用生成模型将气候模拟数据高精度还原，保持时空一致性。
A Generative Framework for Probabilistic, Spatiotemporally Coherent Downscaling of Climate Simulation
- 基于扩散模型学习高分辨率气象数据的统计特性。
- 在粗粒度气候数据上生成多条符合物理规律的高分辨率路径。
- 适合需要长期、多变量气候预测的研究与决策场景。
局部气候信息对影响评估和决策至关重要，但全球气候模拟分辨率过低，难以捕捉小尺度现象。现有统计降尺度方法将这些现象视为时间解耦的空间片段，无法保持物理一致性。本文提出一种新型生成框架，利用基于评分的扩散模型在高分辨率再分析数据上训练，捕捉局地天气动力学的统计特征。训练完成后，以粗粒度气候模型输出为条件，生成与整体信息一致的高分辨率天气模式。由于该任务存在固有不确定性，我们借助扩散模型的随机性，生成多条轨迹进行采样。我们在高分辨率再分析数据上验证了方法有效性，随后将其应用于气候模型降尺度任务。结果表明，模型生成的天气动态在空间和时间上均具高度一致性，且与全球气候输出吻合。
原文摘要 · Abstract (English)
Local climate information is crucial for impact assessment and decision-making, yet coarse global climate simulations cannot capture small-scale phenomena. Current statistical downscaling methods infer these phenomena as temporally decoupled spatial patches. However, to preserve physical properties, estimating spatio-temporally coherent high-resolution weather dynamics for multiple variables across long time horizons is crucial. We present a novel generative framework that uses a score-based diffusion model trained on high-resolution reanalysis data to capture the statistical properties of local weather dynamics. After training, we condition on coarse climate model data to generate weather patterns consistent with the aggregate information. As this predictive task is inherently uncertain, we leverage the probabilistic nature of diffusion models and sample multiple trajectories. We evaluate our approach with high-resolution reanalysis information before applying it to the climate model downscaling task. We then demonstrate that the model generates spatially and temporally coherent weather dynamics that align with global climate output.
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。