Supplementary Tables for "A 3.3-Million-Year Record of Antarctic Iceberg Rafted Debris and Ice Sheet Evolution Quantified by Machine Learning"

Supplementary Tables for "A 3.3-Million-Year Record of Antarctic Iceberg Rafted Debris and Ice Sheet Evolution Quantified by Machine Learning" Table Captions: Table S1. Site U1537 Age Model Tie Points from Weber et al. (2022) and Reilly et al. (2021) Table S2. Site U1537 Age Model used in this study, applying both the age tie points from Weber et al. (2022) and Reilly et al. (2021) Table S3. Hole U1538A correlation to the Dove Basin Stack from Bailey et al. (2022), and the addition of the U1538 splice CCSF-A depth to the Dove Basin CCSF-A Table S4. Site U1538 splice table used in this study, note the continuation down Hole A after Core 14H Table S5. New top core section offsets for Site U1536 cores added to the Reilly et al. (2021) extended splice table Table S6. New top core section offsets for Site U1537 cores added to Reilly et al. (2021) extended splice table Table S7. Comparison of Convolutional Neural Network IRD counts to shipboard eye counts of IRD at Site U1536 Table S8. Site U1537 CNN IRD Counts per 50 cm bins Table S9. Site U1536 IRD Fluxes Per 5 kyr Quantified by a Convolutional Neural Network (0-3.3 Ma) Table S10. Site U1537 IRD Fluxes Per 5 kyr Quantified by a Convolutional Neural Network (0-3.3 Ma) Table S11. Site U1536 IRD Fluxes Per 1 kyr Quantified by a Convolutional Neural Network (0-1.2 Ma) Table S12. Site U1537 IRD Fluxes Per 1 kyr Quantified by a Convolutional Neural Network (0-1.2 Ma) Table S13. Site U1538 IRD Fluxes Per 1 kyr Quantified by a Convolutional Neural Network (0-1.2 Ma)《通过机器学习量化的南极冰山浮运碎屑与冰盖演化的330万年记录》（A 3.3-Million-Year Record of Antarctic Iceberg Rafted Debris and Ice Sheet Evolution Quantified by Machine Learning）补充表格 表格说明： 表S1. 取自Weber等人（2022）与Reilly等人（2021）的U1537站位年龄定标点 表S2. 本研究采用的U1537站位年龄模型，整合了Weber等人（2022）与Reilly等人（2021）的年龄定标点 表S3. U1538A钻孔与Bailey等人（2022）提出的多夫盆地堆叠层序对比，以及将U1538拼接层的CCSF-A深度纳入多夫盆地CCSF-A深度体系 表S4. 本研究使用的U1538站位岩心拼接表，请注意14H岩心之后延伸至A钻孔的岩心序列 表S5. 纳入Reilly等人（2021）扩展拼接表的U1536站位岩心新顶芯段偏移量 表S6. 纳入Reilly等人（2021）扩展拼接表的U1537站位岩心新顶芯段偏移量 表S7. 卷积神经网络（Convolutional Neural Network）统计的U1536站位冰筏碎屑（Ice Rafted Debris, IRD）数量与船载目视计数结果的对比 表S8. 按50厘米分箱统计的U1537站位卷积神经网络冰筏碎屑数量 表S9. 经卷积神经网络量化的U1536站位每5千年冰筏碎屑通量（0-3.3 Ma） 表S10. 经卷积神经网络量化的U1537站位每5千年冰筏碎屑通量（0-3.3 Ma） 表S11. 经卷积神经网络量化的U1536站位每1千年冰筏碎屑通量（0-1.2 Ma） 表S12. 经卷积神经网络量化的U1537站位每1千年冰筏碎屑通量（0-1.2 Ma） 表S13. 经卷积神经网络量化的U1538站位每1千年冰筏碎屑通量（0-1.2 Ma）