Climate‐Driven Straightening of Western North Pacific Tropical Cyclone Trajectories

Abstract Tropical cyclone (TC) trajectory shifts remain a critical uncertainty in climate risk assessment. However, due to the absence of effective quantification methods for TC trajectories, our understanding of the overall characteristics of their trajectory shifts and the underlying impact mechanisms remains quite limited. Here, we introduce an important metric, sinuosity, as the parameter to describe the overall trajectory of TCs, and propose a new sinuosity‐based machine learning framework for their classification. The findings reveal a significant decrease in the sinuosity of TC trajectories over the Western North Pacific. This reduction is mainly due to straighter trajectories within the recurving cluster TCs, which follow conventional southeast‐to‐northeast recurving trajectories, with most dissipating in Northeast Asia. This “straightening” phenomenon of recurving TCs is jointly driven by the shortening of TC duration and the westward expansion of the Western Pacific Subtropical High (WPSH). As the trajectories of recurving TCs are becoming straighter, they are more likely to make landfall in Northeast Asia, significantly increasing the regional risks of TCs. These findings unravel a previously unobserved dimension of TC climate response, demonstrating how changes in subtropical circulations are reshaping regional hazard patterns with profound implications for disaster preparedness, mitigation, and adaptation strategies.