用生成流模型加速气候敏感性分析，计算速度提升百倍。
Sensitivity Analysis for Climate Science with Generative Flow Models
- 将伴随状态法引入生成流模型，高效计算梯度。
- 在GPU上仅需数小时，比超算快数周。
- 适合需要快速气候响应分析的研究者。
敏感性分析是气候科学的核心，对理解风暴强度到长期气候反馈等现象至关重要。然而，传统物理模型计算敏感性成本极高，耗时且开发复杂。现代人工智能生成模型虽评估速度快，但其敏感性计算仍存在瓶颈。本文将伴随状态法应用于生成流模型，以计算任意大气变量对海表温度的敏感性。我们使用在ERA5和ICON数据上训练的cBottle模型进行验证，并通过模型自身输出定量评估结果。实验表明该方法可产生可靠梯度，将敏感性分析的计算成本从超算上的数周降至GPU上的数小时，显著简化气候科学中的关键工作流程。代码已公开于https://github.com/Kwartzl8/cbottle_adjoint_sensitivity。
原文摘要 · Abstract (English)
Sensitivity analysis is a cornerstone of climate science, essential for understanding phenomena ranging from storm intensity to long-term climate feedbacks. However, computing these sensitivities using traditional physical models is often prohibitively expensive in terms of both computation and development time. While modern AI-based generative models are orders of magnitude faster to evaluate, computing sensitivities with them remains a significant bottleneck. This work addresses this challenge by applying the adjoint state method for calculating gradients in generative flow models. We apply this method to the cBottle generative model, trained on ERA5 and ICON data, to perform sensitivity analysis of any atmospheric variable with respect to sea surface temperatures. We quantitatively validate the computed sensitivities against the model's own outputs. Our results provide initial evidence that this approach can produce reliable gradients, reducing the computational cost of sensitivity analysis from weeks on a supercomputer with a physical model to hours on a GPU, thereby simplifying a critical workflow in climate science. The code can be found at https://github.com/Kwartzl8/cbottle_adjoint_sensitivity.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。