Nonuniform Mantle Suture Beneath Northern Tibet Revealed by Helium Isotopes and Attention‐Based Machine Learning: Implications for Crustal Rheology and Growth of the Tibetan Plateau

Abstract

The tectonic evolution of northern Tibet is crucial for unraveling the uplift and growth mechanisms of the Qinghai–Tibet Plateau. However, inconsistent geophysical observations have led to long‐standing controversies over whether the Asian lithosphere beneath northern Tibet is undergoing underthrusting or southward subduction. In this study, we compiled 846 helium isotope ( R _C / R _A ) data points from across the Qinghai–Tibet Plateau and developed a two‐stage gradient boosted tree–probabilistic fusion network (2S‐GBTPFN) method that integrates geological, geophysical, geochemical, and hydrological parameters. The results show that attention‐based machine learning (ML) can reveal the spatial pattern of helium isotope ratios across the plateau and delineate a helium boundary in northern Tibet ( R _C / R _A  > 0.1 R _A ). This boundary is interpreted as the geochemical expression of a mantle suture and is consistent with independent geophysical observations. The newly identified helium boundary reveals a nonuniform rheological boundary across the Kunlun suture and indicates uneven, limited southward subduction of the Asian lithosphere beneath the northern Tibetan Plateau. When combined with evidence for subduction beneath southern Tibet, our results support a geodynamic framework involving dual‐slab subduction, multistage lithospheric delamination, and asthenospheric upwelling. These findings provide new geochemical evidence for the oblique subduction processes that have contributed to the uplift and growth of the Qinghai–Tibet Plateau.