人工智能能否实现动力降尺度？一种用于模拟公里级 COSMO5.0_CLM9 模拟的潜在扩散模型
Can AI be enabled to dynamical downscaling? A Latent Diffusion Model to mimic km-scale COSMO5.0\_CLM9 simulations
AI总结：
本文提出用潜在扩散模型结合残差方法将 ERA5 数据降尺度至 2 公里，以模拟 COSMO_CLM 动力降尺度结果，并证明其性能优于插值、UNET 和 GAN 基线。
摘要：
降尺度技术是深度学习在地球系统建模中最突出的应用之一。一个稳健的深度学习降尺度模型能够从粗尺度数值模式模拟中生成高分辨率场，从而节省区域/局地模式在时间和资源上的大量消耗。此外，生成式深度学习模型有潜力通过生成类似集合的情景池来提供不确定性信息，而这一任务对于传统数值模拟而言在计算上是难以承受的。在本研究中，我们应用一种潜在扩散模型（Latent Diffusion Model，LDM）将意大利上空的 ERA5 数据降尺度至 2 公里分辨率。高分辨率目标数据包括由 COSMO_CLM 动力降尺度得到的 2 米气温和 10 米水平风分量。我们的目标是证明：在给定相同输入数据的情况下，生成建模的最新进展使深度学习能够取得与数值动力模式相当的结果，同时保持细尺度特征和流场特征的真实性。我们从 ERA5 中选取一组预测因子作为 LDM 的输入，并在应用 LDM 时利用相对于参考 UNET 的残差方法。生成式 LDM 的性能与复杂度逐步增加的参考基线进行了比较：ERA5 的二次插值、UNET，以及基于同一参考 UNET 构建的生成对抗网络（GAN）。结果凸显了 LDM 架构和残差方法相对于这些基线所带来的改进。模型在年度测试数据集上进行评估，通过确定性指标、误差的空间分布以及频率和功率谱分布的重建来评价模型性能。
英文摘要：
Downscaling techniques are one of the most prominent applications of Deep Learning (DL) in Earth System Modeling. A robust DL downscaling model can generate high-resolution fields from coarse-scale numerical model simulations, saving the timely and resourceful applications of regional/local models. Additionally, generative DL models have the potential to provide uncertainty information, by generating ensemble-like scenario pools, a task that is computationally prohibitive for traditional numerical simulations. In this study, we apply a Latent Diffusion Model (LDM) to downscale ERA5 data over Italy up to a resolution of 2 km. The high-resolution target data consists of 2-m temperature and 10-m horizontal wind components from a dynamical downscaling performed with COSMO_CLM. Our goal is to demonstrate that recent advancements in generative modeling enable DL to deliver results comparable to those of numerical dynamical models, given the same input data, preserving the realism of fine-scale features and flow characteristics. A selection of predictors from ERA5 is used as input to the LDM, and a residual approach against a reference UNET is leveraged in applying the LDM. The performance of the generative LDM is compared with reference baselines of increasing complexity: quadratic interpolation of ERA5, a UNET, and a Generative Adversarial Network (GAN) built on the same reference UNET. Results highlight the improvements introduced by the LDM architecture and the residual approach over these baselines. The models are evaluated on a yearly test dataset, assessing the models' performance through deterministic metrics, spatial distribution of errors, and reconstruction of frequency and power spectra distributions.
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。
