753202010441192020.7896411.12414.73411.02414.62411.12414.73
754202011441502020.8743412.88415.15412.57414.81412.88415.15
755202012441802020.9563413.89414.81414.08414.99413.89414.81
75620211442112021.0411415.15415.08415.25415.16415.15415.08
75720212442422021.1260416.47415.69416.13415.34416.47415.69
75820213442702021.2027417.16415.62417.06415.50417.16415.62
75920214443012021.2877418.24415.46418.47415.67418.24415.46
76020215443312021.3699418.95415.55419.24415.84418.95415.55
76120216443622021.4548418.70416.12418.58416.02418.70416.12
76220217443922021.5370416.65415.84416.97416.20416.65415.84
76320218444232021.6219414.34415.89414.79416.39414.34415.89
76420219444542021.7068412.90416.41413.05416.58412.90416.41
765202110444842021.7890413.55417.17413.15416.76413.55417.17
766202111445152021.8740414.82417.09414.70416.94414.82417.09
767202112445452021.9562416.43417.36416.20417.11416.43417.36
76820221445762022.0411418.01417.94417.35417.26418.01417.94
76920222446072022.1260418.99418.20418.20417.40418.99418.20
77020223446352022.2027418.45416.90419.08417.51418.45416.90
77120224446662022.2877420.02417.23420.44417.63420.02417.23
77220225446962022.3699420.78417.36421.16417.75420.78417.36
77320226447272022.4548420.68418.09420.43417.86420.68418.09
77420227447572022.5370418.66417.85418.75417.98418.66417.85
\n", "

773 rows × 10 columns

\n", "
" ], "text/plain": [ " Yr Mn EDate DDate C02 SA Fit SAFit Filled \\\n", "2 1958 3 21259 1958.2027 315.71 314.44 316.20 314.91 315.71 \n", "3 1958 4 21290 1958.2877 317.45 315.16 317.30 314.99 317.45 \n", "4 1958 5 21320 1958.3699 317.51 314.70 317.88 315.07 317.51 \n", "5 1958 6 21351 1958.4548 -99.99 -99.99 317.26 315.15 317.26 \n", "6 1958 7 21381 1958.5370 315.87 315.20 315.86 315.22 315.87 \n", "7 1958 8 21412 1958.6219 314.93 316.21 313.98 315.29 314.93 \n", "8 1958 9 21443 1958.7068 313.21 316.10 312.45 315.36 313.21 \n", "9 1958 10 21473 1958.7890 -99.99 -99.99 312.43 315.41 312.43 \n", "10 1958 11 21504 1958.8740 313.33 315.21 313.61 315.46 313.33 \n", "11 1958 12 21534 1958.9562 314.67 315.43 314.77 315.52 314.67 \n", "12 1959 1 21565 1959.0411 315.58 315.52 315.64 315.58 315.58 \n", "13 1959 2 21596 1959.1260 316.49 315.84 316.29 315.64 316.49 \n", "14 1959 3 21624 1959.2027 316.65 315.37 316.99 315.70 316.65 \n", "15 1959 4 21655 1959.2877 317.72 315.42 318.09 315.77 317.72 \n", "16 1959 5 21685 1959.3699 318.29 315.48 318.67 315.85 318.29 \n", "17 1959 6 21716 1959.4548 318.15 316.01 318.06 315.94 318.15 \n", "18 1959 7 21746 1959.5370 316.54 315.87 316.67 316.03 316.54 \n", "19 1959 8 21777 1959.6219 314.80 316.08 314.81 316.13 314.80 \n", "20 1959 9 21808 1959.7068 313.84 316.74 313.30 316.22 313.84 \n", "21 1959 10 21838 1959.7890 313.33 316.33 313.32 316.31 313.33 \n", "22 1959 11 21869 1959.8740 314.81 316.69 314.54 316.40 314.81 \n", "23 1959 12 21899 1959.9562 315.58 316.35 315.73 316.48 315.58 \n", "24 1960 1 21930 1960.0410 316.43 316.37 316.63 316.56 316.43 \n", "25 1960 2 21961 1960.1257 316.98 316.33 317.30 316.64 316.98 \n", "26 1960 3 21990 1960.2049 317.58 316.27 318.04 316.72 317.58 \n", "27 1960 4 22021 1960.2896 319.03 316.70 319.14 316.80 319.03 \n", "28 1960 5 22051 1960.3716 320.03 317.21 319.69 316.87 320.03 \n", "29 1960 6 22082 1960.4563 319.59 317.46 319.03 316.93 319.59 \n", "30 1960 7 22112 1960.5383 318.18 317.53 317.60 316.98 318.18 \n", "31 1960 8 22143 1960.6230 315.90 317.22 315.67 317.02 315.90 \n", ".. ... .. ... ... ... ... ... ... ... \n", "745 2020 2 43876 2020.1257 414.05 413.27 413.83 413.04 414.05 \n", "746 2020 3 43905 2020.2049 414.45 412.88 414.82 413.23 414.45 \n", "747 2020 4 43936 2020.2896 416.11 413.31 416.25 413.43 416.11 \n", "748 2020 5 43966 2020.3716 417.15 413.76 417.02 413.63 417.15 \n", "749 2020 6 43997 2020.4563 416.29 413.74 416.36 413.83 416.29 \n", "750 2020 7 44027 2020.5383 414.42 413.64 414.77 414.03 414.42 \n", "751 2020 8 44058 2020.6230 412.52 414.10 412.61 414.23 412.52 \n", "752 2020 9 44089 2020.7077 411.18 414.69 410.90 414.43 411.18 \n", "753 2020 10 44119 2020.7896 411.12 414.73 411.02 414.62 411.12 \n", "754 2020 11 44150 2020.8743 412.88 415.15 412.57 414.81 412.88 \n", "755 2020 12 44180 2020.9563 413.89 414.81 414.08 414.99 413.89 \n", "756 2021 1 44211 2021.0411 415.15 415.08 415.25 415.16 415.15 \n", "757 2021 2 44242 2021.1260 416.47 415.69 416.13 415.34 416.47 \n", "758 2021 3 44270 2021.2027 417.16 415.62 417.06 415.50 417.16 \n", "759 2021 4 44301 2021.2877 418.24 415.46 418.47 415.67 418.24 \n", "760 2021 5 44331 2021.3699 418.95 415.55 419.24 415.84 418.95 \n", "761 2021 6 44362 2021.4548 418.70 416.12 418.58 416.02 418.70 \n", "762 2021 7 44392 2021.5370 416.65 415.84 416.97 416.20 416.65 \n", "763 2021 8 44423 2021.6219 414.34 415.89 414.79 416.39 414.34 \n", "764 2021 9 44454 2021.7068 412.90 416.41 413.05 416.58 412.90 \n", "765 2021 10 44484 2021.7890 413.55 417.17 413.15 416.76 413.55 \n", "766 2021 11 44515 2021.8740 414.82 417.09 414.70 416.94 414.82 \n", "767 2021 12 44545 2021.9562 416.43 417.36 416.20 417.11 416.43 \n", "768 2022 1 44576 2022.0411 418.01 417.94 417.35 417.26 418.01 \n", "769 2022 2 44607 2022.1260 418.99 418.20 418.20 417.40 418.99 \n", "770 2022 3 44635 2022.2027 418.45 416.90 419.08 417.51 418.45 \n", "771 2022 4 44666 2022.2877 420.02 417.23 420.44 417.63 420.02 \n", "772 2022 5 44696 2022.3699 420.78 417.36 421.16 417.75 420.78 \n", "773 2022 6 44727 2022.4548 420.68 418.09 420.43 417.86 420.68 \n", "774 2022 7 44757 2022.5370 418.66 417.85 418.75 417.98 418.66 \n", "\n", " SAFitFilled \n", "2 314.44 \n", "3 315.16 \n", "4 314.70 \n", "5 315.15 \n", "6 315.20 \n", "7 316.21 \n", "8 316.10 \n", "9 315.41 \n", "10 315.21 \n", "11 315.43 \n", "12 315.52 \n", "13 315.84 \n", "14 315.37 \n", "15 315.42 \n", "16 315.48 \n", "17 316.01 \n", "18 315.87 \n", "19 316.08 \n", "20 316.74 \n", "21 316.33 \n", "22 316.69 \n", "23 316.35 \n", "24 316.37 \n", "25 316.33 \n", "26 316.27 \n", "27 316.70 \n", "28 317.21 \n", "29 317.46 \n", "30 317.53 \n", "31 317.22 \n", ".. ... \n", "745 413.27 \n", "746 412.88 \n", "747 413.31 \n", "748 413.76 \n", "749 413.74 \n", "750 413.64 \n", "751 414.10 \n", "752 414.69 \n", "753 414.73 \n", "754 415.15 \n", "755 414.81 \n", "756 415.08 \n", "757 415.69 \n", "758 415.62 \n", "759 415.46 \n", "760 415.55 \n", "761 416.12 \n", "762 415.84 \n", "763 415.89 \n", "764 416.41 \n", "765 417.17 \n", "766 417.09 \n", "767 417.36 \n", "768 417.94 \n", "769 418.20 \n", "770 416.90 \n", "771 417.23 \n", "772 417.36 \n", "773 418.09 \n", "774 417.85 \n", "\n", "[773 rows x 10 columns]" ] }, "execution_count": 80, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data = raw_data.drop(missing_data.index).copy()\n", "data" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Nous indexons les données par le mois durant lequel elles ont été relevées. Nous remarquons qu'elles sont déjà ordonnées par ordre chronologique." ] }, { "cell_type": "code", "execution_count": 81, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
YrMnEDateDDateC02SAFitSAFitFilledSAFitFilled
1958-0319583212591958.2027315.71314.44316.20314.91315.71314.44
1958-0419584212901958.2877317.45315.16317.30314.99317.45315.16
1958-0519585213201958.3699317.51314.70317.88315.07317.51314.70
1958-0619586213511958.4548-99.99-99.99317.26315.15317.26315.15
1958-0719587213811958.5370315.87315.20315.86315.22315.87315.20
1958-0819588214121958.6219314.93316.21313.98315.29314.93316.21
1958-0919589214431958.7068313.21316.10312.45315.36313.21316.10
1958-10195810214731958.7890-99.99-99.99312.43315.41312.43315.41
1958-11195811215041958.8740313.33315.21313.61315.46313.33315.21
1958-12195812215341958.9562314.67315.43314.77315.52314.67315.43
1959-0119591215651959.0411315.58315.52315.64315.58315.58315.52
1959-0219592215961959.1260316.49315.84316.29315.64316.49315.84
1959-0319593216241959.2027316.65315.37316.99315.70316.65315.37
1959-0419594216551959.2877317.72315.42318.09315.77317.72315.42
1959-0519595216851959.3699318.29315.48318.67315.85318.29315.48
1959-0619596217161959.4548318.15316.01318.06315.94318.15316.01
1959-0719597217461959.5370316.54315.87316.67316.03316.54315.87
1959-0819598217771959.6219314.80316.08314.81316.13314.80316.08
1959-0919599218081959.7068313.84316.74313.30316.22313.84316.74
1959-10195910218381959.7890313.33316.33313.32316.31313.33316.33
1959-11195911218691959.8740314.81316.69314.54316.40314.81316.69
1959-12195912218991959.9562315.58316.35315.73316.48315.58316.35
1960-0119601219301960.0410316.43316.37316.63316.56316.43316.37
1960-0219602219611960.1257316.98316.33317.30316.64316.98316.33
1960-0319603219901960.2049317.58316.27318.04316.72317.58316.27
1960-0419604220211960.2896319.03316.70319.14316.80319.03316.70
1960-0519605220511960.3716320.03317.21319.69316.87320.03317.21
1960-0619606220821960.4563319.59317.46319.03316.93319.59317.46
1960-0719607221121960.5383318.18317.53317.60316.98318.18317.53
1960-0819608221431960.6230315.90317.22315.67317.02315.90317.22
.................................
2020-0220202438762020.1257414.05413.27413.83413.04414.05413.27
2020-0320203439052020.2049414.45412.88414.82413.23414.45412.88
2020-0420204439362020.2896416.11413.31416.25413.43416.11413.31
2020-0520205439662020.3716417.15413.76417.02413.63417.15413.76
2020-0620206439972020.4563416.29413.74416.36413.83416.29413.74
2020-0720207440272020.5383414.42413.64414.77414.03414.42413.64
2020-0820208440582020.6230412.52414.10412.61414.23412.52414.10
2020-0920209440892020.7077411.18414.69410.90414.43411.18414.69
2020-10202010441192020.7896411.12414.73411.02414.62411.12414.73
2020-11202011441502020.8743412.88415.15412.57414.81412.88415.15
2020-12202012441802020.9563413.89414.81414.08414.99413.89414.81
2021-0120211442112021.0411415.15415.08415.25415.16415.15415.08
2021-0220212442422021.1260416.47415.69416.13415.34416.47415.69
2021-0320213442702021.2027417.16415.62417.06415.50417.16415.62
2021-0420214443012021.2877418.24415.46418.47415.67418.24415.46
2021-0520215443312021.3699418.95415.55419.24415.84418.95415.55
2021-0620216443622021.4548418.70416.12418.58416.02418.70416.12
2021-0720217443922021.5370416.65415.84416.97416.20416.65415.84
2021-0820218444232021.6219414.34415.89414.79416.39414.34415.89
2021-0920219444542021.7068412.90416.41413.05416.58412.90416.41
2021-10202110444842021.7890413.55417.17413.15416.76413.55417.17
2021-11202111445152021.8740414.82417.09414.70416.94414.82417.09
2021-12202112445452021.9562416.43417.36416.20417.11416.43417.36
2022-0120221445762022.0411418.01417.94417.35417.26418.01417.94
2022-0220222446072022.1260418.99418.20418.20417.40418.99418.20
2022-0320223446352022.2027418.45416.90419.08417.51418.45416.90
2022-0420224446662022.2877420.02417.23420.44417.63420.02417.23
2022-0520225446962022.3699420.78417.36421.16417.75420.78417.36
2022-0620226447272022.4548420.68418.09420.43417.86420.68418.09
2022-0720227447572022.5370418.66417.85418.75417.98418.66417.85
\n", "

773 rows × 10 columns

\n", "
" ], "text/plain": [ " Yr Mn EDate DDate C02 SA Fit SAFit Filled \\\n", "1958-03 1958 3 21259 1958.2027 315.71 314.44 316.20 314.91 315.71 \n", "1958-04 1958 4 21290 1958.2877 317.45 315.16 317.30 314.99 317.45 \n", "1958-05 1958 5 21320 1958.3699 317.51 314.70 317.88 315.07 317.51 \n", "1958-06 1958 6 21351 1958.4548 -99.99 -99.99 317.26 315.15 317.26 \n", "1958-07 1958 7 21381 1958.5370 315.87 315.20 315.86 315.22 315.87 \n", "1958-08 1958 8 21412 1958.6219 314.93 316.21 313.98 315.29 314.93 \n", "1958-09 1958 9 21443 1958.7068 313.21 316.10 312.45 315.36 313.21 \n", "1958-10 1958 10 21473 1958.7890 -99.99 -99.99 312.43 315.41 312.43 \n", "1958-11 1958 11 21504 1958.8740 313.33 315.21 313.61 315.46 313.33 \n", "1958-12 1958 12 21534 1958.9562 314.67 315.43 314.77 315.52 314.67 \n", "1959-01 1959 1 21565 1959.0411 315.58 315.52 315.64 315.58 315.58 \n", "1959-02 1959 2 21596 1959.1260 316.49 315.84 316.29 315.64 316.49 \n", "1959-03 1959 3 21624 1959.2027 316.65 315.37 316.99 315.70 316.65 \n", "1959-04 1959 4 21655 1959.2877 317.72 315.42 318.09 315.77 317.72 \n", "1959-05 1959 5 21685 1959.3699 318.29 315.48 318.67 315.85 318.29 \n", "1959-06 1959 6 21716 1959.4548 318.15 316.01 318.06 315.94 318.15 \n", "1959-07 1959 7 21746 1959.5370 316.54 315.87 316.67 316.03 316.54 \n", "1959-08 1959 8 21777 1959.6219 314.80 316.08 314.81 316.13 314.80 \n", "1959-09 1959 9 21808 1959.7068 313.84 316.74 313.30 316.22 313.84 \n", "1959-10 1959 10 21838 1959.7890 313.33 316.33 313.32 316.31 313.33 \n", "1959-11 1959 11 21869 1959.8740 314.81 316.69 314.54 316.40 314.81 \n", "1959-12 1959 12 21899 1959.9562 315.58 316.35 315.73 316.48 315.58 \n", "1960-01 1960 1 21930 1960.0410 316.43 316.37 316.63 316.56 316.43 \n", "1960-02 1960 2 21961 1960.1257 316.98 316.33 317.30 316.64 316.98 \n", "1960-03 1960 3 21990 1960.2049 317.58 316.27 318.04 316.72 317.58 \n", "1960-04 1960 4 22021 1960.2896 319.03 316.70 319.14 316.80 319.03 \n", "1960-05 1960 5 22051 1960.3716 320.03 317.21 319.69 316.87 320.03 \n", "1960-06 1960 6 22082 1960.4563 319.59 317.46 319.03 316.93 319.59 \n", "1960-07 1960 7 22112 1960.5383 318.18 317.53 317.60 316.98 318.18 \n", "1960-08 1960 8 22143 1960.6230 315.90 317.22 315.67 317.02 315.90 \n", "... ... .. ... ... ... ... ... ... ... \n", "2020-02 2020 2 43876 2020.1257 414.05 413.27 413.83 413.04 414.05 \n", "2020-03 2020 3 43905 2020.2049 414.45 412.88 414.82 413.23 414.45 \n", "2020-04 2020 4 43936 2020.2896 416.11 413.31 416.25 413.43 416.11 \n", "2020-05 2020 5 43966 2020.3716 417.15 413.76 417.02 413.63 417.15 \n", "2020-06 2020 6 43997 2020.4563 416.29 413.74 416.36 413.83 416.29 \n", "2020-07 2020 7 44027 2020.5383 414.42 413.64 414.77 414.03 414.42 \n", "2020-08 2020 8 44058 2020.6230 412.52 414.10 412.61 414.23 412.52 \n", "2020-09 2020 9 44089 2020.7077 411.18 414.69 410.90 414.43 411.18 \n", "2020-10 2020 10 44119 2020.7896 411.12 414.73 411.02 414.62 411.12 \n", "2020-11 2020 11 44150 2020.8743 412.88 415.15 412.57 414.81 412.88 \n", "2020-12 2020 12 44180 2020.9563 413.89 414.81 414.08 414.99 413.89 \n", "2021-01 2021 1 44211 2021.0411 415.15 415.08 415.25 415.16 415.15 \n", "2021-02 2021 2 44242 2021.1260 416.47 415.69 416.13 415.34 416.47 \n", "2021-03 2021 3 44270 2021.2027 417.16 415.62 417.06 415.50 417.16 \n", "2021-04 2021 4 44301 2021.2877 418.24 415.46 418.47 415.67 418.24 \n", "2021-05 2021 5 44331 2021.3699 418.95 415.55 419.24 415.84 418.95 \n", "2021-06 2021 6 44362 2021.4548 418.70 416.12 418.58 416.02 418.70 \n", "2021-07 2021 7 44392 2021.5370 416.65 415.84 416.97 416.20 416.65 \n", "2021-08 2021 8 44423 2021.6219 414.34 415.89 414.79 416.39 414.34 \n", "2021-09 2021 9 44454 2021.7068 412.90 416.41 413.05 416.58 412.90 \n", "2021-10 2021 10 44484 2021.7890 413.55 417.17 413.15 416.76 413.55 \n", "2021-11 2021 11 44515 2021.8740 414.82 417.09 414.70 416.94 414.82 \n", "2021-12 2021 12 44545 2021.9562 416.43 417.36 416.20 417.11 416.43 \n", "2022-01 2022 1 44576 2022.0411 418.01 417.94 417.35 417.26 418.01 \n", "2022-02 2022 2 44607 2022.1260 418.99 418.20 418.20 417.40 418.99 \n", "2022-03 2022 3 44635 2022.2027 418.45 416.90 419.08 417.51 418.45 \n", "2022-04 2022 4 44666 2022.2877 420.02 417.23 420.44 417.63 420.02 \n", "2022-05 2022 5 44696 2022.3699 420.78 417.36 421.16 417.75 420.78 \n", "2022-06 2022 6 44727 2022.4548 420.68 418.09 420.43 417.86 420.68 \n", "2022-07 2022 7 44757 2022.5370 418.66 417.85 418.75 417.98 418.66 \n", "\n", " SAFitFilled \n", "1958-03 314.44 \n", "1958-04 315.16 \n", "1958-05 314.70 \n", "1958-06 315.15 \n", "1958-07 315.20 \n", "1958-08 316.21 \n", "1958-09 316.10 \n", "1958-10 315.41 \n", "1958-11 315.21 \n", "1958-12 315.43 \n", "1959-01 315.52 \n", "1959-02 315.84 \n", "1959-03 315.37 \n", "1959-04 315.42 \n", "1959-05 315.48 \n", "1959-06 316.01 \n", "1959-07 315.87 \n", "1959-08 316.08 \n", "1959-09 316.74 \n", "1959-10 316.33 \n", "1959-11 316.69 \n", "1959-12 316.35 \n", "1960-01 316.37 \n", "1960-02 316.33 \n", "1960-03 316.27 \n", "1960-04 316.70 \n", "1960-05 317.21 \n", "1960-06 317.46 \n", "1960-07 317.53 \n", "1960-08 317.22 \n", "... ... \n", "2020-02 413.27 \n", "2020-03 412.88 \n", "2020-04 413.31 \n", "2020-05 413.76 \n", "2020-06 413.74 \n", "2020-07 413.64 \n", "2020-08 414.10 \n", "2020-09 414.69 \n", "2020-10 414.73 \n", "2020-11 415.15 \n", "2020-12 414.81 \n", "2021-01 415.08 \n", "2021-02 415.69 \n", "2021-03 415.62 \n", "2021-04 415.46 \n", "2021-05 415.55 \n", "2021-06 416.12 \n", "2021-07 415.84 \n", "2021-08 415.89 \n", "2021-09 416.41 \n", "2021-10 417.17 \n", "2021-11 417.09 \n", "2021-12 417.36 \n", "2022-01 417.94 \n", "2022-02 418.20 \n", "2022-03 416.90 \n", "2022-04 417.23 \n", "2022-05 417.36 \n", "2022-06 418.09 \n", "2022-07 417.85 \n", "\n", "[773 rows x 10 columns]" ] }, "execution_count": 81, "metadata": {}, "output_type": "execute_result" } ], "source": [ "def generate_period(yr,mn):\n", " return pd.Period(year = yr, month = mn, freq='M')\n", "periods = pd.PeriodIndex(data = [generate_period(data.at[index,'Yr'],data.at[index,'Mn']) for index in data.index])\n", "indexed_data = data.set_index(periods).copy()\n", "indexed_data" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Premières observations\n", "\n", "Nous utilisons les données complétées pour la suite de cette analyse (colonnes `Filled`).\n", "\n", "Voici tout d'abord une représentation de l'évolution de C02 atmosphérique depuis 1958 :" ] }, { "cell_type": "code", "execution_count": 82, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 82, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "indexed_data['Filled'].plot()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Nous observons une augmentation systématique de la concentration de C02 couplée à une oscillation périodique à plus petite échelle. Zoomons sur une partie du graphique pour observer ces oscillations plus précisément." ] }, { "cell_type": "code", "execution_count": 83, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 83, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "indexed_data[\"Filled\"][-30:].plot()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Nous observons des variations saisonales de la concentration de C02. En ôtant ces fluctuations saisonales des données, (colonne `SAFitFilled`), nous obtenons un graphe témoignant uniquement de l'augmentation systématique de la concentration de C02 de ces 60 dernières années." ] }, { "cell_type": "code", "execution_count": 84, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 84, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "indexed_data['SAFitFilled'].plot()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Superposons enfin ces deux types de données sur un même graphique, globalement puis en se restreignant à une période de quelques années." ] }, { "cell_type": "code", "execution_count": 85, "metadata": { "scrolled": true }, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 85, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "indexed_data.plot(y = [\"SAFitFilled\",\"Filled\"])" ] }, { "cell_type": "code", "execution_count": 86, "metadata": { "scrolled": true }, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 86, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "indexed_data[-30:].plot(y = [\"SAFitFilled\",\"Filled\"])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Fluctuations saisonales\n", "\n", "Nous ponvons isoler les fluctuations saisonales afin de les étudier indépendemment. En voici une représentation (sur une échelle de quelques années par souci de lisibilité)." ] }, { "cell_type": "code", "execution_count": 87, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 87, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "seasonal_fluctuations = [data.at[index,'Filled'] - data.at[index,'SAFitFilled'] for index in data.index]\n", "indexed_data['SeasonalOnly'] = seasonal_fluctuations\n", "indexed_data['SeasonalOnly'][-30:].plot()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Nous essayons de modéliser ces variations saisonnières par un signal triangulaire avec la fonction `curve_fit`, intialisée avec un signal triangulaire de période 12 (étant donné qu'il y a douze mois dans l'année) et d'amplitude 6 (par lecture graphique). Nous choisissons également une phase et un offset vertical correspondants à nos lectures graphiques. Enfin, nous affichons le résultat aux côtés des données initiales." ] }, { "cell_type": "code", "execution_count": 88, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 88, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "def signal_triangle(x, a, b, c, h):\n", " v = np.divide(x - h,b)\n", " return a * np.abs(v - np.floor(v + 0.5)) + c\n", "\n", "x_data = np.array(data.index)\n", "y_data = np.array(seasonal_fluctuations)\n", "\n", "params, _ = curve_fit(signal_triangle, x_data, y_data, p0 = [12, 12, -3, 21])\n", "\n", "fitA = params[0]\n", "fitB = params[1]\n", "fitC = params[2]\n", "fitH = params[3]\n", "\n", "indexed_data['ModelSF'] = signal_triangle(x_data, fitA, fitB, fitC, fitH)\n", "\n", "indexed_data['SeasonalOnly'][-30:].plot(style = 'o')\n", "indexed_data['ModelSF'][-30:].plot()\n", "plt.legend([\"Fluctuations saisonnières\", \"Modèle triangulaire\"])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Cette modélisation a ces limites. Tout d'abord, à chaque période, la concentration en C02 augmente plus lentement qu'elle ne redescend, ce que ce signal triangulaire ne reflète pas. Nous observons de plus que l'amplitude des variations saisonnières augmente systématiquement entre 1958 et 2022, ce qui n'est pas reflété par ce modèle d'amplitude constante.\n", "\n", "## Augmentation systématique\n", "\n", "Nous étudions à présent l'augmentation systématique de la concentration en C02 antmosphérique. Nous tentons de modéliser les données `SAFitFilled`, c'est-à-dire les données complétées et ajustées pour omettre les fluctuations saisonnières.\n", "\n", "Nous essayons des modèles linéaire, quadratique, et exponentiels pour ces données, donnés par les fonctions `linear`, `quadr`, et `expo`." ] }, { "cell_type": "code", "execution_count": 89, "metadata": {}, "outputs": [], "source": [ "def linear(x, a, b):\n", " return a * x + b\n", "\n", "def quadr(x, a, b, h):\n", " return a * np.square(x - h) + b\n", "\n", "def expo(x, a, b, c, h):\n", " return a * np.exp(c * (x - h)) + b" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Nous utilisons de nouveau la fonction `curve_fit` pour trouver quels paramètres de ces modèles correspondent au mieux à nos données." ] }, { "cell_type": "code", "execution_count": 90, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/opt/conda/lib/python3.6/site-packages/ipykernel_launcher.py:8: RuntimeWarning: overflow encountered in exp\n", " \n" ] }, { "data": { "text/plain": [ "" ] }, "execution_count": 90, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "x_data = np.array(data.index)\n", "y_data = np.array(data['SAFitFilled'])\n", "\n", "params_lin, extras_lin = curve_fit(linear, x_data, y_data)\n", "params_q, extras_q = curve_fit(quadr, x_data, y_data)\n", "params_e, extras_e = curve_fit(expo, x_data, y_data)\n", "\n", "fitA_lin, fitB_lin = params_lin[0], params_lin[1]\n", "fitA_q, fitB_q, fitH_q = params_q[0], params_q[1], params_q[2]\n", "fitA_e, fitB_e, fitC_e, fitH_e = params_e[0], params_e[1], params_e[2], params_e[3]\n", "\n", "indexed_data['ModelSA_lin'] = linear(x_data, fitA_lin, fitB_lin)\n", "indexed_data['ModelSA_q'] = quadr(x_data, fitA_q, fitB_q, fitH_q)\n", "indexed_data['ModelSA_e'] = expo(x_data, fitA_e, fitB_e, fitC_e, fitH_e)\n", "\n", "indexed_data['SAFitFilled'].plot()\n", "indexed_data['ModelSA_lin'].plot()\n", "indexed_data['ModelSA_q'].plot()\n", "plt.legend([\"Augmentation observée\", \"Modèle linéaire\", \"Modèle quadratique\"])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Nous représentons le modèle exponentiel calculé séparément par souci de lisibilité." ] }, { "cell_type": "code", "execution_count": 91, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 91, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "indexed_data['SAFitFilled'].plot()\n", "indexed_data['ModelSA_e'].plot()\n", "plt.legend([\"Augmentation observée\", \"Modèle exponentiel\"])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Le modèle quadratique semble correspondre assez précisément aux données, nous allons donc l'utiliser pour extrapoler l'évolution future de la concentration de C02 atmosphérique jusqu'à l'année 2050. Nous calculons le nombre de mois supplémentaires à générer par le modèle, et affichons le résultat." ] }, { "cell_type": "code", "execution_count": 92, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 92, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "first_year, first_month = data.iat[0,0], data.iat[0,1]\n", "months = range((2050 - first_year) * 12 + 12 - first_month)\n", "\n", "extrapolation_sys_augm = quadr(months, fitA_q, fitB_q, fitH_q)\n", "plt.plot(x_data, data['SAFitFilled'], '*', color = \"lightgreen\")\n", "plt.plot(months, extrapolation_sys_augm, '-', color = \"red\")\n", "plt.legend([\"Augmentation observée\", \"Extrapolation -- modèle quadratique\"])" ] } ], "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 }