Integrating Weather Station Data and Radar for Precipitation Nowcasting: SmaAt-fUsion and SmaAt-Krige-GNet

2025-02-22

AI Summary

This paper addresses the limited accuracy of radar-only nowcasting models for precipitation due to the neglect of ground-based multivariate meteorological observations. To overcome this limitation, we propose a dual deep learning fusion framework. Methodologically, we introduce two novel architectures: (1) SmaAt-fUsion—an end-to-end convolutional-level fusion network—and (2) SmaAt-Krige-GNet—a Kriging interpolation–enhanced dual-encoder architecture—enabling the first implementation of hierarchical, variable-aware integration of sparse, multivariate in-situ station data into radar-based forecasting models. Evaluated on four years of high-resolution observational data from the Netherlands, SmaAt-fUsion consistently outperforms radar-only baselines across all precipitation intensity levels, while SmaAt-Krige-GNet achieves particularly pronounced gains in light-precipitation regimes. Results demonstrate that assimilating heterogeneous ground observations significantly improves the spatiotemporal modeling capability and robustness of precipitation nowcasting systems.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Multimodal Learning
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

In recent years, data-driven, deep learning-based approaches for precipitation nowcasting have attracted significant attention, showing promising results. However, many existing models fail to fully exploit the extensive atmospheric information available, relying primarily on precipitation data alone. This study introduces two novel deep learning architectures, SmaAt-fUsion and SmaAt-Krige-GNet, specifically designed to enhance precipitation nowcasting by integrating multi-variable weather station data with radar datasets. By leveraging additional meteorological information, these models improve representation learning in the latent space, resulting in enhanced nowcasting performance. The SmaAt-fUsion model extends the SmaAt-UNet framework by incorporating weather station data through a convolutional layer, integrating it into the bottleneck of the network. Conversely, the SmaAt-Krige-GNet model combines precipitation maps with weather station data processed using Kriging, a geo-statistical interpolation method, to generate variable-specific maps. These maps are then utilized in a dual-encoder architecture based on SmaAt-GNet, allowing multi-level data integration. Experimental evaluations were conducted using four years (2016--2019) of weather station and precipitation radar data from the Netherlands. Results demonstrate that SmaAt-Krige-GNet outperforms the standard SmaAt-UNet, which relies solely on precipitation radar data, in low precipitation scenarios, while SmaAt-fUsion surpasses SmaAt-UNet in both low and high precipitation scenarios. This highlights the potential of incorporating discrete weather station data to enhance the performance of deep learning-based weather nowcasting models.

Problem

• Enhance precipitation nowcasting accuracy
• Integrate multi-variable weather station data
• Leverage deep learning for atmospheric information

Innovation

• Integrates weather station and radar data
• Uses SmaAt-fUsion and SmaAt-Krige-GNet models
• Enhances precipitation nowcasting accuracy