1. A 3.3‐Million‐Year Record of Antarctic Iceberg Rafted Debris and Ice Sheet Evolution Quantified by Machine Learning.
- Author
-
Jasper, Claire E., Dyer, Blake, Reilly, Brendan T., Williams, Trevor, Hemming, Sidney, and Raymo, Maureen E.
- Subjects
CONVOLUTIONAL neural networks ,ICE calving ,MACHINE learning ,ICE sheets ,ANTARCTIC ice ,ICE shelves - Abstract
Over the last 3.3 million years, the Antarctic Ice Sheet (AIS) has undergone phases of ice sheet growth and decay, impacting sea level and climate globally. Presently, the largely marine‐terminating AIS loses mass primarily by iceberg calving and basal melt of ice shelves. Quantifying past rates and timing of AIS melt is vital to understanding future cryosphere and sea level changes. One proxy for past ice sheet instabilities is iceberg rafted debris (IRD) fluxes. However, traditional methods of IRD quantification are labor‐intensive. Here, we present a new method of identifying IRD grains in sediment core X‐ray images using a convolutional neural network machine learning algorithm. We present a 3.3‐million‐year record of AIS IRD melt events using sediment cores from International Ocean Discovery Program Sites U1536, U1537, and U1538 in the Southern Ocean's "Iceberg Alley." We identify two increases in the IRD fluxes throughout this period, at ∼1.8 and 0.43 Ma. We propose that after 1.8 Ma, the AIS expanded and transitioned from a primarily terrestrial‐terminating to a primarily marine‐terminating ice sheet. Therefore, after 1.8 Ma, glacial terminations and AIS iceberg discharge are associated with variations in global ice volume, presumably through the mechanism of sea level and, therefore, grounding line change. The second AIS regime change occurs during the Mid‐Brunhes Event (∼0.43 Ma). After this time, there are heightened and continuous IRD fluxes at each glacial termination, indicating increased AIS size and instability after this time. Plain Language Summary: Understanding the timing and magnitude of the melt and retreat of the Antarctic Ice Sheet (AIS) in the past can help us understand its sensitivity to future global climate change. One way to understand ice sheets in the past is to identify and quantify sand to pebble‐sized rock and mineral grains incorporated into ice sheets and exported to the ocean by icebergs. These so‐called iceberg rafted debris (IRD) grains get deposited on the seafloor when icebergs melt, and they are easily identifiable in deep‐sea sediment core X‐ray images. Here, we present an artificial intelligence image detection method to identify IRD in sediment cores collected in the heart of "Iceberg Alley," a region that today receives the majority of Antarctic icebergs. We find that over the past 3.3 million years, there have been two increases in the amount of IRD deposited. We propose that the first major increase in IRD just after 1.8 Ma is evidence of an expansion of the AIS to an ice sheet with marine margins and ice shelves. We further propose that after 430,000 years ago, the AIS undergoes additional expansion and sees consistent, rapid retreat at the end of each global glaciation. Key Points: We present a machine learning method to quantify iceberg rafted debris (IRD) in deep‐sea sediment coresWe quantify IRD at three International Ocean Discovery Program sites in the heart of Iceberg Alley over 3.3 million yearsWe propose that Antarctica transitions from a primarily terrestrial‐terminating to marine‐terminating ice sheet 1.8 million years ago [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF