Ancient Tibetan Ice Cap Holds Climate Secrets Older Than 100,000 Years

Aerial view of a snowy mountainous landscape with a drilling rig and several ice structures in the foreground, under a clear blue sky.
AI generated by Grok

Researchers have confirmed that ancient ice in the high- altitude Guliya Ice Cap (northwestern Tibet) preserves climate records from the last ice age and beyond, making these the oldest ice cores recovered from the region.

The study, led by Lonnie Thompson (professor of earth sciences and senior research scientist at Ohio State’s Byrd Polar and Climate Research Center) and published in Science Advances, compared ice cores drilled from the Guliya Plateau in 1992 and 2015. Both showed matching oxygen isotope patterns, confirming that the environmental signals are consistent and reproducible over decades.

To date the ice, the team used radioactive isotopes beryllium-10 and chlorine-36. These marked the Laschamp Geomagnetic Excursion (a major shift in Earth’s magnetic field more than 41,000 years ago). Combined with oxygen- isotope records from nearby Tibetan cave deposits, the data indicate the Guliya record extends well over 128,000 years (and more than 100,000 years overall).

This is significant because, outside the polar regions, it is the only mountaintop ice core that reaches so far back. For comparison, cores from Peru’s Nevado Huascarán go back only a bit over 30,000 years. Earlier work had suggested Guliya’s record ended in the Mid- Holocene (starting ~12,000 years ago), but the new evidence shows some ice is much older and that portions of the ice cap survived the last ice age.

Accurate timescales like this help answer broader questions about when major glaciations began and ended and how natural ice loss processes worked. The team plans further analyses of the cores for additional climate and environmental indicators. The research was supported by the U.S. National Science Foundation; Ohio State co- authors include Mary Davis and Ellen Mosley- Thompson, with international collaborators from Switzerland and China.

Sources: Ohio State News (Tatyana Woodall) and the Byrd Polar and Climate Research Center announcement (both dated around September 30- October 1, 2026).

Map showing the Guliya ice cap and surrounding geographical features, with arrows indicating monsoon wind patterns and detailed inset of the Chongge ice cap.
Geographic setting of the GIC. Credit: Science Advances (2026). DOI: 10.1126/sciadv.aeh2912

Additional evidence supports a timescale exceeding 100,000 years for the Guliya ice cap, Tibetan Plateau

The oxygen isotope (δ¹⁸O_ice) record from a ~310 m core drilled on the Guliya ice cap (GIC) in 2015 (GP2015) is reproducible with a nearby ~309 m core from 1992. The timescale for a composite of the GIC cores was constructed using ¹⁰Be and ³⁶Cl to identify the ~41 ka Laschamp geomagnetic excursion, plus a match to a δ¹⁸O_atm- dated core from the Guliya Summit. Comparison of the GIC δ¹⁸O_ice with an Asian Monsoon speleothem δ¹⁸O record supports a timescale back to 128 ka, the oldest known so far from Tibet. The δ¹⁸O_ice record shows orbital- scale variations in phase with June insolation and polar ice- core stable-isotope records through the last glacial cycle (with a phase reversal after the deglaciation that requires further study). Radiocarbon- dated ice from the glacier margin confirms glacial- stage ice in the GIC. The ³⁶Cl/¹⁰Be ratios in GP2015 confirm that GIC ice is older than 100 ka.

Key background and context

The Guliya ice cap (western Kunlun Mountains, northwestern Tibetan Plateau) is a cold- based (polar- type) glacier. Cores to bedrock were recovered in 1992 (GP1992, 308.7 m) and 2015 (GP2015, 309.7 m) on the plateau (~6200 m asl), plus thinner summit cores in 2015. Earlier work suggested ages >110 ka (and possibly much older near the bed based on ³⁶Cl).

Recent studies of nearby cores (Chongce ice cap and a 2021 Guliya Plateau core drilled close to the earlier sites) produced much younger (Holocene or Late Holocene) ages, challenging the long timescale. This paper extends and independently supports the longer chronology for the 1992 and 2015 plateau cores.

Main results and methods

  • Reproducibility: High-resolution δ¹⁸O_ice profiles from GP1992 and GP2015 match closely by depth and can be stratigraphically aligned, allowing transfer of the timescale.
  • Timescale construction (to 128 ka):
    • ¹⁰Be and ³⁶Cl peaks at ~187 m mark the Laschamp excursion (~41 ka).
    • Matching of the upper part to the δ¹⁸O_atm- and ¹⁴C- dated Guliya Summit composite.
    • Further extension by matching the δ¹⁸O_ice record (pre- 41 ka) to the continuous, ²³⁰Th-dated Southeast Asian Monsoon speleothem δ¹⁸O composite from SE China caves.
  • Supporting evidence of pre- Holocene ice: Two ¹⁴C- WIOC samples from ice ~50 cm above bedrock at the southeastern margin of the GIC yielded mean calibrated ages of ~24.1 ka and ~24.8 ka (last glacial stage).
  • Cosmogenic isotopes: Combined ³⁶Cl (from both cores) and ¹⁰Be (GP2015) data, placed on the new timescale, show the Laschamp peak and a possible Mono Lake excursion (~33 ka). Nuclear-weapons- testing ³⁶Cl is confined to the near- surface. The ³⁶Cl/¹⁰Be ratio decreases with depth in a manner consistent with radioactive decay and ages >100 ka (effective half-life ~384 kyr under standard assumptions). Establishing a precise timescale beyond 128 ka remains challenging due to possible post- depositional mobility or production- rate issues near the bed.

Climatic implications (discussion highlights)

The composite GP1992/2015 δ¹⁸O_ice and deuterium-excess records track orbital (insolation) and millennial- scale features seen in polar ice cores and (with the expected antiphase relationship) Asian monsoon speleothems through much of the last glacial cycle. A phase shift relative to the speleothem records occurs around the deglaciation (~18- 19 ka), possibly linked to changing atmospheric circulation (westerlies vs. Asian monsoon) and moisture sources as Northern Hemisphere ice sheets retreated. Cooler conditions around the Laschamp excursion are consistent with modeled climate responses to geomagnetic field weakening.

The paper addresses prior challenges to the long timescale and concludes that the Guliya plateau cores preserve a continuous climate archive spanning the last glacial cycle and beyond 100 ka, unique among non- polar, high- elevation ice cores. Future work is suggested to refine ages deeper than 128 ka and to investigate the post- deglaciation isotopic phase change.


Discover more from Climate- Science.press

Subscribe to get the latest posts sent to your email.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.