发表机构
Faculty of Geosciences and Environment, University of Lausanne; Expertise Center for Climate Extremes, University of Lausanne; Max Planck Institute for Biogeochemistry; Institute of Geography - Oeschger Centre for Climate Change Research, University of Bern; Institute for Atmospheric and Climate Science, ETH Zurich（洛桑大学地球科学与环境学院; 洛桑大学气候极端现象研究中心; 马克斯·普朗克生物地球化学研究所; 伯尔尼大学地理研究所-厄施格气候变化研究中心; 苏黎世联邦理工学院大气与气候科学研究所）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文对比WRF动力降尺度与扩散生成模型对次季节极端降水的降尺度能力，发现两者均优于原始预报，且WRF擅长非平稳事件，扩散模型在平稳事件中更优。
AI 中文摘要
粗空间分辨率限制了次季节预测模型解析极端降水的能力。使用动力或深度生成模型进行降尺度可以克服这一问题，但不同大气环流背景下这些模型在极端事件上的比较性能仍知之甚少。在本工作中，我们将天气研究与预报（WRF）模型与基于扩散的生成模型进行对比，对两个物理性质不同的极端降水事件进行降尺度，预报提前期长达3周。为了与WRF进行公平比较——WRF无需针对特定模型训练即可对来自不同驱动模型的边界条件进行降尺度——扩散模型以非配对方式进行训练。与瑞士融合雨量计-雷达观测数据（CombiPrecip）相比，两种方法均优于欧洲中期天气预报中心的原始预报，但表现出依赖于环流背景的优势。WRF在一个多单体、非平稳事件中取得了最高的概率预报技巧。相反，扩散模型在两个事件的不同性能指标上更为一致，在一个更平稳的超单体事件中优于WRF。这些结果表明，显式动力建模可为特定降水事件在次季节提前期增加价值，而生成式降尺度在不同情境下更广泛地增加价值。
英文摘要
Coarse spatial resolution limits the ability of subseasonal prediction models to resolve extreme precipitation. Downscaling with either dynamical or deep generative models can overcome this issue, but the comparative performance of these models for extremes across different atmospheric regimes remains poorly understood. In this work, we evaluate the Weather Research and Forecasting (WRF) model against a diffusion-based generative model by downscaling two physically distinct, extreme precipitation events up to lead times of 3 weeks. For a fair comparison with WRF, which can downscale boundary conditions from different driving models without model-specific training, the diffusion model is trained in an unpaired fashion. Both approaches improve upon the raw European Centre for Medium-Range Weather Forecasts forecasts, in comparison to fused rain gauge-radar observations in Switzerland (CombiPrecip), but exhibit regime-dependent strengths. WRF achieves the highest probabilistic skill for a multicell, non-stationary event. Conversely, the diffusion model is more consistent across different performance metrics for the two events, outperforming WRF in a more stationary supercell event. These results demonstrate that explicit dynamical modeling can add value for specific precipitation events for subseasonal lead times, and that generative downscaling adds value more broadly in different situations.
Comments36 pages, 12 figures, 6 tables, submitted to JAMES
