arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2602.01436physics.ao-ph
收藏

MSWEP V3:基于机器学习的全球降水估计,0.1度每小时分辨率(1979年至今)

MSWEP V3: Machine Learning-Powered Global Precipitation Estimates at 0.1$^\circ$ Hourly Resolution (1979-Present)

King Abdullah University of Science and Technology (KAUST)(阿卜杜拉国王科技大学);Department of Civil Engineering, College of Engineering, King Saud University(沙特国王大学工程学院土木工程系);Hydro-Climate Extremes Lab (H-CEL), Ghent University(根特大学水文气候极端实验室);ECMWF(欧洲中期天气预报中心);Fenner School of Environment & Society, Australian National University(澳大利亚国立大学芬纳环境与社会学院);CSIRO Environment(联邦科学与工业研究组织环境部门);Centre of Studies in Resources Engineering, IIT Bombay(印度理工学院孟买分校资源工程研究中心);Centre for Climate Studies, IIT Bombay(印度理工学院孟买分校气候研究中心);School of Engineering, Newcastle University(纽卡斯尔大学工程学院);Center for Western Weather and Water Extremes, Scripps Institution of Oceanography, University of California, San Diego(加州大学圣地亚哥分校斯克里普斯海洋研究所西部天气和水极端现象中心);School of Geography and the Environment, University of Oxford(牛津大学地理与环境学院);School of Geography and Environmental Science, University of Southampton(南安普顿大学地理与环境科学学院)
Xuetong Wang, Raied S. Alharbi, Oscar M. Baez-Villanueva, Diego G. Miralles, Jun Ma, Shiqin Xu, Matthew F. McCabe, Florian Pappenberger, Albert I. J. M. van Dij… 展开作者

Xuetong Wang, Raied S. Alharbi, Oscar M. Baez-Villanueva, Diego G. Miralles, Jun Ma, Shiqin Xu, Matthew F. McCabe, Florian Pappenberger, Albert I. J. M. van Dijk, Tim R. McVicar, Lanka Karthikeyan, Hayley J. Fowler, Ming Pan, Solomon H. Gebrechorkos, Hylke E. Beck

AI总结:

MSWEP V3通过机器学习提供全球高分辨率降水数据,提升气候监测和水管理的可靠性。

摘要:

我们介绍了MSWEP产品第三版(V3)——首个完全全球、历史性的基于机器学习的降水(P)数据集,旨在满足日益增长的对及时准确降水估计的需求,以应对日益加剧的气候变化挑战。MSWEP V3提供从1979年至今每小时0.1度分辨率的数据,持续更新,延迟约为两小时。开发过程分为两个阶段。首先,使用机器学习模型堆栈生成基线降水场,这些模型整合了卫星和(再)分析的降水和气温产品,以及静态变量。模型使用来自全球15959个降水站的每小时和每日观测数据进行训练。第二,这些基线降水场通过来自全球57666和86000个站点的每日和月度降水站观测数据进行修正。为了评估MSWEP V3的基线性能,我们使用一个独立的15958个降水站观测数据集(排除在第一训练阶段之外)评估了19个(准)全球格点降水产品——包括未修正和基于降水站的产品。MSWEP V3基线实现了每日Kling-Gupta效率(KGE)的中位数为0.69,优于所有评估的产品。其他未修正的产品每日KGE中位数分别为0.61(ERA5)、0.46(IMERG-L V7)、0.38(GSMaP V8)和0.31(CHIRP)。通过留一法交叉验证,每日降水站修正被发现提高了每日相关性的中位数,但受限于基线性能已经很强。我们预计MSWEP V3——可通过www.gloh2o.org/mswep访问——将能够更可靠地监测、预测和管理气候变化中的水相关风险。

英文摘要:

We introduce Version 3 (V3) of the gridded near real-time Multi-Source Weighted-Ensemble Precipitation (MSWEP) product -- the first fully global, historical machine learning powered precipitation (P) dataset, developed to meet the growing demand for timely and accurate P estimates amid escalating climate challenges. MSWEP V3 provides hourly data at 0.1$^\circ$ resolution from 1979 to the present, continuously updated with a latency of approximately two hours. Development follows a two-stage process. First, baseline P fields are generated using machine learning model stacks that integrate satellite- and (re)analysis-based P and air-temperature products, along with static variables. The models are trained using hourly and daily observations from 15,959 P gauges worldwide. Second, these baseline P fields are corrected using daily and monthly gauge observations from 57,666 and 86,000 stations globally. To assess MSWEP V3's baseline performance, we evaluated 19 (quasi-) global gridded P products -- including both uncorrected and gauge-based products -- using observations from an independent set of 15,958 gauges excluded from the first training stage. The MSWEP V3 baseline achieved a median daily Kling-Gupta Efficiency (KGE) of 0.69, outperforming all evaluated products. Other uncorrected products achieved median daily KGE values of 0.61 (ERA5), 0.46 (IMERG-L V7), 0.38 (GSMaP V8), and 0.31 (CHIRP). Using leave-one-out cross-validation, the daily gauge correction was found to improve the median daily correlation by 0.09, constrained by the already strong baseline performance. We anticipate that MSWEP V3 -- accessible at www.gloh2o.org/mswep -- will enable more reliable monitoring, forecasting, and management of water-related risks in a variable and changing climate.

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。

↑