Interpolação e Previsão de Precipitação por Redes Neurais Convolucionais para Grafos

  • Augusto Fonseca CEFET/RJ
  • Ronaldo Goldschimidt IME
  • Eduardo Ogasawara CEFET/RJ
  • Mariza Ferro UFF
  • Fábio Porto LNCC
  • Eduardo Bezerra CEFET/RJ

Resumo


Statistical interpolation methods used to adapt data do not adequately capture the spatiotemporal dependencies of meteorological data. This work proposes the use of Graph Convolutional Neural Networks (GCNs) for precipitation interpolation and forecasting. The experiments utilized radar data and in-situ station data. Comparing GCN interpolation with the statistical Inverse Distance Weighting (IDW) method, preliminary results indicate that GCNs exhibit better accuracy in extreme precipitation events, also showing promise in events of lesser magnitude.

Palavras-chave: redes neurais, interpolação de dados, previsão de precipitação

Referências

Jiandong Bai, Jiawei Zhu, Yujiao Song, Ling Zhao, Zhixiang Hou, Ronghua Du, and Haifeng Li. 2021. A3T-GCN: Attention Temporal Graph Convolutional Network for Traffic Forecasting. ISPRS International Journal of Geo-Information 10, 7 (2021). DOI: 10.3390/ijgi10070485

Rafaela Castro, Yania M Souto, Eduardo Ogasawara, Fabio Porto, and Eduardo Bezerra. 2021. STconvS2S: Spatiotemporal convolutional sequence to sequence network for weather forecasting. Neurocomputing 426 (2021), 285–298.

ECMWF. 2024. Coupled Earth-system reanalysis. European Centre for Medium-Range Weather Forecasts. [link] Acessado: 22-06-2024.

Porto et al. 2022. Machine Learning Approaches to Extreme Weather Events Forecast in Urban Areas: Challenges and Initial Results. Supercomputing Frontiers and Innovations 9 (2022), 49–73. DOI: 10.14529/jsfi220104

Chris Funk, Pete Peterson, Martin Landsfeld, Diego Pedreros, James Verdin, Shraddhanand Shukla, Gregory Husak, James Rowland, Laura Harrison, Andrew Hoell, et al. 2015. The climate hazards infrared precipitation with stations—a new environmental record for monitoring extremes. Scientific data 2, 1 (2015), 1–21.

Gabriela Lewenfus, Wallace Alves Martins, Symeon Chatzinotas, and Björn Ottersten. 2020. Joint Forecasting and Interpolation of Graph Signals Using Deep Learning. arXiv preprint arXiv:2006.01536 (2020).

Brant Liebmann and Dave Allured. 2005. Daily precipitation grids for South America. Bulletin of the American Meteorological Society 86, 11 (2005), 1567–1570.

Xiaoli Ren, Xiaoyong Li, Kaijun Ren, Junqiang Song, Zichen Xu, Kefeng Deng, and Xiang Wang. 2021. Deep Learning-Based Weather Prediction: A Survey. Big Data Research 54 (2021), 100178. DOI: 10.1016/j.bdr.2020.100178

Donald Shepard. 1968. A two-dimensional interpolation function for irregularly-spaced data. In Proceedings of the 1968 23rd ACM National Conference (ACM ’68). Association for Computing Machinery, New York, NY, USA, 517–524. DOI: 10.1145/800186.810616

Selim Furkan Tekin, Oguzhan Karaahmetoglu, Fatih Ilhan, Ismail Balaban, and Suleyman Serdar Kozat. 2021. Spatio-temporal Weather Forecasting and Attention Mechanism on Convolutional LSTMs. arXiv preprint arXiv:2102.00696 (2021).

Hongye Zhou, Feng Zhang, Zhenhong Du, and Renyi Liu. 2021. Forecasting PM2. 5 using hybrid graph convolution-based model considering dynamic wind-field to offer the benefit of spatial interpretability. Environmental Pollution 273 (2021), 116473.
Publicado
14/10/2024
FONSECA, Augusto; GOLDSCHIMIDT, Ronaldo; OGASAWARA, Eduardo; FERRO, Mariza; PORTO, Fábio; BEZERRA, Eduardo. Interpolação e Previsão de Precipitação por Redes Neurais Convolucionais para Grafos. In: BRAZILIAN SYMPOSIUM ON MULTIMEDIA AND THE WEB (WEBMEDIA), 30. , 2024, Juiz de Fora/MG. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2024 . p. 395-399. DOI: https://doi.org/10.5753/webmedia.2024.242049.

Artigos mais lidos do(s) mesmo(s) autor(es)

<< < 1 2