中国北方5种栎属树木多度分布及其对未来气候变化的响应
张雪皎1, 高贤明2, 吉成均1, 康慕谊3,4, 王仁卿5, 岳明6, 张峰7, 唐志尧1,*(
)
)
Response of abundance distribution of five species of Quercus to climate change in northern China
ZHANG Xue-Jiao1, GAO Xian-Ming2, JI Cheng-Jun1, KANG Mu-Yi3,4, WANG Ren-Qing5, YUE Ming6, ZHANG Feng7, TANG Zhi-Yao1,*()
)
图2. 现有气候条件下随机森林模型(RF)、广义加性模型(GAM)和广义线性模型(GLM)对预测的栎属5种植物的潜在分布多度与实际观测值的关系及其准确度。A, 栓皮栎(Qva)。B, 麻栎(Qac)。C, 槲栎(Qal)。D, 锐齿槲栎(Qaa)。E, 蒙古栎(Qm)。F, 三个模型的准确度比较, 不同小写字母表示模型间差异显著(p < 0.05)。
Fig. 2. Relationship between observed and predicted species abundances for five species in Quercus with random forest model (RF), generalized additive model (GAM), generalized linear model (GLM) and comparison of the accuracy of five species based on general linear model and random forest. A, Q. variabilis (Qva). B, Q. acutissima (Qac). C, Q. aliena (Qal). D, Q. aliena var. acuteserrata (Qaa). E, Q. mongolica (Qm). F, Model fitness of the three models, differed lowercase letters indicated significant differences between the models (p < 0.05).
