NGS-2020-10-02584-T_S3.xlsx

Data S3 from the manuscript "Global chemical weathering dominated by continental arcs since the mid-Paleozoic"<br>Gernon et al. 2021. Nature Geoscience.<br><br>Filename:NGS-2020-10-02584-T_S3.xlsx<br><br>S3 tabulates the empirical rank correlations (CEmp) and BN correlations (CBN) for all individual input parameters and lagged parameters (the predictor nodes) with (87Sr/86Sr)sw (the node of interest). These were calculated by generating a saturated Bayesian Network with all 254 nodes (including lags), and computing CEmp and CBN for all predictor nodes with (87Sr/86Sr)sw . Correlations have been calculated using both the original (‘best estimate’) input data set and also the simulated data set, to evaluate the effect of input uncertainty (see Methods and Extended Data Fig. 2 in manuscript).

本数据集为手稿《自中古生代以来以大陆弧为主导的全球化学风化作用》（Global chemical weathering dominated by continental arcs since the mid-Paleozoic）的补充数据S3，作者为Gernon等人，发表于2021年《自然·地球科学》（Nature Geoscience）。
文件名为NGS-2020-10-02584-T_S3.xlsx。
S3表格列示了所有单个输入参数与滞后参数（即预测变量节点）与目标节点(87Sr/86Sr)sw之间的经验秩相关系数（CEmp）与贝叶斯网络相关系数（CBN）。两类相关系数的计算过程为：构建包含全部254个节点（含滞后项）的饱和贝叶斯网络，随后针对所有预测变量节点与(87Sr/86Sr)sw分别计算CEmp与CBN。为评估输入不确定性的影响，研究分别使用原始（"最优估计"）输入数据集与模拟数据集完成相关系数计算，具体细节详见手稿中的方法部分与扩展数据图2。