arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2511.01468cs.LGcs.AI
收藏

DAMBench:一个基于深度学习的大气数据同化多模态基准

DAMBench: A Multi-Modal Benchmark for Deep Learning-based Atmospheric Data Assimilation

The Hong Kong University of Science and Technology (Guangzhou)(香港科学与技术大学(广州));The Hong Kong University of Science and Technology(香港科学与技术大学)

Hao Wang, Zixuan Weng, Jindong Han, Wei Fan, Hao Liu

AI总结:

针对现有深度学习数据同化研究依赖简化场景且缺乏标准基准的问题,本文提出首个大规模多模态基准DAMBench,整合真实气象站与卫星观测数据,并提供统一评估协议与轻量级多模态插件,为该领域奠定了可复现与公平比较的严格基础。

摘要:

数据同化是大气系统建模的基石,其任务是通过将稀疏、含噪的观测与先验估计相结合来重构系统状态。尽管变分和集合卡尔曼滤波等传统方法已被证明有效,但深度学习的最新进展提供了更具可扩展性、高效性和灵活性的替代方案,更适合涉及大规模和多模态观测的复杂现实世界数据同化。然而,现有的基于深度学习的DA研究存在两个关键局限:(1)依赖对观测进行合成扰动的过度简化场景,以及(2)缺乏用于公平模型比较的标准化基准。为了填补这些空白,在这项工作中,我们引入了DAMBench,这是首个旨在评估真实大气条件下数据驱动DA模型的大规模多模态基准。DAMBench整合了来自最先进预报系统的高质量背景状态和现实世界的多模态观测(即现实世界的气象站和卫星图像)。所有数据都被重采样到通用网格并进行时间对齐,以支持系统的训练、验证和测试。我们提供了统一的评估协议,并对具有代表性的数据同化方法进行了基准测试,包括潜在生成模型和神经过程框架。此外,我们提出了一个轻量级多模态插件,以展示如何通过整合现实观测来增强即使是简单的基线。通过全面的实验,DAMBench为未来研究奠定了严格的基础,促进了可复现性、公平比较以及对现实世界多模态场景的可扩展性。我们的数据集和代码已在 https://github.com/figerhaowang/DAMBench 公开。

英文摘要:

Data Assimilation is a cornerstone of atmospheric system modeling, tasked with reconstructing system states by integrating sparse, noisy observations with prior estimation. While traditional approaches like variational and ensemble Kalman filtering have proven effective, recent advances in deep learning offer more scalable, efficient, and flexible alternatives better suited for complex, real-world data assimilation involving large-scale and multi-modal observations. However, existing deep learning-based DA research suffers from two critical limitations: (1) reliance on oversimplified scenarios with synthetically perturbed observations, and (2) the absence of standardized benchmarks for fair model comparison. To address these gaps, in this work, we introduce DAMBench, the first large-scale multi-modal benchmark designed to evaluate data-driven DA models under realistic atmospheric conditions. DAMBench integrates high-quality background states from state-of-the-art forecasting systems and real-world multi-modal observations (i.e., real-world weather stations and satellite imagery). All data are resampled to a common grid and temporally aligned to support systematic training, validation, and testing. We provide unified evaluation protocols and benchmark representative data assimilation approaches, including latent generative models and neural process frameworks. Additionally, we propose a lightweight multi-modal plugin to demonstrate how integrating realistic observations can enhance even simple baselines. Through comprehensive experiments, DAMBench establishes a rigorous foundation for future research, promoting reproducibility, fair comparison, and extensibility to real-world multi-modal scenarios. Our dataset and code are publicly available at https://github.com/figerhaowang/DAMBench.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑