Arctic: ecology and economy
ISSN 2223-4594 | ISSN 2949-110X
Home Archive of journals No. 2(38) 2020 Two-step combined algorithm for improving the accuracy of predicting methane concentration in atmospheric air based on the NARX neural network and subsequent prediction of residuals


JOURNAL: No. 2(38) 2020, p. 59-67

HEADING: Research activities in the Arctic

AUTHORS: Subbotina, I.E., Buevich, A.G., Sergeev, A.P., Shichkin, A.V., Baglaeva, E.M., Remezova, M.S.

ORGANIZATIONS: Institute of Industrial Ecology, Ural Branch of the RAS, Ural Federal University

DOI: 10.25283/2223-4594-2020-2-59-67

UDC: 504.3.054(985)

The article was received on: 19.11.2019

Keywords: greenhouse gases, artificial neural networks, NARX, residuals

Bibliographic description: Subbotina, I.E., Buevich, A.G., Sergeev, A.P., Shichkin, A.V., Baglaeva, E.M., Remezova, M.S. Two-step combined algorithm for improving the accuracy of predicting methane concentration in atmospheric air based on the NARX neural network and subsequent prediction of residuals. Arctic: ecology and economy, 2020, no. 2(38), pp. 59-67. DOI: 10.25283/2223-4594-2020-2-59-67. (In Russian).


Climate change in the Arctic is great and can have a significant inverse effect on the global climate, which determines the global significance of climate change in the Arctic. To date, many issues regarding the mechanisms responsible for the rapid melting of Arctic ice and permafrost degradation have not been resolved. It is not known when and what consequences these changes will lead to. Assessing the relationship between global warming and greenhouse gas emissions is an important environmental challenge. Among the main greenhouse gases, the evolution and climate-forming role of the carbon dioxide have been studied. The data on the methane subcycle of the carbon cycle is much less. In the paper, the authors propose a two-step combined algorithm (NARXR) to improve the accuracy of predicting methane concentration in atmospheric air based on the NARX neural network and subsequent prediction of the residuals. Two commonly used models based on artificial neural networks (ANN) for predicting time series are compared to determine the most appropriate base model. Nonlinear autoregressive neural network with external input (NARX) and Elman’s neural network are used. For the forecast, the authors use data on the methane concentration (CH4) in the atmospheric surface layer on the Arctic Island of Bely (Russia). Data is selected for a time interval of 192 hours, because it is characterized by significant daily fluctuations in the concentration of CH4. Values corresponding to the first 168 hours of the interval are used to train the ANN, and then concentrations are predicted for the next 24 hours. The proposed approach shows more accurate forecast results.


1. Arktika: zona mira i sotrudnichestva. [Arctic: zone of peace and cooperation]. Ed. by A. V. Zagorsky. Moscow, IMEMO RAN, 2011, 195 p. (In Russian).

2. Vtoroy otsenochnyy doklad ob izmeneniyakh klimata i ikh posledstviyakh na territorii Rossiyskoy Federatsii. [Evaluation report The second assessment report of Roshydromet on climate change and its consequences on the territory of the Russian Federation]. [G. V. Alekseyev i dr.]. Feder. sluzhba po gidrometeorologii i monitoringu okruzhayushchey sredy (Rosgidromet). Moscow, Roshydromet, 2014, 1007 p. (In Russian).

3. AMAP Climate Change Update 2019: An Update to Key Findings of Snow, Water, Ice and Permafrost in the Arctic (SWIPA) 2017. Arctic Monitoring and Assessment Programme (AMAP). Oslo, Norway, 2019.

4. Serreze M. C., Barry R. G. Processes and impacts of Arctic amplification: A research synthesis. Global and Planetary Change, 2011, vol. 77 (1-2), . 85—96.

5. Perlshtein G. Z., Sergeev D. O., Tipenko G. S., Tumskoy V. E., Khimenkov A. N., Vlasov A. N., Merzlyakov V. P., Stanilovskaya Yu. V. Uglevodorodnyye gazy i kriolitozona shel’fa Arktiki. [Hydrocarbon gases and cryolithozone of the Arctic shelf]. Arktika: ekologiya i ekonomika, 2015, no. 2 (18), . 35—44. (In Russian).

6. Arzhanov M. M., Mokhov I. I., Denisov S. N. Destabilizatsiya reliktovykh metangidratov pri nablyudayemykh regional’nykh izmeneniyakh klimata. [Destabilization of relict methanhydrates with observed regional climate changes]. Arktika: ekologiya i ekonomika, 2016, no. 4 (24), . 46—51. (In Russian).

7. Dzyuba A. V., Eliseev A. V., Mokhov I. I. Otsenka izmeneniy skorosti stoka metana iz atmosfery pri poteplenii klimata. [Estimates of changes in the rate of methane sink from the atmosphere under climate warming]. Izv. RAN. Ser. Fizika atmosfery i okeana, 2012, no. 48 (3), . 372—382. (In Russian).

8. Chu N., Kadane J. B., Davidson C. I. Using statistical regressions to identify factors influencing PM2.5 concentrations: The Pittsburgh supersite as a case study. Aerosol Science and Technology, 2010, vol. 44, . 766—774.

9. Cobourn W. G. An enhanced PM2.5 air quality forecast model based on nonlinear regression and back- trajectory concentrations. Atmospheric Environment, 2010, vol. 44, P. 3015—3023.

10. Chatfield Ch. The Analysis of Time Series: An Introduction, Sixth Edition. New York, Chapman & Hall/CRC, 2016, 352 p.

11. McNorton J., Wilson C., Gloor M., Parker R. J., Boesch H., Feng,W., Hossaini R., and Chipperfield M. P. Attribution of recent increases in atmospheric methane through 3-D inverse modelling. Atmospheric Chemistry and Physics, 2018, vol. 18, . 18149—18168. Available at: https://doi.org/10.5194/acp-18-18149-2018.

12. Russo A., Raischel F., Lind P. Air quality prediction using optimal neural networks with stochastic variables. Atmospheric Environment, 2013, vol. 79, . 822—830.

13. Zhang G. Time series forecasting using a combined ARIMA and neural network model. Neurocomputing, 2003, vol. 50, . 159—175.

14. Zhou Q., Jiang H., Wang J., Zhou J. A combined model for PM2.5 forecasting based on ensemble empirical mode decomposition and a general regression neural network. Science of the Total Environment, 2014, vol. 496, . 264—274

15. Tabassum-Abbasi, Abbasi T., Luithui C., Abbasi S. A. A Model to Forecast Methane Emissions from Tropical and Subtropical Reservoirs on the Basis of Artificial Neural Networks. Water, 2020, vol. 12, . 145—153.

16. Ardalani-Farsa M., Zolfaghari S. Chaotic time series prediction with residual analysis method using combined Elman–NARX neural networks. Neurocomputing, 2010, vol. 73, . 2540—2553.

17. Erdil A., Arcaklioglu E. The prediction of meteorological variables using artificial neural network. Neural Computing & Applications, 2013, vol. 22, . 1677—1683.

18. Menezes J. M. Jr, Barreto G. A. Long-term time series prediction with the NARX network: An empirical evaluation. Neurocomputing, 2008, vol. 71, . 3335—3343.

19. Pisoni E., Farina M., Carnevale C., Piroddi L. Forecasting peak air pollution levels using NARX models. Engineering Applications of Artificial Intelligence, 2009, vol. 22, . 593—602.

20. Zemouri R., Gouriveau R., Zerhouni N. Defining and applying prediction performance metrics on a recurrent NARX time series model. Neurocomputing, 2010, vol. 73, . 2506—2521.

21. Sergeev A. P., Buevich A. G., Baglaeva E. M., Shichkin A. V. Combining spatial autocorrelation with machine learning increases prediction accuracy of soil heavy metals. Catena, 2019, vol. 174, . 425—435.

22. Tarasov D. A., Buevich A. G., Sergeev A. P., Shichkin A. V. High Variation Topsoil Pollution Forecasting in the Russian Subarctic: Using Artificial Neural Networks Combined with Residual Kriging. Applied Geochemistry, 2018, vol. 88(B), . 188—197.

23. Dai F., Zhoua O., Lva Z., Wang X., Liu G. Spatial prediction of soil organic matter concentration integrating artificial neural network and ordinary kriging in Tibetan Plateau. Ecological Indicators, 2014, vol. 45, . 184—194.

24. Kanevski M., Pozdnoukhov A., Timonin V. Machine Learning for Spatial Environmental Data: Theory, Applications and Software. [S. l.], EPFL Press, 2009, p. 377.

25. Lakes T., Müller D., Krüger C. Cropland change in southern Romania: a comparison of logistic regressions and artificial neural networks. Landscape Ecology, 2009, vol. 24 (9), . 1195—1206.

26. Seo Y., Kim S., Singh V. P. Estimating spatial precipitation using regression Kriging and Artificial Neural Network Residual Kriging (RKNNRK) hybrid approach. Water Resources Management, 2015, vol. 28, . 2189—2204.

27. Song Y.-Q., Yang L.-A., Li B. et al. Spatial prediction of soil organic matter using a hybrid geostatistical model of an extreme learning machine and ordinary kriging. Sustainability, 2017, vol. 9, . 754. Available at: https://doi.org/10.3390/su9050754.

28. Antonov K. L., Poddubny V. A., Markelov Yu. I., Buevich A. G., Medvedev A. N., Manzhurov I. L. Nekotoryye itogi monitoringa parnikovykh gazov v Arkticheskom regione Rossii. [Some Results of the Monitoring of Greenhouse Gases in the Arctic Region of Russia]. Arktika: ekologiya i ekonomika, 2018, vol. 1 (30), . 56—67. DOI: 10.25283/2223-4594-2018-1-56-67. (In Russian).

29. Shepherd A. J. Second-Order Methods for Neural Networks: Fast and Reliable Training Methods for Multi-Layer Perceptrons. [S. l.], Springer-Verlag, 1997, 145 p.

30. Shichkin A. V., Buevich A. G., Sergeev A. P. et al. Training algorithms for artificial neural networks for time series forecasting of greenhouse gas concentrations. AIP Conference Proceedings, 2019, 2116 (1), 200019. DOI: 10.1063/1.5114200.

31. Willmott C. J. On the Validation of Models. Physical Geography, 1981, vol. 2, . 184—194.

32. Willmott C. J., Robeson S. M., Matsuura K. A refined index of model performance. International J. of Climatology, Wiley Online Library, 2011. Available at: http:// wileyonlinelibrary.com. DOI: 10.1002/joc.2419.

Download »

© 2011-2024 Arctic: ecology and economy
DOI 10.25283/2223-4594