{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Concentration de CO2 dans l'atmosphère depuis 1958" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Introduction\n", "Le but de cette étude est d'analyser l'évolution de la concentration en CO2 dans l'atmosphère, en mettant en pratique les outils de recherche reproductible." ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "%matplotlib inline\n", "import matplotlib.pyplot as plt\n", "import pandas as pd\n", "import isoweek\n", "import numpy as np" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": [ "#raw_data = pd.read_csv(\"https://scrippsco2.ucsd.edu/assets/data/atmospheric/stations/in_situ_co2/monthly/monthly_in_situ_co2_mlo.csv\", skiprows = 54, sep=r'\\s*,\\s*', engine='python')\n", "raw_data = pd.read_csv(\"monthly_in_situ_co2_mlo.csv\", skiprows = 54, sep=r'\\s*,\\s*', engine='python')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Les données ont été extraites le 11/05/2020. On travaillera avec une copie locale mais la ligne commentée permet le téléchargement des données à la source. \n", "Les 54 premières lignes correspondent à du texte contenant les références à citer, des explications sur la forme des données ... On les supprime donc pour permettre à Pandas de lire les données sous forme de tableau. " ] }, { "cell_type": "code", "execution_count": 3, "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", "
YrMnDateDate.1CO2seasonallyfitseasonally.1CO2.1seasonally.2
0NaNNaNNaNNaNNaNadjustedNaNadjusted fitfilledadjusted filled
1NaNNaNExcelNaN[ppm][ppm][ppm][ppm][ppm][ppm]
21958.01.0212001958.0411-99.99-99.99-99.99-99.99-99.99-99.99
31958.02.0212311958.1260-99.99-99.99-99.99-99.99-99.99-99.99
41958.03.0212591958.2027315.70314.44316.18314.90315.70314.44
51958.04.0212901958.2877317.46315.16317.29314.98317.46315.16
61958.05.0213201958.3699317.51314.71317.86315.06317.51314.71
71958.06.0213511958.4548-99.99-99.99317.24315.14317.24315.14
81958.07.0213811958.5370315.86315.19315.86315.21315.86315.19
91958.08.0214121958.6219314.93316.19313.99315.28314.93316.19
101958.09.0214431958.7068313.21316.08312.45315.35313.21316.08
111958.010.0214731958.7890-99.99-99.99312.43315.40312.43315.40
121958.011.0215041958.8740313.33315.20313.61315.46313.33315.20
131958.012.0215341958.9562314.67315.43314.76315.51314.67315.43
141959.01.0215651959.0411315.58315.54315.62315.57315.58315.54
151959.02.0215961959.1260316.49315.86316.27315.63316.49315.86
161959.03.0216241959.2027316.65315.38316.98315.69316.65315.38
171959.04.0216551959.2877317.72315.42318.09315.77317.72315.42
181959.05.0216851959.3699318.29315.49318.65315.85318.29315.49
191959.06.0217161959.4548318.15316.03318.04315.94318.15316.03
201959.07.0217461959.5370316.54315.86316.67316.03316.54315.86
211959.08.0217771959.6219314.80316.06314.82316.12314.80316.06
221959.09.0218081959.7068313.84316.73313.31316.22313.84316.73
231959.010.0218381959.7890313.33316.33313.32316.30313.33316.33
241959.011.0218691959.8740314.81316.68314.54316.39314.81316.68
251959.012.0218991959.9562315.58316.35315.72316.47315.58316.35
261960.01.0219301960.0410316.43316.39316.61316.56316.43316.39
271960.02.0219611960.1257316.98316.35317.27316.64316.98316.35
281960.03.0219901960.2049317.58316.28318.03316.71317.58316.28
291960.04.0220211960.2896319.03316.70319.14316.79319.03316.70
.................................
7282018.07.0432962018.5370408.90408.08409.44408.65408.90408.08
7292018.08.0433272018.6219407.10408.63407.34408.91407.10408.63
7302018.09.0433582018.7068405.59409.08405.67409.19405.59409.08
7312018.010.0433882018.7890405.99409.61405.85409.45405.99409.61
7322018.011.0434192018.8740408.12410.38407.49409.73408.12410.38
7332018.012.0434492018.9562409.23410.15409.08409.99409.23410.15
7342019.01.0434802019.0411410.92410.87410.31410.25410.92410.87
7352019.02.0435112019.1260411.66410.90411.26410.49411.66410.90
7362019.03.0435392019.2027412.00410.46412.26410.70412.00410.46
7372019.04.0435702019.2877413.52410.72413.75410.93413.52410.72
7382019.05.0436002019.3699414.83411.42414.55411.15414.83411.42
7392019.06.0436312019.4548413.96411.38413.92411.37413.96411.38
7402019.07.0436612019.5370411.85411.03412.37411.58411.85411.03
7412019.08.0436922019.6219410.08411.62410.23411.80410.08411.62
7422019.09.0437232019.7068408.55412.06408.50412.03408.55412.06
7432019.010.0437532019.7890408.43412.06408.63412.24408.43412.06
7442019.011.0437842019.8740410.29412.56410.22412.47410.29412.56
7452019.012.0438142019.9562411.85412.78411.77412.68411.85412.78
7462020.01.0438452020.0410413.37413.32412.96412.89413.37413.32
7472020.02.0438762020.1257414.09413.33413.87413.10414.09413.33
7482020.03.0439052020.2049414.51412.94414.88413.29414.51412.94
7492020.04.0439362020.2896416.18413.35-99.99-99.99416.18413.35
7502020.05.0439662020.3716-99.99-99.99-99.99-99.99-99.99-99.99
7512020.06.0439972020.4563-99.99-99.99-99.99-99.99-99.99-99.99
7522020.07.0440272020.5383-99.99-99.99-99.99-99.99-99.99-99.99
7532020.08.0440582020.6230-99.99-99.99-99.99-99.99-99.99-99.99
7542020.09.0440892020.7077-99.99-99.99-99.99-99.99-99.99-99.99
7552020.010.0441192020.7896-99.99-99.99-99.99-99.99-99.99-99.99
7562020.011.0441502020.8743-99.99-99.99-99.99-99.99-99.99-99.99
7572020.012.0441802020.9563-99.99-99.99-99.99-99.99-99.99-99.99
\n", "

758 rows × 10 columns

\n", "
" ], "text/plain": [ " Yr Mn Date Date.1 CO2 seasonally fit seasonally.1 \\\n", "0 NaN NaN NaN NaN NaN adjusted NaN adjusted fit \n", "1 NaN NaN Excel NaN [ppm] [ppm] [ppm] [ppm] \n", "2 1958.0 1.0 21200 1958.0411 -99.99 -99.99 -99.99 -99.99 \n", "3 1958.0 2.0 21231 1958.1260 -99.99 -99.99 -99.99 -99.99 \n", "4 1958.0 3.0 21259 1958.2027 315.70 314.44 316.18 314.90 \n", "5 1958.0 4.0 21290 1958.2877 317.46 315.16 317.29 314.98 \n", "6 1958.0 5.0 21320 1958.3699 317.51 314.71 317.86 315.06 \n", "7 1958.0 6.0 21351 1958.4548 -99.99 -99.99 317.24 315.14 \n", "8 1958.0 7.0 21381 1958.5370 315.86 315.19 315.86 315.21 \n", "9 1958.0 8.0 21412 1958.6219 314.93 316.19 313.99 315.28 \n", "10 1958.0 9.0 21443 1958.7068 313.21 316.08 312.45 315.35 \n", "11 1958.0 10.0 21473 1958.7890 -99.99 -99.99 312.43 315.40 \n", "12 1958.0 11.0 21504 1958.8740 313.33 315.20 313.61 315.46 \n", "13 1958.0 12.0 21534 1958.9562 314.67 315.43 314.76 315.51 \n", "14 1959.0 1.0 21565 1959.0411 315.58 315.54 315.62 315.57 \n", "15 1959.0 2.0 21596 1959.1260 316.49 315.86 316.27 315.63 \n", "16 1959.0 3.0 21624 1959.2027 316.65 315.38 316.98 315.69 \n", "17 1959.0 4.0 21655 1959.2877 317.72 315.42 318.09 315.77 \n", "18 1959.0 5.0 21685 1959.3699 318.29 315.49 318.65 315.85 \n", "19 1959.0 6.0 21716 1959.4548 318.15 316.03 318.04 315.94 \n", "20 1959.0 7.0 21746 1959.5370 316.54 315.86 316.67 316.03 \n", "21 1959.0 8.0 21777 1959.6219 314.80 316.06 314.82 316.12 \n", "22 1959.0 9.0 21808 1959.7068 313.84 316.73 313.31 316.22 \n", "23 1959.0 10.0 21838 1959.7890 313.33 316.33 313.32 316.30 \n", "24 1959.0 11.0 21869 1959.8740 314.81 316.68 314.54 316.39 \n", "25 1959.0 12.0 21899 1959.9562 315.58 316.35 315.72 316.47 \n", "26 1960.0 1.0 21930 1960.0410 316.43 316.39 316.61 316.56 \n", "27 1960.0 2.0 21961 1960.1257 316.98 316.35 317.27 316.64 \n", "28 1960.0 3.0 21990 1960.2049 317.58 316.28 318.03 316.71 \n", "29 1960.0 4.0 22021 1960.2896 319.03 316.70 319.14 316.79 \n", ".. ... ... ... ... ... ... ... ... \n", "728 2018.0 7.0 43296 2018.5370 408.90 408.08 409.44 408.65 \n", "729 2018.0 8.0 43327 2018.6219 407.10 408.63 407.34 408.91 \n", "730 2018.0 9.0 43358 2018.7068 405.59 409.08 405.67 409.19 \n", "731 2018.0 10.0 43388 2018.7890 405.99 409.61 405.85 409.45 \n", "732 2018.0 11.0 43419 2018.8740 408.12 410.38 407.49 409.73 \n", "733 2018.0 12.0 43449 2018.9562 409.23 410.15 409.08 409.99 \n", "734 2019.0 1.0 43480 2019.0411 410.92 410.87 410.31 410.25 \n", "735 2019.0 2.0 43511 2019.1260 411.66 410.90 411.26 410.49 \n", "736 2019.0 3.0 43539 2019.2027 412.00 410.46 412.26 410.70 \n", "737 2019.0 4.0 43570 2019.2877 413.52 410.72 413.75 410.93 \n", "738 2019.0 5.0 43600 2019.3699 414.83 411.42 414.55 411.15 \n", "739 2019.0 6.0 43631 2019.4548 413.96 411.38 413.92 411.37 \n", "740 2019.0 7.0 43661 2019.5370 411.85 411.03 412.37 411.58 \n", "741 2019.0 8.0 43692 2019.6219 410.08 411.62 410.23 411.80 \n", "742 2019.0 9.0 43723 2019.7068 408.55 412.06 408.50 412.03 \n", "743 2019.0 10.0 43753 2019.7890 408.43 412.06 408.63 412.24 \n", "744 2019.0 11.0 43784 2019.8740 410.29 412.56 410.22 412.47 \n", "745 2019.0 12.0 43814 2019.9562 411.85 412.78 411.77 412.68 \n", "746 2020.0 1.0 43845 2020.0410 413.37 413.32 412.96 412.89 \n", "747 2020.0 2.0 43876 2020.1257 414.09 413.33 413.87 413.10 \n", "748 2020.0 3.0 43905 2020.2049 414.51 412.94 414.88 413.29 \n", "749 2020.0 4.0 43936 2020.2896 416.18 413.35 -99.99 -99.99 \n", "750 2020.0 5.0 43966 2020.3716 -99.99 -99.99 -99.99 -99.99 \n", "751 2020.0 6.0 43997 2020.4563 -99.99 -99.99 -99.99 -99.99 \n", "752 2020.0 7.0 44027 2020.5383 -99.99 -99.99 -99.99 -99.99 \n", "753 2020.0 8.0 44058 2020.6230 -99.99 -99.99 -99.99 -99.99 \n", "754 2020.0 9.0 44089 2020.7077 -99.99 -99.99 -99.99 -99.99 \n", "755 2020.0 10.0 44119 2020.7896 -99.99 -99.99 -99.99 -99.99 \n", "756 2020.0 11.0 44150 2020.8743 -99.99 -99.99 -99.99 -99.99 \n", "757 2020.0 12.0 44180 2020.9563 -99.99 -99.99 -99.99 -99.99 \n", "\n", " CO2.1 seasonally.2 \n", "0 filled adjusted filled \n", "1 [ppm] [ppm] \n", "2 -99.99 -99.99 \n", "3 -99.99 -99.99 \n", "4 315.70 314.44 \n", "5 317.46 315.16 \n", "6 317.51 314.71 \n", "7 317.24 315.14 \n", "8 315.86 315.19 \n", "9 314.93 316.19 \n", "10 313.21 316.08 \n", "11 312.43 315.40 \n", "12 313.33 315.20 \n", "13 314.67 315.43 \n", "14 315.58 315.54 \n", "15 316.49 315.86 \n", "16 316.65 315.38 \n", "17 317.72 315.42 \n", "18 318.29 315.49 \n", "19 318.15 316.03 \n", "20 316.54 315.86 \n", "21 314.80 316.06 \n", "22 313.84 316.73 \n", "23 313.33 316.33 \n", "24 314.81 316.68 \n", "25 315.58 316.35 \n", "26 316.43 316.39 \n", "27 316.98 316.35 \n", "28 317.58 316.28 \n", "29 319.03 316.70 \n", ".. ... ... \n", "728 408.90 408.08 \n", "729 407.10 408.63 \n", "730 405.59 409.08 \n", "731 405.99 409.61 \n", "732 408.12 410.38 \n", "733 409.23 410.15 \n", "734 410.92 410.87 \n", "735 411.66 410.90 \n", "736 412.00 410.46 \n", "737 413.52 410.72 \n", "738 414.83 411.42 \n", "739 413.96 411.38 \n", "740 411.85 411.03 \n", "741 410.08 411.62 \n", "742 408.55 412.06 \n", "743 408.43 412.06 \n", "744 410.29 412.56 \n", "745 411.85 412.78 \n", "746 413.37 413.32 \n", "747 414.09 413.33 \n", "748 414.51 412.94 \n", "749 416.18 413.35 \n", "750 -99.99 -99.99 \n", "751 -99.99 -99.99 \n", "752 -99.99 -99.99 \n", "753 -99.99 -99.99 \n", "754 -99.99 -99.99 \n", "755 -99.99 -99.99 \n", "756 -99.99 -99.99 \n", "757 -99.99 -99.99 \n", "\n", "[758 rows x 10 columns]" ] }, "execution_count": 3, "metadata": {}, "output_type": "execute_result" } ], "source": [ "raw_data" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Les deux premières lignes contiennent des unités et non des valeurs, on les retire du tableau pour l'instant." ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [], "source": [ "data = raw_data.iloc[2:]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Pour ce jeu de données, les 4 premières colonnes sont des dates, et seule la colonne 5 contient des mesures brutes. Nous allons conserver uniquement les informations sur l'année, le mois, et la valeur brute de la mesure." ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [], "source": [ "useful_data = data.iloc[0:len(data.index), [0,1,4]]\n", "#useful_data" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "On vérifie que les données ont un type approprié." ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ " 1958.0\n", " 2.0\n", " -99.99\n" ] } ], "source": [ "print(type(useful_data['Yr'][3]), useful_data['Yr'][3])\n", "print(type(useful_data['Mn'][3]), useful_data['Mn'][3])\n", "print(type(useful_data['CO2'][3]), useful_data['CO2'][3])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "On voit que la troisième colonne n'est pas bien interprétée, peut être à cause du signe '-'. On essaye de convertir les données." ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [], "source": [ "useful_data['CO2'] = useful_data['CO2'].astype(float)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Les explications jointes au fichier indiquent que les valeurs manquantes sont remplacées par la valeur -99.99. On souhaite donc supprimer chaque ligne comportant cette valeur." ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "[2, 3, 7, 11, 75, 76, 77, 750, 751, 752, 753, 754, 755, 756, 757]\n" ] } ], "source": [ "liste = []\n", "for i in range(len(useful_data.index)):\n", " try:\n", " if(useful_data['CO2'][useful_data.index[i]] == -99.99):\n", " liste.append(useful_data.index[i])\n", " except:\n", " print(i, ' ', end='')\n", "print(liste)\n", "useful_data.drop(liste, inplace=True)" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [], "source": [ "useful_data\n", "useful_data_copie = pd.DataFrame.copy(useful_data, deep = True)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "On souhaite maintenant convertir l'année et le mois en un format plus adapté à Pandas, et à l'utiliser comme index. Un méthode possible est présentée ici, en rassemblant les deux informations puis en appliquant une fonction pour une mise au format Pandas." ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [], "source": [ "useful_data['period'] = useful_data['Yr']*100 + useful_data['Mn']" ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [], "source": [ "useful_data['period'] = useful_data['period'].astype(int)" ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [], "source": [ "useful_data = useful_data.iloc[0:len(useful_data.index), [2,3]]" ] }, { "cell_type": "code", "execution_count": 13, "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", "
CO2
period
1958-03315.70
1958-04317.46
1958-05317.51
1958-07315.86
1958-08314.93
1958-09313.21
1958-11313.33
1958-12314.67
1959-01315.58
1959-02316.49
1959-03316.65
1959-04317.72
1959-05318.29
1959-06318.15
1959-07316.54
1959-08314.80
1959-09313.84
1959-10313.33
1959-11314.81
1959-12315.58
1960-01316.43
1960-02316.98
1960-03317.58
1960-04319.03
1960-05320.04
1960-06319.58
1960-07318.18
1960-08315.90
1960-09314.17
1960-10313.83
......
2017-11405.17
2017-12406.75
2018-01408.05
2018-02408.34
2018-03409.25
2018-04410.30
2018-05411.30
2018-06410.88
2018-07408.90
2018-08407.10
2018-09405.59
2018-10405.99
2018-11408.12
2018-12409.23
2019-01410.92
2019-02411.66
2019-03412.00
2019-04413.52
2019-05414.83
2019-06413.96
2019-07411.85
2019-08410.08
2019-09408.55
2019-10408.43
2019-11410.29
2019-12411.85
2020-01413.37
2020-02414.09
2020-03414.51
2020-04416.18
\n", "

741 rows × 1 columns

\n", "
" ], "text/plain": [ " CO2\n", "period \n", "1958-03 315.70\n", "1958-04 317.46\n", "1958-05 317.51\n", "1958-07 315.86\n", "1958-08 314.93\n", "1958-09 313.21\n", "1958-11 313.33\n", "1958-12 314.67\n", "1959-01 315.58\n", "1959-02 316.49\n", "1959-03 316.65\n", "1959-04 317.72\n", "1959-05 318.29\n", "1959-06 318.15\n", "1959-07 316.54\n", "1959-08 314.80\n", "1959-09 313.84\n", "1959-10 313.33\n", "1959-11 314.81\n", "1959-12 315.58\n", "1960-01 316.43\n", "1960-02 316.98\n", "1960-03 317.58\n", "1960-04 319.03\n", "1960-05 320.04\n", "1960-06 319.58\n", "1960-07 318.18\n", "1960-08 315.90\n", "1960-09 314.17\n", "1960-10 313.83\n", "... ...\n", "2017-11 405.17\n", "2017-12 406.75\n", "2018-01 408.05\n", "2018-02 408.34\n", "2018-03 409.25\n", "2018-04 410.30\n", "2018-05 411.30\n", "2018-06 410.88\n", "2018-07 408.90\n", "2018-08 407.10\n", "2018-09 405.59\n", "2018-10 405.99\n", "2018-11 408.12\n", "2018-12 409.23\n", "2019-01 410.92\n", "2019-02 411.66\n", "2019-03 412.00\n", "2019-04 413.52\n", "2019-05 414.83\n", "2019-06 413.96\n", "2019-07 411.85\n", "2019-08 410.08\n", "2019-09 408.55\n", "2019-10 408.43\n", "2019-11 410.29\n", "2019-12 411.85\n", "2020-01 413.37\n", "2020-02 414.09\n", "2020-03 414.51\n", "2020-04 416.18\n", "\n", "[741 rows x 1 columns]" ] }, "execution_count": 13, "metadata": {}, "output_type": "execute_result" } ], "source": [ "def convertIntoPeriod(anneeEtMois):\n", " y = (int)(anneeEtMois/100)\n", " m = (int)(anneeEtMois%100)\n", " return pd.Period(pd.Timestamp(y,m,1), 'M')\n", "useful_data['period'] = [convertIntoPeriod(date) for date in useful_data['period']]\n", "useful_data.set_index('period')" ] }, { "cell_type": "code", "execution_count": 14, "metadata": { "scrolled": true }, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 14, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "useful_data['CO2'].plot()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "On a un premier aperçu de nos données, mais l'échelle ne correspond pas à ce que nous voulons. De plus, il va être difficile avec des données manquantes de travailler proprement avec ces indices. On va donc repartir d'une copie de useful_data, et renseigner la date sous la forme du nombre de mois en partant de l'an 1958. Janvier 1959 sera donc référencé par \"13\", etc." ] }, { "cell_type": "code", "execution_count": 15, "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", "
YrMnCO2IndexMois
41958.03.0315.703
51958.04.0317.464
61958.05.0317.515
81958.07.0315.867
91958.08.0314.938
101958.09.0313.219
121958.011.0313.3311
131958.012.0314.6712
141959.01.0315.5813
151959.02.0316.4914
161959.03.0316.6515
171959.04.0317.7216
181959.05.0318.2917
191959.06.0318.1518
201959.07.0316.5419
211959.08.0314.8020
221959.09.0313.8421
231959.010.0313.3322
241959.011.0314.8123
251959.012.0315.5824
261960.01.0316.4325
271960.02.0316.9826
281960.03.0317.5827
291960.04.0319.0328
301960.05.0320.0429
311960.06.0319.5830
321960.07.0318.1831
331960.08.0315.9032
341960.09.0314.1733
351960.010.0313.8334
...............
7202017.011.0405.17719
7212017.012.0406.75720
7222018.01.0408.05721
7232018.02.0408.34722
7242018.03.0409.25723
7252018.04.0410.30724
7262018.05.0411.30725
7272018.06.0410.88726
7282018.07.0408.90727
7292018.08.0407.10728
7302018.09.0405.59729
7312018.010.0405.99730
7322018.011.0408.12731
7332018.012.0409.23732
7342019.01.0410.92733
7352019.02.0411.66734
7362019.03.0412.00735
7372019.04.0413.52736
7382019.05.0414.83737
7392019.06.0413.96738
7402019.07.0411.85739
7412019.08.0410.08740
7422019.09.0408.55741
7432019.010.0408.43742
7442019.011.0410.29743
7452019.012.0411.85744
7462020.01.0413.37745
7472020.02.0414.09746
7482020.03.0414.51747
7492020.04.0416.18748
\n", "

741 rows × 4 columns

\n", "
" ], "text/plain": [ " Yr Mn CO2 IndexMois\n", "4 1958.0 3.0 315.70 3\n", "5 1958.0 4.0 317.46 4\n", "6 1958.0 5.0 317.51 5\n", "8 1958.0 7.0 315.86 7\n", "9 1958.0 8.0 314.93 8\n", "10 1958.0 9.0 313.21 9\n", "12 1958.0 11.0 313.33 11\n", "13 1958.0 12.0 314.67 12\n", "14 1959.0 1.0 315.58 13\n", "15 1959.0 2.0 316.49 14\n", "16 1959.0 3.0 316.65 15\n", "17 1959.0 4.0 317.72 16\n", "18 1959.0 5.0 318.29 17\n", "19 1959.0 6.0 318.15 18\n", "20 1959.0 7.0 316.54 19\n", "21 1959.0 8.0 314.80 20\n", "22 1959.0 9.0 313.84 21\n", "23 1959.0 10.0 313.33 22\n", "24 1959.0 11.0 314.81 23\n", "25 1959.0 12.0 315.58 24\n", "26 1960.0 1.0 316.43 25\n", "27 1960.0 2.0 316.98 26\n", "28 1960.0 3.0 317.58 27\n", "29 1960.0 4.0 319.03 28\n", "30 1960.0 5.0 320.04 29\n", "31 1960.0 6.0 319.58 30\n", "32 1960.0 7.0 318.18 31\n", "33 1960.0 8.0 315.90 32\n", "34 1960.0 9.0 314.17 33\n", "35 1960.0 10.0 313.83 34\n", ".. ... ... ... ...\n", "720 2017.0 11.0 405.17 719\n", "721 2017.0 12.0 406.75 720\n", "722 2018.0 1.0 408.05 721\n", "723 2018.0 2.0 408.34 722\n", "724 2018.0 3.0 409.25 723\n", "725 2018.0 4.0 410.30 724\n", "726 2018.0 5.0 411.30 725\n", "727 2018.0 6.0 410.88 726\n", "728 2018.0 7.0 408.90 727\n", "729 2018.0 8.0 407.10 728\n", "730 2018.0 9.0 405.59 729\n", "731 2018.0 10.0 405.99 730\n", "732 2018.0 11.0 408.12 731\n", "733 2018.0 12.0 409.23 732\n", "734 2019.0 1.0 410.92 733\n", "735 2019.0 2.0 411.66 734\n", "736 2019.0 3.0 412.00 735\n", "737 2019.0 4.0 413.52 736\n", "738 2019.0 5.0 414.83 737\n", "739 2019.0 6.0 413.96 738\n", "740 2019.0 7.0 411.85 739\n", "741 2019.0 8.0 410.08 740\n", "742 2019.0 9.0 408.55 741\n", "743 2019.0 10.0 408.43 742\n", "744 2019.0 11.0 410.29 743\n", "745 2019.0 12.0 411.85 744\n", "746 2020.0 1.0 413.37 745\n", "747 2020.0 2.0 414.09 746\n", "748 2020.0 3.0 414.51 747\n", "749 2020.0 4.0 416.18 748\n", "\n", "[741 rows x 4 columns]" ] }, "execution_count": 15, "metadata": {}, "output_type": "execute_result" } ], "source": [ "udc = useful_data_copie\n", "udc['IndexMois'] = [(int)(udc['Mn'][x] + (udc['Yr'][x] - 1958)*12) for x in udc.index]\n", "udc" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "On vérifie à l'aide de la dernière valeur que tout est correct :" ] }, { "cell_type": "code", "execution_count": 16, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "748" ] }, "execution_count": 16, "metadata": {}, "output_type": "execute_result" } ], "source": [ "testIndex = (2020 - 1958)*12 + 4\n", "testIndex" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "On utilise notre nouvelle colonne comme index et on supprime les autres." ] }, { "cell_type": "code", "execution_count": 17, "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", "
CO2IndexMois
4315.703
5317.464
6317.515
8315.867
9314.938
10313.219
12313.3311
13314.6712
14315.5813
15316.4914
16316.6515
17317.7216
18318.2917
19318.1518
20316.5419
21314.8020
22313.8421
23313.3322
24314.8123
25315.5824
26316.4325
27316.9826
28317.5827
29319.0328
30320.0429
31319.5830
32318.1831
33315.9032
34314.1733
35313.8334
.........
720405.17719
721406.75720
722408.05721
723408.34722
724409.25723
725410.30724
726411.30725
727410.88726
728408.90727
729407.10728
730405.59729
731405.99730
732408.12731
733409.23732
734410.92733
735411.66734
736412.00735
737413.52736
738414.83737
739413.96738
740411.85739
741410.08740
742408.55741
743408.43742
744410.29743
745411.85744
746413.37745
747414.09746
748414.51747
749416.18748
\n", "

741 rows × 2 columns

\n", "
" ], "text/plain": [ " CO2 IndexMois\n", "4 315.70 3\n", "5 317.46 4\n", "6 317.51 5\n", "8 315.86 7\n", "9 314.93 8\n", "10 313.21 9\n", "12 313.33 11\n", "13 314.67 12\n", "14 315.58 13\n", "15 316.49 14\n", "16 316.65 15\n", "17 317.72 16\n", "18 318.29 17\n", "19 318.15 18\n", "20 316.54 19\n", "21 314.80 20\n", "22 313.84 21\n", "23 313.33 22\n", "24 314.81 23\n", "25 315.58 24\n", "26 316.43 25\n", "27 316.98 26\n", "28 317.58 27\n", "29 319.03 28\n", "30 320.04 29\n", "31 319.58 30\n", "32 318.18 31\n", "33 315.90 32\n", "34 314.17 33\n", "35 313.83 34\n", ".. ... ...\n", "720 405.17 719\n", "721 406.75 720\n", "722 408.05 721\n", "723 408.34 722\n", "724 409.25 723\n", "725 410.30 724\n", "726 411.30 725\n", "727 410.88 726\n", "728 408.90 727\n", "729 407.10 728\n", "730 405.59 729\n", "731 405.99 730\n", "732 408.12 731\n", "733 409.23 732\n", "734 410.92 733\n", "735 411.66 734\n", "736 412.00 735\n", "737 413.52 736\n", "738 414.83 737\n", "739 413.96 738\n", "740 411.85 739\n", "741 410.08 740\n", "742 408.55 741\n", "743 408.43 742\n", "744 410.29 743\n", "745 411.85 744\n", "746 413.37 745\n", "747 414.09 746\n", "748 414.51 747\n", "749 416.18 748\n", "\n", "[741 rows x 2 columns]" ] }, "execution_count": 17, "metadata": {}, "output_type": "execute_result" } ], "source": [ "del udc['Yr']\n", "del udc['Mn']\n", "udc.reset_index()\n", "udc.set_index('IndexMois')\n", "udc" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Pour une raison quelconque Pandas refuse d'indexer correctement le tableau ... Tant pis. On utilisera la colonne IndexMois en guise d'abscisses pour les plots." ] }, { "cell_type": "code", "execution_count": 18, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[]" ] }, "execution_count": 18, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "plt.plot(udc['IndexMois'], udc['CO2'])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "On vérifie la façon dont les données manquantes sont gérées." ] }, { "cell_type": "code", "execution_count": 19, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[]" ] }, "execution_count": 19, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "plt.plot(udc['IndexMois'][0:10], udc['CO2'][0:10])" ] }, { "cell_type": "code", "execution_count": 20, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "4 3\n", "5 4\n", "6 5\n", "8 7\n", "9 8\n", "10 9\n", "12 11\n", "13 12\n", "14 13\n", "15 14\n", "Name: IndexMois, dtype: int64 4 315.70\n", "5 317.46\n", "6 317.51\n", "8 315.86\n", "9 314.93\n", "10 313.21\n", "12 313.33\n", "13 314.67\n", "14 315.58\n", "15 316.49\n", "Name: CO2, dtype: float64\n" ] } ], "source": [ "print(udc['IndexMois'][0:10], udc['CO2'][0:10])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "On voit que la valeur 6 en abscisse n'a pas d'ordonnée, et que la droite est tracée entre les points 5 et 7. Il n'y a pas de problème. " ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "On peut s'intéresser maintenant aux résultats. On voit une croissance globale, et des oscillations locales. On peut zoomer sur trois années pour voir ce qu'il se passe localement par exemple." ] }, { "cell_type": "code", "execution_count": 21, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[]" ] }, "execution_count": 21, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "plt.plot(udc['IndexMois'][24:60], udc['CO2'][24:60])" ] }, { "cell_type": "code", "execution_count": 22, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "29 5 / 30 6 / 31 7 / 32 8 / 33 9 / 34 10 / 35 11 / 36 0 / 37 1 / 38 2 / 39 3 / 40 4 / 41 5 / 42 6 / 43 7 / 44 8 / 45 9 / 46 10 / 47 11 / 48 0 / 49 1 / 50 2 / 51 3 / 52 4 / 53 5 / 54 6 / 55 7 / 56 8 / 57 9 / 58 10 / 59 11 / 60 0 / " ] } ], "source": [ "for i in range(29,61):\n", " print(i, i%12, end=' / ')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "On voit des minima locaux aux abscisses 34, 45, 58 qui correspondent aux mois Octobre, Septembre, Octobre. Il semble donc que la concentration en CO2 soit périodiquement minimale à cette période de l'année. De même, on voit des maxima locaux aux abscisses 41 et 54, soit en mai et juin. On peut faire une autre vérification par précaution." ] }, { "cell_type": "code", "execution_count": 23, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "400 4 / 401 5 / 402 6 / 403 7 / 404 8 / 405 9 / 406 10 / 407 11 / 408 0 / 409 1 / 410 2 / 411 3 / 412 4 / 413 5 / 414 6 / 415 7 / 416 8 / 417 9 / 418 10 / 419 11 / 420 0 / 421 1 / 422 2 / 423 3 / 424 4 / 425 5 / 426 6 / 427 7 / 428 8 / 429 9 / 430 10 / 431 11 / 432 0 / 433 1 / 434 2 / 435 3 / 436 4 / " ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "plt.plot(udc['IndexMois'][400:436], udc['CO2'][400:436])\n", "for i in range(400,437):\n", " print(i, i%12, end=' / ')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "On voit les maxima en 413 (mai), 425 (mai), 437(mai) et des les minima en 417 (septembre), 429 (septembre), 441 (septembre). L'hypothèse se tient." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Pour caractériser la croissance et faire des prévisions pour les années à venir, on souhaite joindre une courbe de tendance et son équation à ces données. On s'intéressera juste aux moyennes annuelles ici. \n", "On présente ici 3 \"fit\", respectivement linéaire, polynomial de degré 2 et exponentiel. On choisira par la suite celui qui nous semble le plus en adéquation." ] }, { "cell_type": "code", "execution_count": 24, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 24, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "#Fit lineaire avec numpy\n", "a, b = np.polyfit(udc['IndexMois'], udc['CO2'], 1)\n", "yLin = a*udc['IndexMois'] + b\n", "\n", "#Fit de degré 2\n", "a2, b2, c2 = np.polyfit(udc['IndexMois'], udc['CO2'], 2)\n", "yCarre = a2*udc['IndexMois']**2 + b2*udc['IndexMois'] + c2\n", "\n", "#Fit exponentiel\n", "aExp, bExp = np.polyfit(udc['IndexMois'], [np.log(y) for y in udc['CO2']], 1)\n", "yExp = np.exp(bExp)*np.exp(aExp*udc['IndexMois'])\n", "\n", "plt.plot(udc['IndexMois'], udc['CO2'], label='data')\n", "plt.plot(udc['IndexMois'], yLin, label='lin')\n", "plt.plot(udc['IndexMois'], yCarre, label='deg2')\n", "plt.plot(udc['IndexMois'], yExp, label='exp')\n", "plt.legend()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Le polynôme de degré 2 est plus adapté à nos données ici. Pour faire des extrapolations sur les années à suivre, il suffit de tracer la courbe en étendant sur la plage de valeurs des abscisses qui nous intéresse. Pour obtenir des valeurs annuelles moyennes, il suffit d'intégrer la fonction sur 12 mois." ] }, { "cell_type": "code", "execution_count": 25, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[]" ] }, "execution_count": 25, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "#on va afficher les valeurs de 2010 à 2025, sachant que nos données s'arrêtent à Avril 2020 (748)\n", "borneJanvier2010 = (2010-1958)*12 + 1\n", "borneDecembre2025 = (2025-1958)*12 +12\n", "x1 = [x for x in range(borneJanvier2010, 749)]\n", "x2 = [x for x in range(borneJanvier2010, borneDecembre2025+1)]\n", "\n", "y1 = udc['CO2'][-(749-borneJanvier2010):]\n", "y2 = a2*np.asarray(x2)**2 + b2*np.asarray(x2) + c2\n", "\n", "plt.plot(x1, y1)\n", "plt.plot(x2, y2)" ] }, { "cell_type": "code", "execution_count": 26, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "valeur moyenne pour l'annee 2020 : [CO2] = 412.69 ppm\n", "valeur moyenne pour l'annee 2021 : [CO2] = 415.09 ppm\n", "valeur moyenne pour l'annee 2022 : [CO2] = 417.52 ppm\n", "valeur moyenne pour l'annee 2023 : [CO2] = 419.97 ppm\n", "valeur moyenne pour l'annee 2024 : [CO2] = 422.45 ppm\n", "valeur moyenne pour l'annee 2025 : [CO2] = 424.96 ppm\n" ] } ], "source": [ "#Valeurs moyennes : on peut se contenter d'une intégration manuelle ici\n", "for x in range(2020, 2026):\n", " borneInf = (x-1958)*12\n", " borneSup = borneInf + 12\n", " Y2 = (a2*borneSup**3)/3 + (b2*borneSup**2)/2 + c2*borneSup\n", " Y1 = (a2*borneInf**3)/3 + (b2*borneInf**2)/2 + c2*borneInf\n", " meanValue = (Y2-Y1)/(borneSup-borneInf)\n", " print(\"valeur moyenne pour l'annee \", x, \" : [CO2] = \", format(meanValue, '0.2f'), \" ppm\")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Pour finir, on peut vérifier nos résultats en comparant le fit avec les données réelles pour une année entièrement renseignée, 2019 par exemple." ] }, { "cell_type": "code", "execution_count": 27, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "733 734 735 736 737 738 739 740 741 742 743 744 \n", "733 744\n", "[CO2] (moyenne réelle) : \t 411.28\n", "[CO2] (moyenne fit) : \t 410.31\n" ] } ], "source": [ "#calcul avec data\n", "newBorneInf = (2019-1958)*12+1 \n", "newBorneSup = newBorneInf + 11\n", "somme = 0\n", "for x in range(newBorneInf, newBorneSup+1):\n", " somme += udc['CO2'][x]\n", " print(x, end = ' ')\n", "print(\"\")\n", "newMean = somme/12\n", "\n", "#calcul avec fit\n", "newY2 = (a2*newBorneSup**3)/3 + (b2*newBorneSup**2)/2 + c2*newBorneSup\n", "newY1 = (a2*(newBorneInf-1)**3)/3 + (b2*(newBorneInf-1)**2)/2 + c2*(newBorneInf-1) \n", "# on ne peut pas utiliser la même borne pour une intégration ou une somme ici sans omettre une valeur, d'où le -1\n", "newMeanValue = (newY2-newY1)/12\n", "\n", "print(newBorneInf, newBorneSup)\n", "print(\"[CO2] (moyenne réelle) : \\t\", format(newMean, '0.2f'))\n", "print(\"[CO2] (moyenne fit) : \\t\", format(newMeanValue, '0.2f'))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Le résultat est acceptable, on peut calculer l'erreur relative." ] }, { "cell_type": "code", "execution_count": 28, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "-0.0023505910290639647" ] }, "execution_count": 28, "metadata": {}, "output_type": "execute_result" } ], "source": [ "erreurRelative = (newMeanValue - newMean)/newMean\n", "erreurRelative" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Conclusion\n", "Le but de cette étude était de produire une analyse reproductible de l'évolution de la concentration en CO2 dans l'atmosphère. Les données de base permettent une étude de 1958 à 2020, modulo les données manquantes.\n", "En utilisant les librairies Pandas et Numpy pour traiter les données, nous avons constaté une croissance globale de la concentration en CO2 d'année en année, couplée à une oscillation de cette concentration avec des maxima autour de Mai et des minima autour de Septembre. Après quelques recherches, sur [cette page](https://en.wikipedia.org/wiki/Keeling_Curve), l'augmentation globale serait due à l'utilisation des énergies fossiles, et l'oscillation annuelle à l'effet de la photosynthèse de la flore terrestre.\n", "Nous avons ensuite appliqué 3 fonctions différentes pour trouver une courbe de tendance raisonnable pour notre jeu de données. Le choix s'est fait sur un polynôme de degré 2, qui est un bon compromis car il suit de près les données sans pour autant être trop complexe à traiter.\n", "Ce fit a ensuite permis d'extrapoler les valeurs moyennes des concentrations en CO2 pour les années 2020 à 2025. Une vérification sur l'année 2019 pour laquelle l'intégralité des données sont disponibles suggère une erreur relative inférieure au pourcent (0.2% en l'occurence), ce qui est semble acceptable.\n", "Pour compléter cette étude, un travail possible serait de caractériser l'oscillation de la concentration en CO2 dans l'année par une fonction sinusoidale par exemple." ] }, { "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 }