Articles | Volume 18, issue 16
https://doi.org/10.5194/gmd-18-5351-2025
© Author(s) 2025. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
https://doi.org/10.5194/gmd-18-5351-2025
© Author(s) 2025. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Model description paper
| Highlight paper
| 
27 Aug 2025
Model description paper | Highlight paper |
| 27 Aug 2025
| 27 Aug 2025
GPTCast: a weather language model for precipitation nowcasting
Fondazione Bruno Kessler, Trento, Italy
Elena Tomasi
Fondazione Bruno Kessler, Trento, Italy
Rishabh Wanjari
Fondazione Bruno Kessler, Trento, Italy
Virginia Poli
Arpae Emilia-Romagna, Bologna, Italy
Chiara Cardinali
Arpae Emilia-Romagna, Bologna, Italy
Pier Paolo Alberoni
Arpae Emilia-Romagna, Bologna, Italy
Marco Cristoforetti
Fondazione Bruno Kessler, Trento, Italy
Related authors
Elena Tomasi, Gabriele Franch, and Marco Cristoforetti
Geosci. Model Dev., 18, 2051–2078, https://doi.org/10.5194/gmd-18-2051-2025, https://doi.org/10.5194/gmd-18-2051-2025, 2025
Short summary
Short summary
High-resolution weather data are crucial for many applications, typically generated via resource-intensive numerical models through dynamical downscaling. We developed an AI model using latent diffusion models (LDMs) to mimic this process, increasing weather data resolution over Italy from 25 to 2 km. LDM outperforms other methods, accurately capturing local patterns and extreme events. This approach offers a cost-effective alternative, with potential disruptive application in climate sciences.
Elena Tomasi, Gabriele Franch, and Marco Cristoforetti
Geosci. Model Dev., 18, 2051–2078, https://doi.org/10.5194/gmd-18-2051-2025, https://doi.org/10.5194/gmd-18-2051-2025, 2025
Short summary
Short summary
High-resolution weather data are crucial for many applications, typically generated via resource-intensive numerical models through dynamical downscaling. We developed an AI model using latent diffusion models (LDMs) to mimic this process, increasing weather data resolution over Italy from 25 to 2 km. LDM outperforms other methods, accurately capturing local patterns and extreme events. This approach offers a cost-effective alternative, with potential disruptive application in climate sciences.
Giacomo Roversi, Pier Paolo Alberoni, Anna Fornasiero, and Federico Porcù
Atmos. Meas. Tech., 13, 5779–5797, https://doi.org/10.5194/amt-13-5779-2020, https://doi.org/10.5194/amt-13-5779-2020, 2020
Short summary
Short summary
The microwave signal travelling between two antennas of the commercial mobile backhaul network is strongly attenuated by rainfall. The open-source RAINLINK algorithm extracts rainfall rate maps, processing the attenuation data recorded by the transmission system. In this work, we applied RAINLINK to 357 Vodafone links in northern Italy and compared the outputs with the operational rain products of the local weather service (Arpae), outlining pros and cons and discussing error structure.
Cited articles
Agrawal, S., Barrington, L., Bromberg, C., Burge, J., Gazen, C., and Hickey, J.: Machine Learning for Precipitation Nowcasting from Radar Images, CoRR, arXiv [preprint], https://doi.org/10.48550/arXiv.1912.12132, 2019. a
Ayzel, G., Scheffer, T., and Heistermann, M.: RainNet v1.0: a convolutional neural network for radar-based precipitation nowcasting, Geosci. Model Dev., 13, 2631–2644, https://doi.org/10.5194/gmd-13-2631-2020, 2020. a, b
Bellon, A. and Austin, G. L.: The evaluation of two years of real-time operation of a short-term precipitation forecasting procedure (SHARP), J. Appl. Meteorol., 17, 1778–1787, 1978. a
Bojinski, S., Blaauboer, D., Calbet, X., de Coning, E., Debie, F., Montmerle, T., Nietosvaara, V., Norman, K., Bañón Peregrín, L., Schmid, F., Strelec Mahović, N., and Wapler, K.: Towards nowcasting in Europe in 2030, Meteorol. Appl., 30, e2124, https://doi.org/10.1002/met.2124, 2023. a
Bowler, N. E., Pierce, C. E., and Seed, A. W.: STEPS: A probabilistic precipitation forecasting scheme which merges an extrapolation nowcast with downscaled NWP, Q. J. Roy. Meteor. Soc., 132, 2127–2155, https://doi.org/10.1256/qj.04.100, 2006. a, b, c
Dao, T., Fu, D. Y., Ermon, S., Rudra, A., and Ré, C.: FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness, Advances in Neural Information Processing Systems (NeurIPS), https://dl.acm.org/doi/10.5555/3600270.3601459 (last access: 20 August 2025), 2022. a
Dixon, M. and Wiener, G.: TITAN: Thunderstorm Identification, Tracking, Analysis, and Nowcasting – A radar-based methodology, J. Atmos. Ocean. Tech., 10, 785–797, 1993. a
Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., Minderer, M., Heigold, G., Gelly, S., Uszkoreit, J., and Houlsby, N.: An image is worth 16x16 words: Transformers for image recognition at scale, arXiv [preprint], https://doi.org/10.48550/arXiv.2010.11929, 2020. a
Esser, P., Rombach, R., and Ommer, B.: Taming transformers for high-resolution image synthesis, in: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, Nashville, TN, USA, 20–25 June 2021, 12873–12883, https://doi.org/10.1109/CVPR46437.2021.01268, 2021. a, b, c, d
Falcon, W. and The PyTorch Lightning team: PyTorch Lightning, Zenodo [code], https://doi.org/10.5281/zenodo.3828935, 2019. a
Fan, A., Lewis, M., and Dauphin, Y.: Hierarchical Neural Story Generation, in: Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (ACL), https://doi.org/10.18653/v1/P18-1082, 2018. a
Foresti, L., Sideris, I. V., Panziera, L., Nerini, D., and Germann, U.: A 10-year radar-based analysis of orographic precipitation growth and decay patterns over the Swiss Alpine region, Q. J. Roy. Meteor. Soc., 144, 2277–2301, https://doi.org/10.1002/qj.3364, 2018. a
Fornasiero, A., Bech, J., and Alberoni, P. P.: Enhanced radar precipitation estimates using a combined clutter and beam blockage correction technique, Nat. Hazards Earth Syst. Sci., 6, 697–710, https://doi.org/10.5194/nhess-6-697-2006, 2006. a
Fornasiero, A., Amorati, R., and Alberoni, P. P.: Radar Quantitative Precipitation Estimation at Arpa-Sim: A Critical Approach to Retrieve the Rainfall Rate at the Ground Level, in: Proceedings of the 5th European Radar Conference, Helsinki, vol. 30, ISBN 9789516976764, 2008. a
Franch, G., Nerini, D., Pendesini, M., Coviello, L., Jurman, G., and Furlanello, C.: Precipitation Nowcasting with Orographic Enhanced Stacked Generalization: Improving Deep Learning Predictions on Extreme Events, Atmosphere, 11, 267, https://doi.org/10.3390/atmos11030267, 2020. a
Franch, G., Tomasi, E., Cardinali, C., Poli, V., Alberoni, P. P., and Cristoforetti, M.: Dataset for “GPTCast: a weather language model for precipitation nowcasting”, Zenodo [data set], https://doi.org/10.5281/zenodo.13692016, 2024a. a
Franch, G., Tomasi, E., and Cristoforetti, M.: Code for “GPTCast: a weather language model for precipitation nowcasting”, Zenodo [code], https://doi.org/10.5281/zenodo.13832526, 2024b. a
Franch, G., Tomasi, E., and Cristoforetti, M.: Pretrained models for “GPTCast: a weather language model for precipitation nowcasting”, Zenodo [code], https://doi.org/10.5281/zenodo.13594332, 2024c. a
Gao, Z., Shi, X., Han, B., Wang, H., Jin, X., Maddix, D., Zhu, Y., Li, M., and Wang, Y.: PreDiff: precipitation nowcasting with latent diffusion models, in: Proceedings of the 37th International Conference on Neural Information Processing Systems (NeurIPS '23), New Orleans, LA, USA, 10–16 December 2023, Curran Associates, Inc., Red Hook, NY, USA, 3439, 36 pp., ISBN 9781713899921, 2023. a
Gholami, A., Kim, S., Dong, Z., Yao, Z., Mahoney, M. W., and Keutzer, K.: A Survey of Quantization Methods for Efficient Neural Network Inference, arXiv [preprint], https://doi.org/10.48550/arXiv.2103.13630, 2021. a
Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., and Bengio, Y.: Generative adversarial nets, Association for Computing Machinery, New York, NY, USA, 139–144, https://doi.org/10.1145/3422622, 2014. a
Göber, M., Christel, I., Hoffmann, D., Mooney, C. J., Rodriguez, L., Becker, N., Ebert, E. E., Fearnley, C., Fundel, V. J., Geiger, T., Golding, B., Jeurig, J., Kelman, I., Kox, T., Magro, F.-A., Perrels, A., Postigo, J. C., Potter, S. H., Robbins, J., Rust, H., Schoster, D., Tan, M. L., Taylor, A., and Williams, H.: Enhancing the Value of Weather and Climate Services in Society: Identified Gaps and Needs as Outcomes of the First WMO WWRP/SERA Weather and Society Conference, B. Am. Meteor. Soc., 104, E645–E651, https://doi.org/10.1175/BAMS-D-22-0199.1, 2023. a
Kuzmin, A., Van Baalen, M., Ren, Y., Nagel, M., Peters, J., and Blankevoort, T.: FP8 quantization: the power of the exponent, in: Proceedings of the 36th International Conference on Neural Information Processing Systems (NeurIPS '22), New Orleans, LA, USA, 28 November–9 December 2022, Curran Associates, Inc., Red Hook, NY, USA, 1065, 12 pp., ISBN 9781713871088, 2022. a
Lam, R., Pascanu, R., Puigdomènech Gimenez, M., Agrawal, S., Dapogny, C., Schmidt, M., Keck, T., Mudigonda, M., Brutlag, P., Wang, J., Chantry, M., Norman, C., Dudhia, A., Clark, R., Otte, N., Tirilly, P., Wiklendt, S., Zimmer, A., Merose, A., Petersen, S., Visram, R., Valter, D., Hess, F., See, A., Fritz, F., Bodin, T., Untema, B., Thurman, R., Targett, P., Ravenscroft, A., McGuire, P., Kabra, M., Keeling, J., Gopal, A., Cheng, H., Piotrowski, T., Battaglia, P., Kohli, P., Heess, N., and Hassabis, D.: GraphCast: AI model for faster and more accurate global weather forecasting, Science, 382, 1416–1421, https://doi.org/10.1126/science.adi2336, 2023. a
Lang, S., Alexe, M., Chantry, M., Dramsch, J., Pinault, F., Raoult, B., Clare, M. C. A., Lessig, C., Maier-Gerber, M., Magnusson, L., Ben Bouallègue, Z., Prieto Nemesio, A., Dueben, P. D., Brown, A., Pappenberger, F., and Rabier, F.: AIFS-ECMWF's data-driven forecasting system, arXiv [preprint], https://doi.org/10.48550/arXiv.2406.01465, 2024. a, b
Leinonen, J., Hamann, U., Nerini, D., Germann, U., and Franch, G.: Latent diffusion models for generative precipitation nowcasting with accurate uncertainty quantification, arXiv [preprint], https://doi.org/10.48550/arXiv.2304.12891, 2023. a, b
Lessig, C., Luise, I., Gong, B., Langguth, M., Stadler, S., and Schultz, M.: AtmoRep: A stochastic model of atmosphere dynamics using large scale representation learning, arXiv [preprint], https://doi.org/10.48550/arXiv.2308.13280, 2023. a
Liu, Z., Lin, Y., Cao, Y., Hu, H., Wei, Y., Zhang, Z., Lin, S., and Guo, B.: Swin transformer: Hierarchical vision transformer using shifted windows, in: Proceedings of the IEEE/CVF international conference on computer vision, 11–17 October 2021, 10012–10022, https://doi.org/10.1109/ICCV48922.2021.00986, 2021. a
Marshall, J. S. and Palmer, W. M. K.: The distribution of raindrops with size, J. Atmos. Sci., 5, 165–166, https://doi.org/10.1175/1520-0469(1948)005<0165:TDORWS>2.0.CO;2, 1948. a
Panziera, L., Germann, U., Gabella, M., and Mandapaka, P. V.: NORA – Nowcasting of Orographic Rainfall by means of Analogues, Q. J. Roy. Meteor. Soc., 137, 2106–2123, https://doi.org/10.1002/qj.878, 2011. a
Pope, R., Douglas, S., Chowdhery, A., Devlin, J., Bradbury, J., Levskaya, A., Heek, J., Xiao, K., Agrawal, S., and Dean, J.: Efficiently scaling transformer inference, arXiv [preprint], https://doi.org/10.48550/arXiv.2211.05102, 2023. a
Pulkkinen, S., Nerini, D., Pérez Hortal, A. A., Velasco-Forero, C., Seed, A., Germann, U., and Foresti, L.: Pysteps: an open-source Python library for probabilistic precipitation nowcasting (v1.0), Geosci. Model Dev., 12, 4185–4219, https://doi.org/10.5194/gmd-12-4185-2019, 2019. a, b
Pulkkinen, S., Chandrasekar, V., von Lerber, A., and Harri, A.-M.: Nowcasting of Convective Rainfall Using Volumetric Radar Observations, IEEE T. Geosci. Remote S., 58, 7845–7859, https://doi.org/10.1109/TGRS.2020.2984594, 2020. a
Pulkkinen, S., Chandrasekar, V., and Niemi, T.: Lagrangian Integro-Difference Equation Model for Precipitation Nowcasting, J. Atmos. Ocean. Tech., 38, 2125–2145, https://doi.org/10.1175/JTECH-D-21-0013.1, 2021. a
Radford, A., Wu, J., Child, R., Luan, D., Amodei, D., and Sutskever, I.: Language Models are Unsupervised Multitask Learners, OpenAI Blog, 1, https://cdn.openai.com/better-language-models/language_models_are_unsupervised_multitask_learners.pdf (last access: 20 August 2025), 2019. a
Ravuri, S., Lenc, K., Willson, M., Kangin, D., Lam, R., Mirowski, P., Fitzsimons, M., Athanassiadou, M., Kashem, S., Madge, S., Prudden, R., Mandhane, A., Clark, A., Brock, A., Simonyan, K., Hadsell, R., Robinson, N., Clancy, E., Arribas, A., and Mohamed, S.: Skilful precipitation nowcasting using deep generative models of radar, Nature, 597, 672–677, 2021. a, b
Ritvanen, J., Pulkkinen, S., Moisseev, D., and Nerini, D.: Cell-tracking-based framework for assessing nowcasting model skill in reproducing growth and decay of convective rainfall, Geosci. Model Dev., 18, 1851–1878, https://doi.org/10.5194/gmd-18-1851-2025, 2025. a
Rombach, R., Blattmann, A., Lorenz, D., Esser, P., and Ommer, B.: High-resolution image synthesis with latent diffusion models, in: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 10684–10695, https://doi.org/10.1109/CVPR46437.2021.01268, 2022. a
Seed, A. W.: A Dynamic and Spatial Scaling Approach to Advection Forecasting, J. Appl. Meteorol., 42, 381–388, https://doi.org/10.1175/1520-0450(2003)042<0381:ADASSA>2.0.CO;2, 2003. a
Seed, A. W., Pierce, C. E., and Norman, K.: Formulation and evaluation of a scale decomposition-based stochastic precipitation nowcast scheme, Water Resour. Res., 49, 6624–6641, https://doi.org/10.1002/wrcr.20536, 2013. a
Shi, X., Chen, Z., Wang, H., Yeung, D.-Y., Wong, W.-K., and Woo, W.-C.: Convolutional LSTM network: A machine learning approach for precipitation nowcasting, Adv. Neur. In., 28, 802–810, ISBN 9781510825024, 2015. a
Sideris, I. V., Foresti, L., Nerini, D., and Germann, U.: NowPrecip: localized precipitation nowcasting in the complex terrain of Switzerland, Q. J. Roy. Meteor. Soc., 146, 1768–1800, https://doi.org/10.1002/qj.3766, 2020. a
Sun, J., Xue, M., Wilson, J. W., Zawadzki, I., Ballard, S. P., Onvlee-Hooimeyer, J., Joe, P., Barker, D. M., Li, P.-W., Golding, B., Xu, M., and Pinto, J.: Use of NWP for Nowcasting Convective Precipitation: Recent Progress and Challenges, B. Am. Meteorol. Soc., 95, 409–426, https://doi.org/10.1175/BAMS-D-11-00263.1, 2014. a
Surcel, M., Zawadzki, I., and Yau, M. K.: A Study on the Scale Dependence of the Predictability of Precipitation Patterns, J. Atmos. Sci., 72, 216–235, https://doi.org/10.1175/JAS-D-14-0071.1, 2015. a
Tomasi, E., Franch, G., and Cristoforetti, M.: Can AI be enabled to perform dynamical downscaling? A latent diffusion model to mimic kilometer-scale COSMO5.0_CLM9 simulations, Geosci. Model Dev., 18, 2051–2078, https://doi.org/10.5194/gmd-18-2051-2025, 2025. a
Turner, B. J., Zawadzki, I., and Germann, U.: Predictability of Precipitation from Continental Radar Images. Part III: Operational Nowcasting Implementation (MAPLE), J. Appl. Meteorol., 43, 231–248, https://doi.org/10.1175/1520-0450(2004)043<0231:POPFCR>2.0.CO;2, 2004. a
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, Ł., and Polosukhin, I.: Attention is all you need, Adv. Neur. In., 30, 5999–6009, ISBN 9781510860964, 2017. a
Wang, Y., Gao, Z., Long, M., Wang, J., and Philip, S. Y.: Predrnn++: Towards a resolution of the deep-in-time dilemma in spatiotemporal predictive learning, in: International conference on machine learning, PMLR, 5123–5132,ISBN 9781510867963, 2018. a
Wang, Z., Bovik, A. C., Sheikh, H. R., and Simoncelli, E. P.: Image quality assessment: from error visibility to structural similarity, IEEE T. Image Process., 13, 600–612, 2004. a
Werner, M. and Cranston, M.: Understanding the value of radar rainfall nowcasts in flood forecasting and warning in flashy catchments, Meteorological Applications: A journal of forecasting, practical applications, Training Techniques And Modelling, 16, 41–55, 2009. a
Wernli, H., Paulat, M., Hagen, M., and Frei, C.: SAL – A novel quality measure for the verification of quantitative precipitation forecasts, Mon. Weather Rev., 136, 4470–4487, 2008. a
Wernli, H., Hofmann, C., and Zimmer, M.: Spatial forecast verification methods intercomparison project: Application of the SAL technique, Weather Forecast., 24, 1472–1484, 2009. a
Wolf, T., Debut, L., Sanh, V., Chaumond, J., Delangue, C., Moi, A., Cistac, P., Rault, T., Louf, R., Funtowicz, M., Davison, J., Shleifer, S., von Platen, P., Ma, C., Jernite, Y., Plu, J., Xu, C., Le Scao, T., Gugger, S., Drame, M., Lhoest, Q., and Rush, A.: Transformers: State-of-the-Art Natural Language Processing, in: Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing: System Demonstrations, edited by: Liu, Q. and Schlangen, D., Association for Computational Linguistics, Online, 38–45, https://doi.org/10.18653/v1/2020.emnlp-demos.6, 2020. a
Woo, W.-C. and Wong, W.-K.: Operational Application of Optical Flow Techniques to Radar-Based Rainfall Nowcasting, Atmosphere, 8, 48, https://doi.org/10.3390/atmos8030048, 2017. a
Yadan, O.: Hydra – A framework for elegantly configuring complex applications, Github [code], https://github.com/facebookresearch/hydra (last access: 20 August 2025), 2019. a
Yu, J., Li, X., Koh, J. Y., Zhang, H., Pang, R., Qin, J., Ku, A., Xu, Y., Baldridge, J., and Wu, Y.: Vector-quantized Image Modeling with Improved VQGAN, in: International Conference on Learning Representations, https://openreview.net/forum?id=pfNyExj7z2 (last access: 20 August 2025), 2022. a
Zhang, R., Isola, P., Efros, A. A., Shechtman, E., and Wang, O.: The unreasonable effectiveness of deep features as a perceptual metric, in: Proceedings of the IEEE conference on computer vision and pattern recognition, 586–595, https://doi.ieeecomputersociety.org/10.1109/CVPR.2018.00068 (last access: 20 August 2025), 2018. a
Executive editor
The application of machine learning techniques to weather forecasting is an exceptionally promising area for this technology. This paper presents an LLM nowcasting tool which outperforms existing technology for short term precipitation forecasting. This is an exciting demonstrator of the possibilities of this novel approach.
The application of machine learning techniques to weather forecasting is an exceptionally...
Short summary
Our research introduces GPTCast, a novel method for very short term precipitation forecasting using radar data. By applying advanced machine learning techniques inspired by large language models, we developed a system that generates accurate and realistic weather predictions. We trained the model using 6 years of radar data from northern Italy, demonstrating its superior performance over leading ensemble extrapolation methods.
Our research introduces GPTCast, a novel method for very short term precipitation forecasting...
