Abstract: The formation of the Qinghai–Tibet Plateau has long been debated, despite the various proxies used to estimate its paleoelevation. Here, we introduce a novel method to calibrate paleoelevation by comparing the fossil and extant plant communities in the Qinghai–Tibet Plateau. Our estimation confirms that the uplift of the plateau was an episodic and heterochronous process. Specifically, the Lhasa Terrane was already elevated by 1 km before the initial India–Asia collision. During the first orogenic stage, the Qiangtang Terrane rose faster than the Lhasa Terrane, attaining 3 km in the late Eocene. In the second stage, the Lhasa Terrane underwent rapid uplift, reaching 3 km in the Oligocene. By the Middle Miocene, both the Qiangtang and Lhasa terranes had achieved an elevation of 4 km. The Himalaya rose by at least 2 km after the Pliocene. Our biological knowledge-based findings contradict the previous geological evidence-based reports, which posited that the plateau had reached an elevation of 4–5 km during the Eocene. We provide a new perspective on the plateau′s uplift history based on biological evidence, which has the potential to reconcile the confusion arising from contradictory proxies.

Key words: cluster analysis, community similarity, fossil assemblages, Himalaya, paleoelevation, Tibetan Plateau