A HYBRID MODEL BASED ON AN ARTIFICIAL NEURAL NETWORK WITH A LONG CHAIN OF SHORT-TERM MEMORY ELEMENTS AND A DISCRETE WAVELET TRANSFORM FOR PREDICTING SURFACE METHANE CONTENT IN THE ARCTIC AREA

The study of the dynamics of greenhouse gases in the Arctic regions of the planet is becoming increasingly important. Such studies are especially relevant due to the climate change observed in this region. The paper propose a hybrid model that combines wavelet transformation of the original data and an artificial neural network with a long chain of short-term memory (LSTM) elements to predict changes in the surface methane concentration in the Arctic latitudes. The methane concentration time series via a discrete wavelet transform was decomposed into four components — one approximating and three detailing ones. These components were used to train the LSTM network. The forecast was calculated as the sum of forecasts for each of the components. Three predictive models were built. In the first, the LSTM network was trained in a non-linear autoregressive mode. The second one was a combination of discrete wavelet transform with LSTM neural network. An additional model based on a non-linear autoregressive neural network (NAR) was also used for comparison. The work is based on data from environmental monitoring of greenhouse gases on Bely Island, Yamalo-Nenets Autonomous Area of Russia. The initial data for building the proposed model were obtained within July-August 2017. The accuracy of the forecast was assessed using several indicators. The hybrid model based on LSTM showed the best accuracy.

1. Tao Q., Liu F., Li Y., Sidorov D. Air Pollution Forecasting Using a Deep Learning Model Based on 1D Convnets and Bidirectional GRU. IEEE Access, 2019, vol. 7, pp. 76690—76698. DOI: 10.1109/ACCESS.2019.2921578.
2. Zhao J., Deng F., Cai Y., Chen J. Long short-term memory — Fully connected (LSTM-FC) neural network for PM2.5 concentration prediction. Chemosphere, 2019, vol. 220, pp. 486—492.
3. Xu S., Niu R. Displacement prediction of Baijiabao landslide based on empirical mode decomposition and long short-term memory neural network in Three Gorges area, China. Computers & Geosciences, 2019, vol. 111, pp. 87—96. Available at: https://doi.org/10.1016/j.cageo.2017.10.013.
4. Tong W., Li L., Zhou X., Hamilton A., Zhang K. Deep learning 578 PM2.5 concentrations with bidirectional LSTM RNN. Air Quality, Atmosphere & 579 Health, 2019, vol. 12 (4), pp. 411—423. Available at: https://doi.org/10.1007/s11869-018-0647-4.
5. Jung D.-H., Kim H. S., Jhin C., Kim H.-J., Park S. H. Time-serial analysis of deep neural network models for prediction of climatic conditions inside a greenhouse. Computers and Electronics in Agriculture, 2020, vol. 173, p. 105402. Available at: https://doi.org/10.1016/j.compag.2020.105402.
6. Hamrani A., Akbarzadeh A., Madramootoo C. A. Machine learning for predicting greenhouse gas emissions from agricultural soils. Science of The Total Environment, 2020, vol. 741, p. 140338. Available at: https://doi.org/10.1016/j.scitotenv.2020.140338.
7. Li W., Kiaghadi A., Dawson C. High temporal resolution rainfall–runoff modeling using long-short-term-memory (LSTM) networks. Neural Comput & Applic Environment, 2021, vol. 33, pp. 1261—1278. Available at: https://doi.org/10.1007/s00521-020-05010-6.
8. Sherstinsky A. Fundamentals of Recurrent Neural Network (RNN) and Long Short-Term Memory (LSTM) network. Physica D: Nonlinear Phenomena, 2020, vol. 404, p. 132306. Available at: https://doi.org/10.1016/j.physd.2019.132306.
9. Memarzadeh G., Keynia F. A new short-term wind speed forecasting method based on fine-tuned LSTM neural network and optimal input sets. Energy Conversion and Management Environment, 2020, vol. 213, p. 112824. https://doi.org/10.1016/j.enconman.2020.112824.
10. Wang Y., Liu P., Xu C., Peng C., Wu J. A deep learning approach to real-time CO concentration prediction at signalized intersection. Atmospheric Pollution Research, 2020, vol. 11 (8), pp. 1370—1378. Available at: https://doi.org/10.1016/j.apr.2020.05.007.
11. Chang Y.-S., Chiao H.-T., Abimannan S., Huang Y.-P., Tsai Y.-T., Lin K.-M. An LSTM-based aggregated model for air pollution forecasting. Atmospheric Pollution Research, 2020, vol. 11 (8), pp. 1451—1463. Available at: https://doi.org/10.1016/j.apr.2020.05.015.
12. Zheng Y., Yi X., Li M., Li R., Shan Z., Chang E., Li T. Forecasting Fine-Grained Air Quality Based on Big Data. KDD ‘15, Australia: ACM, pp. 2267—2276. Available at: https://doi.org/10.1145/2783258.2788573.
13. Soh P., Chang J., Huang J. Adaptive Deep Learning-Based Air Quality Prediction Model Using the Most Relevant Spatial-Temporal Relations. IEEE Access, 2020, vol. 8, pp. 38186—38199. DOI: 10.1109/ACCESS.2018.2849820.
14. Ma J. Ding Y., Gan V. J. L., Lin C., Wan Z. Spatiotemporal Prediction of PM2.5 Concentrations at Different Time Granularities Using IDW-BLSTM. IEEE Access., 2020, vol. 7, p. 107907. DOI: 10.1109/ACCESS.2019.2932445.
15. Adamowski J., Chan H. F. A wavelet neural network conjunction model for groundwater level forecasting. J. of Hydrology, 2011, vol. 407 (1-4), pp. 28—40. Available at: https://doi.org/10.1016/j.jhydrol.2011.06.013.
16. Ebrahimi H., Rajaee T. Simulation of groundwater level variations using wavelet combined with neural network, linear regression and support vector machine. Glob Planet Change, 2017, vol. 148, pp. 181—191. Available at: https://doi.org/10.1016/j.gloplacha.2016.11.014.
17. Kalteh A. M. Monthly River flow forecasting using artificial neural network and support vector regression models coupled with wavelet transform. Computers & Geosciences, 2013, vol. 54, pp. 1—8. Available at: https://doi.org/10.1016/j.cageo.2012.11.015.
18. Graf R., Zhu S., Sivakumar B. Forecasting river water temperature time series using a wavelet–neural network hybrid modelling approach. J. of Hydrology, 2019, p. 124115. Available at: https://doi.org/10.1016/j.jhydrol.2019.124115.
19. Siwek K., Osowski S. Improving the accuracy of prediction of PM10 pollution by the wavelet transformation and an ensemble of neural predictors. Engineering Applications of Artificial Intelligence, 2012, vol. 25 (6), pp. 1246—1258. Available at: https://doi.org/10.1016/j.engappai.2011.10.013.
20. Chen Y., Shi R., Shu S., Gao W. Ensemble and enhanced PM10 concentration forecast model based on stepwise regression and wavelet analysis. Atmospheric Environment, 2013, vol. 74, pp. 346—359. Available at: https://doi.org/10.1016/j.atmosenv.2013.04.002.
21. Feng X., Li Q., Zhu Y., Hou J., Jin L., Wang J. Artificial neural networks forecasting of PM 2.5 pollution using air mass trajectory based geographic model and wavelet transformation. Atmospheric Environment, 2016, vol. 107, pp. 118—128. Available at: https://doi.org/10.1016/j.atmosenv.2015.02.030.
22. Dunea D., Pohoata A., Iordache S. Using wavelet–feedforward neural networks to improve air pollution forecasting in urban environments. Environmental Monitoring and Assessment., 2015, vol. 187 (7), p. 477. Available at: https://doi.org/10.1007/s10661-015-4697-x.
23. Bai Y., Li Y., Wang X., Xie J., Li C. Air pollutants concentrations forecasting using back propagation neural network based on wavelet decomposition with meteorological conditions. Atmospheric Pollution Research, 2016, vol. 7 (3), pp. 557—566. Available at: https://doi.org/10.1016/j.apr.2016.01.004.
24. Cabaneros S. M., Calautit J. K., Hughes B. Spatial estimation of outdoor NO2 levels in Central London using deep neural networks and a wavelet decomposition technique. Ecological Modelling, 2020, vol. 424, p. 109017. Available at: https://doi.org/10.1016/j.ecolmodel.2020.109017.
25. Osowski S., Garanty K. Forecasting of the daily meteorological pollution using wavelets and support vector machine. Engineering Applications of Artificial Intelligence, 2007, vol. 20 (6), pp. 745—755. Available at: https://doi.org/10.1016/j.engappai.2006.10.008.
26. Su X., An J., Zhang Y., Zhu P., Zhu B. Prediction of ozone hourly concentrations by support vector machine and kernel extreme learning machine using wavelet transformation and partial least squares methods. Atmospheric Pollution Research Intelligence, 2020, vol. 11 (6), pp. 51—60. Available at: https://doi.org/10.1016/j.apr.2020.02.024.
27. Mallat S. A theory for multiresolution signal decomposition: the wavelet representation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1989, vol. 11, pp. 674—693. Available at: https://doi.org/10.1109/34.192463.
28. Daubechies I. Ten Lectures on Wavelets. Philadelphia, PA, SIAM Press, 1988.
29. Buevich A., Sergeev A., Shichkin A., Baglaeva E. A two-step combined algorithm based on NARX neural network and the subsequent prediction of the residues improves prediction accuracy of the greenhouse gases concentrations. Neural Computing and Applications, 2021, vol. 33, pp. 1547—1557. Available at: https://doi.org/10.1007/s00521-020-04995-4.
30. Rakhmatova A., Sergeev A., Shichkin A., Buevich A., Baglaeva E. Three-day forecasting of greenhouse gas CH4 in the atmosphere of the arctic Belyy island using discrete wavelet transform and artificial neural networks. Neural Computing & Applications, 2021, vol. 33, pp. 10311—10322.
31. Hochreiter S., Schmidhuber J. Long short-term memory. Neural computation, 1997, vol. 9 (8), pp. 735—1780.
32. Sainath T. N., Vinyals O., Senior A., Sak H. Convolutional, Long Short-Term Memory, fully connected Deep Neural Networks. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), South Brisbane, QLD, Australia. [S. l.], 2015, pp. 4580—4584. DOI: 10.1109/ICASSP.2015.7178838.
33. Fischer T., Krauss C. Deep learning with long short-term memory networks for financial market predictions. European J. of Operational Research, 2018, vol. 270 (2), pp. 654—669. Available at: https://doi.org/10.1016/j.ejor.2017.11.054.
34. Willmott C. J. On the Validation of Models. Physical Geography, 1981, vol. 2, pp. 184—194.
35. Willmott C. J., Robeson S. M., Matsuura K. A refined index of model performance. Intern. J. of Climatology, 2012, vol. 32 (13), pp. 2088—2094. Available at: https://doi.org/10.1002/joc.2419.