CloudNine: Analyzing Meteorological Observation Impact on Weather Prediction Using Explainable Graph Neural Networks (2402.14861v1)
Abstract: The impact of meteorological observations on weather forecasting varies with sensor type, location, time, and other environmental factors. Thus, quantitative analysis of observation impacts is crucial for effective and efficient development of weather forecasting systems. However, the existing impact analysis methods are difficult to be widely applied due to their high dependencies on specific forecasting systems. Also, they cannot provide observation impacts at multiple spatio-temporal scales, only global impacts of observation types. To address these issues, we present a novel system called ``CloudNine,'' which allows analysis of individual observations' impacts on specific predictions based on explainable graph neural networks (XGNNs). Combining an XGNN-based atmospheric state estimation model with a numerical weather prediction model, we provide a web application to search for observations in the 3D space of the Earth system and to visualize the impact of individual observations on predictions in specific spatial regions and time periods.
- Weather impact on retail sales: How can weather derivatives help with adverse weather deviations? Journal of Retailing and Consumer Services, 49:1–10, July 2019.
- On the properties of ensemble forecast sensitivity to observations. Quarterly Journal of the Royal Meteorological Society, 145(722):1897–1914, May 2019.
- Development of an operational hybrid data assimilation system at kiaps. Asia-Pacific Journal of Atmospheric Sciences, 54(S1):319–335, June 2018.
- A simpler formulation of forecast sensitivity to observations: application to ensemble kalman filters. Tellus A: Dynamic Meteorology and Oceanography, 64(1):18462, December 2012.
- A new approach for estimating the observation impact in ensemble–variational data assimilation. Monthly Weather Review, 146(2):447–465, January 2018.
- Day-ahead hourly solar irradiance forecasting based on multi-attributed spatio-temporal graph convolutional network. Sensors, 22(19):7179, sep 2022.
- HiSTGNN: Hierarchical spatio-temporal graph neural network for weather forecasting. Information Sciences, 648:119580, nov 2023.
- A cfcc-lstm model for sea surface temperature prediction. IEEE Geoscience and Remote Sensing Letters, 15(2):207–211, February 2018.
- Graph representation learning and its applications: A survey. Sensors, 23(8):4168, April 2023.
- Learning multi-resolution representations of research patterns in bibliographic networks. Journal of Informetrics, 15(1):101126, February 2021.
- Learning skillful medium-range global weather forecasting. Science, page eadi2336, November 2023.
- Explainable graph neural networks for observation impact analysis in atmospheric state estimation. In Proceedings of the Explainable Machine Learning for Sciences Workshop (XAI4Sci) held in conjunction with the 38th AAAI Conference on Artificial Intelligence (AAAI 2024), Vancouver, BC, Canada, February 2024.
- Explainability methods for graph convolutional neural networks. In Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2019). IEEE, jun 2019.
- Explainability techniques for graph convolutional networks. arXiv preprint arXiv:1905.13686, 2019.
- T-GCN: A temporal graph convolutional network for traffic prediction. IEEE Trans. Intell. Transp. Syst., 21(9):3848–3858, 2020.