A late summer temperature reconstruction based on tree-ring maximum latewood density since AD 1246 on the southeastern Tibetan Plateau

Abstract

The Tibetan Plateau (TP) is a critical component of the Earth's climate system, exhibiting high sensitivity to climate change. However, limited spatiotemporal coverage of meteorological observations hinders a comprehensive understanding of its long-term temperature variations. This study addresses this gap by presenting an August–September minimum temperature reconstruction for the southeastern TP since AD 1246, utilizing tree-ring maximum latewood density (MXD) as a proxy. Our reconstruction explains 42.3% of the variance observed in recent instrumental data. High consistency with published MXD chronologies from nearby regions validates the reconstruction's reliability. Furthermore, the comparison of our reconstruction with summer temperature field reconstructions suggests that those integrating diverse data sources—such as tree-ring width, density, and documentary evidence—through data assimilation techniques perform better in capturing pre-instrumental temperature variability in our study area. This highlights the importance of incorporating long tree-ring density data and other varied proxies into future data assimilation models for climate reconstructions. Additionally, large tropical volcanic eruptions exert a significant cooling effect, with reconstructed temperatures dropping by 0.33 °C and 0.37 °C in the eruption year and following year, respectively. Solar activity also appears to influence regional temperatures, with warm periods (AD 1720–1805 and AD, 1830–2008) generally coinciding with high solar activity and cold episodes (AD 1268–1344, AD 1429–1480, AD 1645–1715, and AD, 1806–1830) aligning with low solar activity. However, an exceptional cool period during strong solar activity (AD 1605–1638) suggests that other factors, such as volcanic eruptions, Atlantic Multidecadal Oscillation and Pacific Decadal Oscillation variability, might also be involved.