Deeplearning-based approach to improving numerical weather forecasts

А.Yu. Doroshenko, V.M. Shpyg, R.V. Kushnirenko


This paper briefly describes the history of numerical weather prediction development. The difficulties, which occur in the modelling of atmospheric processes, their nature and possible ways of their mitigation, are described. It also indicates alternative methods of improving the quality of meteorological forecasts. A brief history of deep learning and possible ways of its application to meteorological problems are given. Then, the paper describes the format used to store the 2m temperature forecasts of the COSMO numerical regional model. The proposed neural network architecture enables correcting the forecast errors of the numerical model. We conducted the experiments on the data of eight meteorological stations of the Kyiv region, so we obtained eight trained neural network models. The results showed that the proposed architecture enables obtaining better-quality forecasts in more than 50% of cases. Root-mean-square errors of the resulting forecasts decreased, and it is a widespread skill-score of improved-quality forecasts in meteorological science.

Prombles in programming 2023; 3: 91-98


deep learning; numerical weather prediction; COSMO; 2m temperature


Bauer, P., Thorpe, A. and Brunet, G., 2015. The quiet revolution of numerical weather prediction. Nature, 525(7567), pp.47-55. CrossRef

Orlanski, I., 1975. A rational subdivision of scales for atmospheric processes. Bulletin of the American Meteorological Society, pp.527-530.

Prusov, V.A., Doroshenko, A.E., Chernysh, R.I. and Guk, L.N., 2007. Efficient difference scheme for numerical solution of a convective diffusion problem. Cybernetics and Systems Analysis, 43, pp.368-376. CrossRef

Prusov, V.A. and Doroshenko, A.Y., 2006. Modeling of Natural and Technogenic Processes in the Atmosphere [in Ukrainian].

LeCun, Y., Bottou, L., Bengio, Y. and Haffner, P., 1998. Gradient based learning applied to document recognition. Proceedings of the IEEE, 86(11), pp.2278-2324. CrossRef

Zhu, X.X., Tuia, D., Mou, L., Xia, G.S., Zhang, L., Xu, F. and Fraundorfer, F., 2017. Deep learning in remote sensing: A comprehensive review and list of resources. IEEE geoscience and remote sensing magazine, 5(4), pp.8-36. CrossRef

Graves, A., Liwicki, M., Fernandez, S., Bertolami, R., Bunke, H. and Schmidhuber, J., 2009. A novel connectionist system for improved unconstrained handwriting. IEEE Transactions on Pattern Analysis and Machine Intelligence, 31(5). CrossRef

Sak, H., Senior, A. and Beaufays, F., 2014. Long short-term memory based recurrent neural network architectures for large vocabulary speech recognition. arXiv preprint arXiv:1402.1128. CrossRef

Hochreiter, S. and Schmidhuber, J., 1997. Long short-term memory. Neural computation, 9(8), pp.1735-1780. CrossRef

Cho, K., Van Merriënboer, B., Gulcehre, C., Bahdanau, D., Bougares, F., Schwenk, H. and Bengio, Y., 2014. Learning phrase representations using RNN encoder-decoder for statistical machine translation. arXiv preprint arXiv:1406.1078. CrossRef

Shi, X., Chen, Z., Wang, H., Yeung, D.Y., Wong, W.K. and Woo, W.C., 2015. Convolutional LSTM network: A machine learning approach for precipitation nowcasting. Advances in neural information processing systems, 28.

Zhou, Y., Wang, H., Xu, F. and Jin, Y.Q., 2016. Polarimetric SAR image classification using deep convolutional neural networks. IEEE Geoscience and Remote Sensing Letters, 13(12), pp.1935-1939. CrossRef

Xu, Z., Du, J., Wang, J., Jiang, C. and Ren, Y., 2019, May. Satellite image prediction relying on GAN and LSTM neural networks. In ICC 2019-2019 IEEE international conference on communications (ICC) (pp. 1-6). IEEE. CrossRef

Schmidt, V., Alghali, M., Sankaran, K., Yuan, T. and Bengio, Y., 2020. Modeling cloud reflectance fields using conditional generative adversarial networks. arXiv preprint arXiv:2002.07579.

Dueben, P.D. and Bauer, P., 2018. Challenges and design choices for global weather and climate models based on machine learning. Geoscientific Model Development, 11(10), pp.3999-4009. CrossRef

Wandel, N., Weinmann, M. and Klein, R., 2020. Unsupervised deep learning of incompressible fluid dynamics. arXiv preprint arXiv:2006.08762.

Schultz, M.G., Betancourt, C., Gong, B., Kleinert, F., Langguth, M., Leufen, L.H., Mozaffari, A. and Stadtler, S., 2021. Can deep learning beat numerical weather prediction?. Philosophical Transactions of the Royal Society A, 379(2194), p.20200097. CrossRef

Bauer, P., Dueben, P.D., Hoefler, T., Quintino, T., Schulthess, T.C. and Wedi, N.P., 2021. The digital revolution of Earth-system science. Nature Computational Science, 1(2), pp.104-113. CrossRef

Prudden, R., Adams, S., Kangin, D., Robinson, N., Ravuri, S., Mohamed, S. and Arribas, A., 2020. A review of radar-based nowcasting of precipitation and applicable machine learning techniques. arXiv preprint arXiv:2005.04988.

Bonavita, M. and Laloyaux, P., 2020. Machine learning for model error inference and correction. Journal of Advances in Modeling Earth Systems, 12(12), p.e2020MS002232. CrossRef

Krasnopolsky, V.M., Fox-Rabinovitz, M.S. and Chalikov, D.V., 2005. New approach to calculation of atmospheric model physics: Accurate and fast neural network emulation of longwave radiation in a climate model. Monthly Weather Review, 133(5), pp.1370-1383. CrossRef

Rasp, S. and Lerch, S., 2018. Neural networks for postprocessing ensemble weather forecasts. Monthly Weather Review, 146(11), pp.3885-3900. CrossRef

Shpyg, V., Budak, I., Pishniak, D. and Poperechnyi, P., 2013, November. The application of regional NWP models to operational weather forecasting in Ukraine. In CAS Technical Conference (TECO) on» Responding to the Environmental Stressors of the 21st Century» Available from: http://www. wmo. int/pages/prog/arep/cas/ documents/Ukraine-NWPModels. pdf [Accessed 27/02/2020].

Doms, G. and Baldauf, M., 2011. A description of the nonhydrostatic regional COSMO-Model Part I: dynamics and numerics. Deutscher Wetterdienst, Offenbach.



  • There are currently no refbacks.