Generalizing Weather Forecast to Fine-grained Temporal Scales via Physics-AI Hybrid Modeling
This work addresses the challenge of fine-grained temporal forecasting in meteorology, offering a solution for more detailed weather predictions, though it is incremental in combining existing physics and AI techniques.
The paper tackles the problem of data-driven weather forecasting models being limited to coarse temporal scales by proposing a physics-AI hybrid model that generalizes forecasts to finer-grained scales, such as 30-minute intervals, even when trained on hourly data.
Data-driven artificial intelligence (AI) models have made significant advancements in weather forecasting, particularly in medium-range and nowcasting. However, most data-driven weather forecasting models are black-box systems that focus on learning data mapping rather than fine-grained physical evolution in the time dimension. Consequently, the limitations in the temporal scale of datasets prevent these models from forecasting at finer time scales. This paper proposes a physics-AI hybrid model (i.e., WeatherGFT) which generalizes weather forecasts to finer-grained temporal scales beyond training dataset. Specifically, we employ a carefully designed PDE kernel to simulate physical evolution on a small time scale (e.g., 300 seconds) and use a parallel neural networks with a learnable router for bias correction. Furthermore, we introduce a lead time-aware training framework to promote the generalization of the model at different lead times. The weight analysis of physics-AI modules indicates that physics conducts major evolution while AI performs corrections adaptively. Extensive experiments show that WeatherGFT trained on an hourly dataset, effectively generalizes forecasts across multiple time scales, including 30-minute, which is even smaller than the dataset's temporal resolution.