Assessing U-Net Model Performance in Weather Radar-Based Precipitation Nowcasting: A Reflectivity Threshold Analysis
Resumo
Severe weather events pose a global threat, causing property damage and endangering lives. Weather-related disasters account for 50% of all natural and technological disasters. The development of accurate prediction systems are crucial for early warnings and mitigation. The present study evaluates the effectiveness of the U-Net model in weather radar-based precipitation nowcasting, considering reflectivity thresholds. Visual comparison and evaluation metrics are used to assess observed and predicted reflectivity. The 10 dBZ threshold achieved a prominent result, accurately predicting over 75% of values above the reflectivity threshold. Results contribute to improving severe weather prediction and decision-making.
Referências
Ayzel, G., Scheffer, T., and Heistermann, M. (2020). Rainnet v1. 0: a convolutional neural network for radar-based precipitation nowcasting. Geoscientific Model Development, 13(6):2631–2644.
Bonnet, S. M., Evsukoff, A., and Morales Rodriguez, C. A. (2020). Precipitation nowcasting with weather radar images and deep learning in são paulo, brasil. Atmosphere, 11(11):1157.
Brasil (2022). Ministério do desenvolvimento regional. Secretaria de Proteção e Defesa Civil. Universidade Federal de Santa Catarina. Centro de Estudos e Pesquisas em Engenharia e Defesa Civil. Atlas Digital de Desastres no Brasil. Brasília: MDR, 2022.
Gao, Z., Shi, X., Wang, H., Zhu, Y., Wang, Y. B., Li, M., and Yeung, D.-Y. (2022). Earth-former: Exploring space-time transformers for earth system forecasting. Advances in Neural Information Processing Systems, 35:25390–25403.
Kim, S., Hong, S., Joh, M., and Song, S.-k. (2017). Deeprain: Convlstm network for precipitation prediction using multichannel radar data. arXiv preprint arXiv:1711.02316.
Ravuri, S., Lenc, K., Willson, M., Kangin, D., Lam, R., Mirowski, P., Fitzsimons, M., Athanassiadou, M., Kashem, S., Madge, S., et al. (2021). Skilful precipitation nowcasting using deep generative models of radar. Nature, 597(7878):672–677.
Ronneberger, O., Fischer, P., and Brox, T. (2015). U-net: Convolutional networks for biomedical image segmentation. In Medical Image Computing and Computer-Assisted Intervention–MICCAI 2015: 18th International Conference, Munich, Germany, October 5-9, 2015, Proceedings, Part III 18, pages 234–241. Springer.
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.
Shi, X., Gao, Z., Lausen, L., Wang, H., Yeung, D.-Y., Wong, W.-k., and Woo, W.-c. (2017). Deep learning for precipitation nowcasting: A benchmark and a new model. Advances in neural information processing systems, 30.
Trebing, K., Stanczyk, T., and Mehrkanoon, S. (2021). Smaat-unet: Precipitation nowcasting using a small attention-unet architecture. Pattern Recognition Letters, 145:178–186.
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, Ł., and Polosukhin, I. (2017). Attention is all you need. Advances in neural information processing systems, 30.
Zhongming, Z., Linong, L., Xiaona, Y., Wangqiang, Z., Wei, L., et al. (2021). Atlas of mortality and economic losses from weather, climate and water extremes (1970-2019). WMO.