Accurate short-term precipitation forecasting is critical for weather-sensitive decision-making in agriculture, transportation, and disaster response. Existing deep learning approaches often struggle to balance global structural consistency with local detail preservation, especially under complex meteorological conditions. We propose DuoCast, a dual-diffusion framework that decomposes precipitation forecasting into low- and high-frequency components modeled in orthogonal latent subspaces. We theoretically prove that this frequency decomposition reduces prediction error compared to conventional single branch U-Net diffusion models. In DuoCast, the low-frequency model captures large-scale trends via convolutional encoders conditioned on weather front dynamics, while the high-frequency model refines fine-scale variability using a self-attention-based architecture. Experiments on four benchmark radar datasets show that DuoCast consistently outperforms state-of-the-art baselines, achieving superior accuracy in both spatial detail and temporal evolution.

翻译：准确的短期降水预报对于农业、交通和灾害响应等天气敏感领域的决策至关重要。现有的深度学习方法往往难以平衡全局结构一致性与局部细节保持，尤其是在复杂气象条件下。我们提出了DuoCast，一种双扩散框架，将降水预报分解为在正交潜在子空间中建模的低频和高频分量。我们从理论上证明，与传统的单分支U-Net扩散模型相比，这种频率分解降低了预测误差。在DuoCast中，低频模型通过以天气锋面动力学为条件的卷积编码器捕捉大尺度趋势，而高频模型则使用基于自注意力的架构细化精细尺度的变异性。在四个基准雷达数据集上的实验表明，DuoCast始终优于最先进的基线方法，在空间细节和时间演化方面均实现了更优的准确性。