Zhang Yufang, Moore John C, Zhao Liyun, Werder Mauro A, Gladstone Rupert, Wolovick Michael
State Key Laboratory of Earth Surface Processes and Resource Ecology, Faculty of Geographical Science, Beijing Normal University, 100875, China.
Arctic Centre, University of Lapland, 96101 Rovaniemi, Finland.
Sci Total Environ. 2024 Jun 1;927:172144. doi: 10.1016/j.scitotenv.2024.172144. Epub 2024 Apr 4.
Global climate warming leads to ever-increasing glacier mass loss. Pine Island Glacier in Antarctica is one of the largest contributors to global sea level rise (SLR). One of the biggest uncertainties in the assessment of glacier contribution to SLR at present are subglacial hydrology processes which are less well known than other ice dynamical processes. We use the Glacier Drainage System (GlaDS) model which couples both distributed and channelized components to simulate the basal hydrology of Pine Island Glacier with basal sliding and meltwater production taken from a full-Stokes Elmer/Ice model fitting observed surface velocities. We find ≈100 km long Rothlisberger channels up to 26 m in diameter extending up glacier from the grounding line along the main trunk of Pine Island Glacier delivering 51 m s of fresh water to the grounding line. Channelization occurs at high water pressure because of high basal melt rates (maximum of 1 m a) caused by high rates of shear heating in regions with fast ice flow (>1000 m a). We simulate a shallow "swamp" of 0.8 m water depth where flow transitions from a distributed system into the channels. We performed a set of 38 sensitivity experiments varying sheet and channel conductivity over 4 orders of magnitude. We find a threshold behavior in distributed sheet conductivity above which basal water pressures are unaffected by changing channel conductivities. Our findings suggest a strong need to better understand controls on basal water conductivity through the distributed system. This issue is critical to improve model-based predictive capability for the Pine Island Glacier and, more generally, the Antarctic Ice Sheet.
全球气候变暖导致冰川质量损失不断增加。南极洲的松岛冰川是全球海平面上升的最大贡献者之一。目前,在评估冰川对海平面上升的贡献时,最大的不确定性之一是冰下水文过程,它比其他冰动力学过程鲜为人知。我们使用冰川排水系统(GlaDS)模型,该模型结合了分布式和渠道化组件,通过从全斯托克斯Elmer/Ice模型中获取的基底滑动和融水产量来模拟松岛冰川的基底水文,该模型拟合了观测到的表面速度。我们发现,沿着松岛冰川主干从接地线向上延伸的直径达26米、长约100公里的罗特利斯伯格渠道,将51立方米每秒的淡水输送到接地线。由于快速冰流区域(>1000米/年)的高剪切加热速率导致高基底融化速率(最大1米/年),在高水压下会发生渠道化。我们模拟了一个水深0.8米的浅“沼泽”,在这里水流从分布式系统转变为渠道。我们进行了一组38次敏感性实验,将表层和渠道的电导率在4个数量级范围内变化。我们发现分布式表层电导率存在阈值行为,高于该阈值,基底水压不受渠道电导率变化的影响。我们的研究结果表明,迫切需要更好地了解通过分布式系统对基底水电导率的控制。这个问题对于提高基于模型的松岛冰川预测能力,更广泛地说,对于南极冰盖的预测能力至关重要。