High‐resolution stable isotopic signals of ground ice indicate freeze–thaw history in permafrost on the northeastern Qinghai–Tibet Plateau

Abstract

Understanding the mechanism of formation of ground ice and the freeze–thaw history of permafrost is essential when assessing the future of permafrost in a changing climate. High‐resolution ground ice records, integrating stable isotopes (δ18O, d‐excess, and δ13C), hydrochemistry (EC and pH) data, and cryostratigraphy at a depth of 4.8 m from two contrasting permafrost profiles (P‐1, P‐2) in the Source Area of the Yellow River (SAYR) on the northeastern Qinghai–Tibet Plateau (QTP), were investigated. The results suggested significant depth variations in the stable isotopes and hydrochemistry of the ground ice. The near‐surface ground ice (NSGI) and deep‐layer ground ice (DLGI) were characterized in terms of variations in stable isotopes and known modern active layer data. By synthesizing the measured δ18O and the modeled isotopic fractionation processes during freezing, we suggest that both the NSGI and DLGI in P‐1 were mainly formed by the segregation mechanism during permafrost aggradation. The NSGI in P‐2, however, exhibited quick freezing origins compared with the predominant ice segregation processes for the DLGI. By combining the evolution of various stable isotopes and hydrochemistry with 14C age data, four historical freeze–thaw stages were identified. Specifically, one thawing–refreezing stage (2.8–2.2 m), one freezing aggradation stage (2.2–1.6 m), and two permafrost aggradation–degradation cycle stages (4.8–2.8 m; 1.6–0.7 m) were differentiated, which emphasize the importance of climate‐induced freeze–thaw transitions and differing permafrost aggradation processes on ground ice formation and resultant isotope hydrochemical behaviors. This study is the first to use high‐resolution data in ground ice to interpret the freeze–thaw history of permafrost in the SAYR. These findings are important for further understanding of past permafrost evolution and projected future permafrost degradation trends on the QTP, and provide an alternative method to explore permafrost history.