The troposphere is one of the atmospheric layers where most weather phenomena occur. Temperature variations in the troposphere, especially at 500 hPa, a typical level of the middle troposphere, are significant indicators of future weather changes. Numerical weather prediction is effective for temperature prediction, but its computational complexity hinders a timely response. This paper proposes a novel temperature prediction approach in framework of physics-informed deep learning. The new model, called PGnet, builds upon a generative neural network with a mask matrix. The mask is designed to distinguish the low-quality predicted regions generated by the first physical stage. The generative neural network takes the mask as prior for the second-stage refined predictions. A mask-loss and a jump pattern strategy are developed to train the generative neural network without accumulating errors during making time-series predictions. Experiments on ERA5 demonstrate that PGnet can generate more refined temperature predictions than the state-of-the-art.

对流层是大多数天气现象发生的大气层之一。对流层的温度变化，尤其是 500 hPa（典型的中对流层水平），是未来天气变化的重要指标。数值天气预报对于温度预测是有效的，但其计算复杂性阻碍了及时响应。本文在基于物理的深度学习框架中提出了一种新的温度预测方法。这个名为 PGnet 的新模型建立在带有掩码矩阵的生成神经网络之上。掩码旨在区分第一物理阶段生成的低质量预测区域。生成神经网络将掩码作为第二阶段精细预测的先验。开发了掩码损失和跳跃模式策略来训练生成神经网络，而不会在进行时间序列预测期间累积错误。ERA5 上的实验表明，PGnet 可以生成比最先进技术更精确的温度预测。