Prospects for short-term forecasting of river streamflow runoff using machine learning methods – a case study of the Protva river

Section
Methods, models and technologies
  • Всеволод Морейдо Институт водных проблем РАН
Keywords: hydrological forecasts, machine learning, river flow forecasts, artificial neural networks, MLP, LSTM
(+) Abstract

The paper addresses prospects for short-term (from 1 to 7 days) forecasting of river streamflow runoff based on several machine learning methods: multiple linear regression (LM) model, an multilayer perceptron (MLP) artificial neural network, and a recurrent artificial neural network with long short-term memory (LSTM). Methods for expanding the set of predictors for model construction are proposed, and the possibility of random shuffling of the time-series of predictors for model calibration and verification are assessed. The object of the study is the small river of Central Russia - r. Protva (Spas-Zagorie gauge). The obtained results show the possibility of constructing an effective operational forecasting system for short-term runoff forecasting. The study revealed the applicability of artificial neural network models, acceptable for operational practice, using all available hydrometeorological information on the catchment, as they showed the most stable results at all lead times from 1 to 7 days.

(+) About the author(s) (+) References
1. Белякова П. А. и др. Прогноз максимального стока рек Черноморского побережья Кавказа // Водное хозяйство России: проблемы, технологии, управление. 2013. № 6. С. 4–16.
2. Abadi M. и др. TensorFlow: A System for Large-Scale Machine Learning // Proceedings of the 12th USENIX Symposium on Operating Systems. , 2016. С. 265–283.
3. Abrahart R. J. и др. Two decades of anarchy? Emerging themes and outstanding challenges for neural network river forecasting // Prog. Phys. Geogr. 2012. Т. 36. № 4. С. 480–513.
4. Beven K. Deep learning, hydrological processes and the uniqueness of place // Hydrol. Process. 2020. Т. 34. № 16. С. 3608–3613.
5. Bhattacharya B., Solomatine D. P. Machine learning in sedimentation modelling // Neural Networks. 2006. Т. 19. № 2. С. 208–214.
6. Emerton R. E. и др. Continental and global scale flood forecasting systems // WILEY Interdiscip. Rev. 2016. Т. 3. № 3. С. 391–418.
7. Fatichi S. и др. An overview of current applications, challenges, and future trends in distributed process-based models in hydrology // J. Hydrol. 2016. Т. 537. С. 45–60.
8. Gelfan A. N., Motovilov Y. G. Long-term hydrological forecasting in cold regions: Retrospect, current status and prospect // Geogr. Compass. 2009. Т. 3. № 5. С. 1841–1864.
9. Halff A. H., Halff H. M., Azmoodeh M. Predicting runoff from rainfall using neural networks // Eng. Hydrol. 1993. С. 760–765.
10. Haykin S. Rosenblatt ’ s Perceptron // Neural Networks Learn. Mach. 2009. № 1943. С. 47–67.
11. Haykin S. S. Neural networks : a comprehensive foundation. : Macmillan, 1994. 696 с.
12. Hochreiter S. The vanishing gradient problem during learning recurrent neural nets and problem solutions // Int. J. Uncertainty, Fuzziness Knowlege-Based Syst. 1998. Т. 6. № 2. С. 107–116.
13. Hochreiter S., Schmidhuber J. Long Short-Term Memory // Neural Comput. 1997. Т. 9. № 8. С. 1735–1780.
14. Kratzert F. и др. Rainfall–runoff modelling using Long Short-Term Memory (LSTM) networks // Hydrol. Earth Syst. Sci. 2018. Т. 22. № 11. С. 6005–6022.
15. Moreydo V. M. github.com/esmoreido [Электронный ресурс]. URL: https://github.com/esmoreido (дата обращения: 01.11.2020).
16. Mozer M. A Focused Backpropagation Algorithm for Temporal Pattern Recognition // Complex Syst. 1989. Т. 3. С. 349–381.
17. Oudin L., Michel C., Anctil F. Which potential evapotranspiration input for a lumped rainfall-runoff model? // J. Hydrol. 2005. Т. 303. № 1–4. С. 275–289.
18. Pedregosa F. и др. Scikit-learn: Machine Learning in Python. , 2011. 2825–2830 с.
19. Werbos P. J. Backpropagation Through Time: What It Does and How to Do It // Proc. IEEE. 1990. Т. 78. № 10. С. 1550–1560.
20. Борщ С. В., Христофоров А. В. ОЦЕНКА КАЧЕСТВА ПРОГНОЗОВ РЕЧНОГО СТОКА // Труды Гидрометеорологического научно-исследовательского центра Российской Федерации. 2015. Т. 355. С. 3–195.
21. Зайков Б. Д. Средний сток и его распределение в году по территории СССР // Труды НИУ ГУГМС Серия IV. Вып. 24. Л.-М.: , 1946. С. 67–95.
22. Кучмент Л. С., Гельфан А. Н. Динамико-стохастические модели формирования речного стока. : Наука М, 1993. 104 с.
23. Мотовилов Ю. Г. Гидрологическое Моделирование Речных Бассейнов В Различных Пространственных Масштабах. 2. Результаты Испытаний // Водные Ресурсы. 2016a. Т. 43. № 5. С. 467–475.
24. Мотовилов Ю. Г. Гидрологическое Моделирование Речных Бассейнов В Различных Пространственных Масштабах 1. Алгоритмы Генерализации И Осреднения // Водные Ресурсы. 2016b. Т. 43. № 3. С. 243–253.
(+) Read online
Published
2022-03-10
How to Cite
Морейдо, В. (2022). Prospects for short-term forecasting of river streamflow runoff using machine learning methods – a case study of the Protva river. Hydrosphere. Hazard Processes and Phenomena, 2(4). Retrieved from http://hydro-sphere.ru/index.php/hydrosphere/article/view/67

Downloads

Download data is not yet available.