Skip to main content
QUICK REVIEW

[论文解读] Fuxi-DA: A Generalized Deep Learning Data Assimilation Framework for Assimilating Satellite Observations

Xiaoze Xu, Xiuyu Sun|arXiv (Cornell University)|Apr 12, 2024
Meteorological Phenomena and Simulations被引用 4
一句话总结

Fuxi-DA 是一种广义的基于深度学习的数据同化框架,能够高效地将风云四号B星AGRI仪器的卫星辐射观测资料同化到数值天气预报系统中。通过采用类似U-Net的网络结构,结合多尺度特征融合与多时间步损失函数,该方法显著降低了分析误差,并提升了6–60小时的天气预报性能,在全球和区域尺度指标上均优于基线模型。

ABSTRACT

Data assimilation (DA), as an indispensable component within contemporary Numerical Weather Prediction (NWP) systems, plays a crucial role in generating the analysis that significantly impacts forecast performance. Nevertheless, the development of an efficient DA system poses significant challenges, particularly in establishing intricate relationships between the background data and the vast amount of multi-source observation data within limited time windows in operational settings. To address these challenges, researchers design complex pre-processing methods for each observation type, leveraging approximate modeling and the power of super-computing clusters to expedite solutions. The emergence of deep learning (DL) models has been a game-changer, offering unified multi-modal modeling, enhanced nonlinear representation capabilities, and superior parallelization. These advantages have spurred efforts to integrate DL models into various domains of weather modeling. Remarkably, DL models have shown promise in matching, even surpassing, the forecast accuracy of leading operational NWP models worldwide. This success motivates the exploration of DL-based DA frameworks tailored for weather forecasting models. In this study, we introduces FuxiDA, a generalized DL-based DA framework for assimilating satellite observations. By assimilating data from Advanced Geosynchronous Radiation Imager (AGRI) aboard Fengyun-4B, FuXi-DA consistently mitigates analysis errors and significantly improves forecast performance. Furthermore, through a series of single-observation experiments, Fuxi-DA has been validated against established atmospheric physics, demonstrating its consistency and reliability.

研究动机与目标

  • 开发一种广义的基于深度学习的数据同化框架,用于将多样化卫星观测资料同化到数值天气预报系统中。
  • 克服传统变分数据同化方法的局限性,包括计算成本高、观测算子设计复杂,以及对多云天气观测利用有限的问题。
  • 通过整合风云四号B星AGRI仪器的辐射资料,提升初始场的准确性,这些资料在业务系统中常被低估或未被充分利用。
  • 通过单观测实验与ERA5再分析数据对比,验证该框架的物理一致性和可靠性。
  • 证明基于深度学习的数据同化方法在中期数值天气预报中可达到甚至超越领先业务NWP模型的性能。

提出的方法

  • Fuxi-DA采用类似U-Net的编码器-解码器结构,包含跳跃连接,使用3×3卷积层、层归一化和SiLU激活函数。
  • 模型采用多尺度特征融合机制,将下采样与上采样阶段的特征进行拼接,以保留空间与通道维度的信息。
  • 像素重排(pixel-shuffle)层在每个上采样阶段实现2倍的特征图上采样。
  • 训练采用多时间步损失函数,结合分析误差(t=0)与预报误差(t=1至t=10),并引入纬度加权L1损失以考虑球面几何特性。
  • 损失函数通过AdamW优化,采用学习率热身与余弦退火调度策略,在4块A100 GPU上训练6000轮迭代。
  • 模型以ERA5再分析数据作为真实标签进行训练,卫星观测作为输入,以重建精确的大气状态场。
Figure 1: The comparison between the Fuxi-DA and the variational data assimilation.
Figure 1: The comparison between the Fuxi-DA and the variational data assimilation.

实验结果

研究问题

  • RQ1基于深度学习的数据同化框架能否有效降低同化风云四号B星AGRI辐射观测资料时的分析误差?
  • RQ2在6–60小时预报范围内,Fuxi-DA与基线模型(如FuXi和仅校正模型)相比,预报精度如何?
  • RQ3在同化单个观测时,Fuxi-DA是否保持物理一致性,是否符合大气物理规律?
  • RQ4卫星资料的引入在多大程度上改善了大气状态的表征,特别是在多云和降水条件下?
  • RQ5多时间步损失策略是否通过联合优化分析场与预报场,提升长期预报性能?

主要发现

  • 与对照实验(EXP_CTRL)相比,Fuxi-DA在分析场及6–60小时预报中均显著降低了纬度加权均方根误差(RMSE),且在全部10个预报时间步长上均观测到误差降低。
  • 同化实验(EXP_ASSI)相比EXP_CTRL的归一化RMSE差异为-12.3%,表明在6–60小时预报窗口内误差相对减少了12.3%。
  • 在东亚区域评估中,Fuxi-DA相比对照模型将RMSE降低了14.1%,表明其在动力活跃区域具有更强的性能表现。
  • 单观测实验结果表明,Fuxi-DA能够产生符合物理规律的响应,验证了其可靠性与可解释性。
  • 仅通过模型校正训练的校正模型(EXP_CORR)表现出更高的基线误差,证实Fuxi-DA的性能提升源于有效同化,而非单纯模型校正。
  • 多时间步损失函数通过联合最小化分析误差与预报误差,提升了预报技巧,表现为所有预报时长上误差持续降低。
Figure 2: Comparison of analysis after a single assimilation. The time series of the globally-averaged latitude-weighted RMSE (a, b, c, d) and the time series of the normalized RMSE difference (e, f, g, h) of EXP_CTRL (black lines), EXP_CORR (blue lines) and EXP_ASSI (red lines), and the spatial dis
Figure 2: Comparison of analysis after a single assimilation. The time series of the globally-averaged latitude-weighted RMSE (a, b, c, d) and the time series of the normalized RMSE difference (e, f, g, h) of EXP_CTRL (black lines), EXP_CORR (blue lines) and EXP_ASSI (red lines), and the spatial dis

更好的研究,从现在开始

从阅读论文到最终审阅,大幅缩短您的研究时间。

无需绑定信用卡

本解读由 AI 生成,并经人工编辑审核。