发表机构
Weather Foundation（天气基金会）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究将GNSS导出的ZWD整合到Aurora天气基础模型中，扩展后的模型学习ZWD能力与预训练变量相当，微调时纳入ZWD可改善降水预报，增益随严重程度增加，能让降水功率谱更现实。
AI 中文摘要
全球导航卫星系统（GNSS）除了用于定位外，还可为气象科学服务，因为大气水汽会延迟其信号，这种延迟即天顶湿延迟（ZWD），是柱水汽的直接全天候测量指标。尽管ZWD已被同化到数值天气预报中数十年，但领先的机器学习天气模型（MLWM）尚未使用它，尽管它能解决已知的严重降水低估问题。本文首次将GNSS导出的ZWD整合到最先进的天气基础模型Aurora中。扩展后的Aurora学习ZWD的能力与其预训练变量相当。更重要的是，在对6小时累积降水进行微调时，纳入ZWD能系统地改善预报。增益随严重程度增加，在第99百分位数处公平威胁评分提高8.8%，同时降水功率谱在天气尺度和行星尺度上变得更现实。因此，直接的GNSS观测编码了MLWM可用于高影响降水的信息。
英文摘要
Global Navigation Satellite Systems (GNSS), best known for positioning, also serve weather science, as atmospheric water vapour delays their signals. This delay, the Zenith Wet Delay (ZWD), is a direct, all-weather measure of column moisture. Although assimilated into numerical weather prediction for decades, ZWD is not yet used by leading Machine Learning Weather Models (MLWM), despite addressing a known deficiency: the underestimation of severe precipitation. Here we present the first integration of GNSS-derived ZWD into Aurora, a state-of-the-art weather foundation model. Our extended Aurora learns ZWD with skill comparable to its pretrained variables. More importantly, including ZWD systematically improves forecasts when fine-tuning for 6-hour accumulated precipitation. Gains grow with severity, reaching an 8.8% increase in Equitable Threat Score at the 99th percentile, while the precipitation power spectrum becomes more realistic at synoptic and planetary scales. GNSS observations therefore encode information that MLWM can exploit for high-impact precipitation.
