GAM+RAC models for species richness.

Statistics acronyms are BIC = Bayesian information criterion, ΔBIC = delta BIC (i.e., the difference in BIC score between the best model and the model being compared). Predictors’ acronyms are: Tmean = mean temperature, S = salinity, Chla = chlorophyll-a concentration, MLD = mixed layer depth, autocovariate = residuals autocovariate. The best model is highlighted in bold.

本数据集涉及的统计学术语缩写如下：BIC为贝叶斯信息准则（Bayesian information criterion），ΔBIC为ΔBIC（delta BIC，即最优模型与待比较模型的BIC得分之差）；预测变量的缩写对应关系为：Tmean代表平均温度（mean temperature），S代表盐度（salinity），Chla代表叶绿素a浓度（chlorophyll-a concentration），MLD代表混合层深度（mixed layer depth），autocovariate代表残差自协变量（residuals autocovariate）。最优模型以粗体标注。