全球AI天气模型中的蝴蝶效应确认：来自热带气旋预报的证据
Butterfly Effect Confirmed in Global AI Weather Models: Evidence from Tropical Cyclone Forecasting
- College of Meteorology and Oceanography, National University of Defense Technology（国防科技大学气象海洋学院）
- Laboratory of Atmospheric Environmental Monitoring and Early Warning for Low-Altitude Economy, National University of Defense Technology（国防科技大学低空经济大气环境监测与预警实验室）
- Hunan Institute of Advanced Technology（湖南先进技术研究院）
- College of Atmospheric Sciences, Lanzhou University（兰州大学大气科学学院）
- Institute of Atmospheric Physics, Chinese Academy of Sciences（中国科学院大气物理研究所）
- National Meteorological Centre, China Meteorological Administration（中国气象局国家气象中心）
- Department of Atmospheric and Oceanic Sciences and Institute of Atmospheric Sciences, Fudan University（复旦大学大气与海洋科学系及大气科学研究所）
- Key Laboratory of High Impact Weather (special), China Meteorological Administration（中国气象局极端天气重点实验室）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
本研究通过超强台风“卡努”的反例证明AI天气模型存在蝴蝶效应，微小扰动引发状态转换导致第7天1006公里路径差异，验证了AI捕捉大气混沌的能力并支撑集合预报。
AI中文摘要：
人工智能（AI）天气预报研究近期出现了一个悖论。一方面，有人声称AI天气模型无法模拟大气中的蝴蝶效应，这与AI模型有限的预测能力及AI集合预报的进展相矛盾。本研究通过反例证明，蝴蝶效应确实存在于AI天气预报中。对于超强台风“卡努”，AI预测受到双吸引子系统的约束。局限于两个区域的微小初始扰动会触发两个局部吸引子之间的状态转换，导致第7天预测风暴位置出现1006公里的差异。这一行为与数值天气预报模型一致，并在过去5年约12%的热带气旋中观测到。这些发现验证了AI捕捉大气混沌的能力，并为AI集合预报提供了物理基础。
英文摘要：
A paradox recently emerged in artificial intelligence (AI) weather prediction research. While some claim AI weather models cannot simulate atmospheric butterfly effect, this conflicts with AI models' limited predictability and advances in AI ensemble forecasting. This study demonstrates via counterexamples that the butterfly effect does exist in AI weather predictions. For Super Typhoon Khanun, AI predictions are constrained by a double-attractor system. Minor initial perturbations confined to two regions trigger state transitions between two local attractors, causing a 1006-km difference in the predicted storm position on Day 7. This behavior is consistent with numerical weather prediction models and observed in ~12% of tropical cyclones in the past 5 years. These findings verify AI's ability to capture atmospheric chaos and provide the physical basis for AI ensemble forecasting.
