Main Article Content

Abstract

CO2 emissions have been an environmental issue for decades. The trigger for the increasing concentration of CO2 in the atmosphere is the growth of industries related to burning fossil fuels for coal, natural gas, and petroleum. For nearly a century, several attempts have been made to suppress the rapid growth of CO2 . This study uses daily atmospheric  CO2  levels observed in  Mauna Loa laboratories. The method used is a Prophet that can handle seasonality and mark the change points. Almost 20% of data was missing value, which was then imputed using spline interpolation. Based on the analysis results,  CO2 levels have an upward trend throughout the year and seasonality. There is no point of change in the last ten years that shows a decrease in  CO2  levels. Using forward chaining cross-validation evaluation and error measurement, the prophet model can follow the pattern of  CO2  levels well. The average RMSE value is less than 2.0, with an MAPE value bellow 0.5%.

Keywords

Carbon dioxide Prophet Spline interpolation Time series

Article Details

Author Biographies

Arum Handini Primandari, Universitas Islam Indonesia, Indonesia

 

 

Achmad Kurniansyah Thalib, Purwadhika Digital Technology School, Indonesia

 

 

Ayundyah Kesumawati, Universitas Islam Indonesia, Indonesia

 

 

How to Cite
Primandari, A. H., Thalib, A. K., & Kesumawati, A. (2022). Analysis of Changes in Atmospheric CO2 Emissions Using Prophet Facebook. Enthusiastic : International Journal of Applied Statistics and Data Science, 2(1), 1–9. https://doi.org/10.20885/enthusiastic.vol2.iss1.art1

References

  1. R. Betts and R. Keeling, “Atmospheric carbon dioxide at record high levels despite reduced emissions in 2020 - Met Office,” 2021. https://www.metoffice.gov.uk/research/news/2021/record-co2-levels-despite-lower-emissions-in-2020 (accessed Feb. 22, 2022).
  2. Z. Z. OO and S. PHYU, “Time Series Prediction Based on Facebook Prophet: A Case Study, Temperature Forecasting in Myintkyina,” Int. J. Appl. Math. Electron. Comput., vol. 8, no. 4, pp. 263–267, 2020, doi: 10.18100/ijamec.816894.
  3. T. Toharudin, R. S. Pontoh, R. E. Caraka, S. Zahroh, Y. Lee, and R. C. Chen, “Employing long short-term memory and Facebook prophet model in air temperature forecasting,” Commun. Stat. Simul. Comput., vol. 0, no. 0, pp. 1–24, 2020, doi: 10.1080/03610918.2020.1854302.
  4. H. Weytjens, E. Lohmann, and M. Kleinsteuber, “Cash flow prediction: MLP and LSTM compared to ARIMA and Prophet,” Electron. Commer. Res., no. 0123456789, 2019, doi: 10.1007/s10660-019-09362-7.
  5. R. S. Pontoh, S. Zahroh, H. R. Nurahman, R. I. Aprillion, A. Ramdani, and D. I. Akmal, “Applied of feed-forward neural network and facebook prophet model for train passengers forecasting,” J. Phys. Conf. Ser., vol. 1776, no. 1, pp. 0–9, 2021, doi: 10.1088/1742-6596/1776/1/012057.
  6. M. Lounis, “Predicting active , death and recovery rates of COVID-19 in Al- geria using Facebook ’ Prophet model,” Preprints, vol. 1, no. March, 2021, doi: 10.20944/preprints202103.0019.v1.
  7. S. Mahmud, “Bangladesh COVID-19 Daily Cases Time Series Analysis using Facebook Prophet Model,” SSRN Electron. J., no. June 2020, 2020, doi: 10.2139/ssrn.3660368.
  8. K. W. Thoning, A. M. Crotwell, and J. W. Mund, “Atmospheric Carbon Dioxide Dry Air Mole Fractions from continuous measurements at Mauna Loa, Hawaii, Barrow, Alaska, American Samoa and South Pole,” Boulder, Colorado, USA, 2021. doi: https://doi.org/10.15138/yaf1-bk21.
  9. C. L. Zhao and P. P. Tans, “Estimating uncertainty of the WMO mole fraction scale for carbon dioxide in air,” J. Geophys. Res. Atmos., vol. 111, no. 8, Apr. 2006, doi: 10.1029/2005JD006003.
  10. L. Guo, W. Fang, Q. Zhao, and X. Wang, “The hybrid PROPHET-SVR approach for forecasting product time series demand with seasonality,” Comput. Ind. Eng., vol. 161, no. February, p. 107598, 2021, doi: 10.1016/j.cie.2021.107598.
  11. G. Rafferty and an O. M. C. Safari, Forecasting Time Series Data with Facebook Prophet. Packt Publishing Ltd., 2021.
  12. S. Moritz and T. Bartz-beielstein, “imputeTS : Time Series Missing Value Imputation in R,” R J. Vol., vol. 9, no. 1, pp. 207–218, 2017.
  13. Z. Liu and X. Yang, “Cross validation for uncertain autoregressive model,” Commun. Stat. Simul. Comput., vol. 0, no. 0, pp. 1–12, 2020, doi: 10.1080/03610918.2020.1747077.
  14. M. Schnaubelt, “A comparison of machine learning model validation schemes for non-stationary time series data,” Nürnberg, No. 11/2019, 2019. [Online]. Available: http://hdl.handle.net/10419/209136.