Abstract
Different methods for the statistical correction of the forecasts of surface parameters using the COSMO-Ru13-ENA model with the lead time up to 117 hours are considered. The methods include the systematic correction using the data from recent observations at a weather station, the correction based on special neural networks as well as different combinations of these two techniques. The study presents the estimates of the results of applying the analyzed correction methods to the forecasts of surface air temperature, dew point, and wind speed modulus based on the independent sample for 2018 with the total volume of 2.34 × 107 forecasts. The correction method based on neural networks reduces forecast errors even at the points where meteorological observations have not been carried out.
Similar content being viewed by others
REFERENCES
A. N. Bagrov, F. L. Bykov, and V. A. Gordin, “Complex Forecast of Surface Meteorological Parameters,” Meteorol. Gidrol., No. 5 (2014) [Russ. Meteorol. Hydrol., No. 5, 39 (2014)].
A. N. Bagrov, F. L. Bykov, and V. A. Gordin, “Operative Scheme for the Short-range Complex Forecasting of Wind,” Meteorol. Gidrol., No. 7 (2018) [Russ. Meteorol. Hydrol., No. 7, 43 (2018)].
A. N. Bagrov, F. L. Bykov, and V. A. Gordin, “Operative Scheme for the Short-range Complex Forecasting of Surface Air Temperature and Humidity,” Meteorol. Gidrol., No. 8 (2018) [Russ. Meteorol. Hydrol., No. 8, 43 (2018)].
A. N. Bagrov, V. A. Gordin, E. A. Loktionova, and N. Yu. Ochan, “Control and Archiving of Global Surface Air Temperatures at the Hydrometeorological Center of the Russian Federation,” Meteorol. Gidrol., No. 2 (1993) [Russ. Meteorol. Hydrol., No. 2 (1993)].
D. V. Blinov and G. S. Rivin, “The COSMO-Ru Short-term Nonhydrostatic Mesoscale Weather Prediction System: The Technological Line,” Trudy Gidromettsentra Rossii, No. 365 (2017) [in Russian].
A. I. Galushkin, Synthesis of Multilayer Image Recognition Systems (Energiya, Moscow, 1974) [in Russian].
A. N. Kolmogorov, “Presentation of Continuous Multivariable Functions in the Form of Superposition of Continuous Single-variable Functions and Summation,” Dokl. AN SSSR, No. 5, 114 (1957) [in Russian].
RD 52.27.724-2009. Manual on General-purpose Short-range Weather Forecasting (IG-SOTsIN, Obninsk, 2009) [in Russian].
G. S. Rivin, I. A. Rozinkina, R. M. Vil’fand, D. Yu. Alferov, E. D. Astakhova, D. V. Blinov, A. Yu. Bundel’, E. V. Kazakova, A. A. Kirsanov, M. A. Nikitin, V. L. Perov, G. V. Surkova, A. P. Revokatova, M. V. Shatunova, and M. M. Chumakov, “The COSMO-Ru System of Nonhydrostatic Mesoscale Short-range Weather Forecasting of the Hydrometcenter of Russia: The Second Stage of Implementation and Development,” Meteorol. Gidrol., No. 6 (2015) [Russ. Meteorol. Hydrol., No. 6, 40 (2015)].
Website of COSMO Consortium, http://cosmo-model.org.
S. Khaikin, Neural Networks (Williams, Moscow, 2008).
Y. Bengio, I. Goodfellow, and A. Courville, Deep Learning (MIT Press, 2015).
A. Cotter, O. Shamir, N. Srebro, and K. Sridharan, “Better Mini-batch Algorithms via Accelerated Gradient Methods,” Adv. Neural Information Processing Systems, 24 (2011).
T. Dozat, “Incorporating Nesterov Momentum into Adam,” inICLR Workshop, Vol. 1 (2016).
X. Glorot and Y. Bengio, “Understanding the Difficulty of Training Deep Feedforward Neural Networks,” in Proceedings of AISTATS, Vol. 9 (2010).
K. He, X. Zhang, S. Ren, and J. Sun, “Delving Deep into Rectifiers: Surpassing Human-level Performance on ImageNet Classification,” ArXiv Preprint (2015).
K. He, X. Zhang, S. Ren, and J. Sun, “Deep Residual Learning for Image Recognition,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (2016).
E. Kalnay, Atmospheric Modeling, Data Assimilation and Predictability (Cambridge Univ. Press, 2003).
E. Kazakova, I. Rozinkina, and M. Chumakov, “Verification of Results of the Working Technology SNOWE for Snow Water Equivalent and Snow Density Fields Determination as Initial Data for COSMO Model,” COSMO Newsletter No. 16 (2016).
V. M. Krasnopolsky, The Application of Neural Networks in the Earth System Sciences: Neural Network Emulations for Complex Multidimensional Mappings (Springer, 2013).
V. M. Krasnopolsky and Y. Lin, “A Neural Network Nonlinear Multimodel Ensemble to Improve Precipitation Forecasts over Continental US,” Adv. Meteorol. (2012).
C. Marzban, “A Neural Network for Post-processing Model Output: ARPS,” Mon. Wea. Rev., No. 4, 131 (2003).
T. T. Warner, Numerical Weather and Climate Prediction (Cambridge Univ. Press, 2010).
C. Ye, C. Zhao, Y. Yang, C. Fermuller, and Y. Aloimonos, “LightNet: A Versatile, Standalone Matlab-based Environment for Deep Learning,” ArXiv Preprint (2016).
J. M. Zurada, Introduction to Artificial Neural Systems (PWS, New York, 1992).
Acknowledgements
The author thanks the specialists of Hydrometcenter of Russia: D.V. Blinov for assistance in the work with the COSMO-Ru model archives, V.A. Gordin, G.S. Rivin, and I.A. Rozinkina for fruitful discussions as well as the anonymous reviewer for useful remarks.
Funding
The paper was prepared in the framework of the “Higher School of Economics” Academic Fund Program in 2018–2019 (project 18-05-0011), in 2020–2021 (project 20-04-021), and in the framework of governmental support of the leading universities of the Russian Federation “5-100.”
Author information
Authors and Affiliations
Corresponding author
Additional information
Russian Text ©The Author(s), 2020, published in Meteorologiya i Gidrologiya, 2020, No. 3, pp. 5–20.
About this article
Cite this article
Bykov, F.L. Statistical Correction of the COSMO Model Weather Forecasts Based on Neural Networks. Russ. Meteorol. Hydrol. 45, 141–152 (2020). https://doi.org/10.3103/S1068373920030012
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.3103/S1068373920030012