第七届青年地学论坛

青藏高原是地球上海拔最高的高原，孕育着现今中国境内最多的冰川。末次盛冰期（也称为末次冰盛期，LGM；约2万1千年前）时全球平均气温比现在低约5-8℃。青藏高原冰川在这一时期显著扩张，但其准确范围很难获得，至今仍未有明确定论。准确地重建LGM时期青藏高原冰川的分布，对于理解该地区冰川对气候变化的敏感性，以及气候系统对青藏高原冰川变化的响应，都具有重要科学价值。
用实地勘察的方法来约束青藏高原所有冰川在过去的变化几乎不可能，本研究尝试通过数值模拟的方法来达到这个目的。前人的研究通常不考虑冰川对气候的反馈，而是用一个给定的气候场驱动冰川模式来模拟LGM时期青藏高原的冰川。本研究自主开发了一个耦合器，将地球系统模式CESM 1.2.2与冰盖模式ISSM 4.17进行耦合。本研究使用现代的冰川分布对冰川模式中的关键参数进行了约束，然后利用约束好的参数首次实现了对LGM青藏高原冰川和气候的耦合模拟。
模拟结果显示，气候与冰川的相互作用对LGM青藏高原冰川规模的影响很大。在不考虑这种相互作用（非耦合模拟）时，模拟得到的冰川面积和体积分别为1.25×10⁶ km² 和1.04×10⁶ km³，在耦合模拟中则下降至1.15×10⁶ km²与0.80×10⁶ km³。耦合模拟中，青藏高原腹部西侧的冰川面积减少约50%，喜马拉雅山脉东侧减少约70%，青藏高原东部唐古拉山脉一带也有减少；而在祁连山脉、天山山脉和帕米尔高原，冰川的面积与厚度均有明显上升。对比非耦合试验，本文的LGM耦合模拟得到的冰川分布与重建资料的吻合程度更高，在天山区域的模拟结果也较前人研究有所改进。
在不同的气候态下，大气中的沙尘量以及沉降都会改变。沙尘不仅会通过其辐射效应影响气候，还可以通过改变雪的物理性质影响积雪的寿命。本研究进一步对比了无沙尘、现代沙尘和模拟的LGM沙尘三种情况下青藏高原的冰川和气候特征。模拟结果显示，沙尘含量和分布的变化对青藏高原冰川面积和厚度都有一定的改变。气候模式中对LGM时期青藏高原的准确模拟离不开对沙尘的分布和含量的有效约束。


Tibetan Plateau is the highest-elevation plateau on Earth, hosting the largest number of glaciers in China. Glaciers on Tibetan Plateau and its surrounding areas was much more extensive than present during Last Glacial Maximum (LGM), a period of time centered around 21ka when global mean temperature was 5-8 Kelvin lower than today. Due to uncertainties and scarcity in reconstruction data, constraints of the exact state of LGM glaciers was far beyond ideal. Accurately reconstructing glaciers on and around Tibetan Plateau remains vital towards understanding not only glaciers’ sensitivity against climate change, but also our climate system’s response against changes in glaciers.
Recognizing the difficulty in constraining all of Tibetan Plateau’s glaciers through field-study, we attempt to achieve this goal through means of numerical simulation. Previous studies commonly simulate LGM glaciers with a prescribed climatology without considering the changes through feedbacks from glaciers. We address this issue by building a coupler that couples a climate model/earth system model to an ice-sheet model, which in this study are CESM 1.2.2 and ISSM 4.17, respectively. This was the first time, to the best of our knowledge, that a LGM simulation was done for Tibetan Plateau with dynamically coupled ice-sheet and climate model. During our investigation, we constrained key parameters in the ice-sheet model by comparing present-day glacier simulations with observations of coverage. This set of parameters was then used in LGM simulations.
Our results show that the interaction between Tibetan Plateau glaciers and climate was significant and should not be ignored in high-accuracy simulations. Uncoupled runs that ignores such interaction yielded glacial coverage and volume of 1.25×10⁶ km² and 1.04×10⁶ km³, respectively, while coupled ice-sheet-climate model yielded 1.15×10⁶ km² and 0.80×10⁶ km³, respectively. Regional glacial features showed more dramatic changes once coupled. Glaciers on the mid-west Tibetan Plateau decreased in area by ~50%, while those north and east of Himalayas decreased by ~70%. In contrast, Qilian Mountains, Tianshan Mountains and Pamir Plateau saw pronounced increase in coverage and thickness. Compared with uncoupled simulations, our coupled results is in better agreement with reconstructions of LGM glaciers and provide significant improvement over previous studies in Tianshan region.
The amount of dust in the atmosphere and its deposition changes under various climate states. Such variations in turn, influence the climate through radiative effects and snow properties. This study further compares LGM glaciers and climate in Tibetan Plateau region under three different conditions, namely no dust, prescribed present-day dust and dynamically modelled LGM dust. Simulation results showed that changes in dust mass and distribution posed non-negligible influence on Tibetan Plateau glaciers and the climate.
 

冰川,青藏高原,末次盛冰期,数值模拟,气候模式,冰盖模式