[The impact of phylogenetic uncertainty on the metrics of community phylogenetic structure].

Research paper by Jia Yao JY Cai, Yuan Yuan YY Ding, Xin X Tong, Ren Ren RR Wang, Xiao Ling XL Gong, Xiao Yong XY Chen, Guo Chun GC Shen

Indexed on: 04 May '18Published on: 04 May '18Published in: Ying yong sheng tai xue bao = The journal of applied ecology / Zhongguo sheng tai xue xue hui, Zhongguo ke xue yuan Shenyang ying yong sheng tai yan jiu suo zhu ban


Phylogeny has been widely used to quantify community phylogenetic structure and to infer the underlying mechanism. Many studies, however, neglected phylogeny uncertainty and its potential impact on community phylogenetic structure. In this study, we explored the potential impact of phylogenetic uncertainty among 150 species in a 20 hm plot in Tiantong, Zhejiang. One consensus tree and 999 phylogenetic trees representing the phylogenetic uncertainty were estimated based on two cpDNA fragments (rbcL and matK). Combined with the species distribution data, community phylogenetic structure was quantified by two common indices (NRI and NTI) and their significances were tested by the independent swap null model. Our results showed that tree topology and node age showed a large uncertainty. The uncertainty was larger for young species and significantly increased with mean phylogenetic distance. Phylogenetic uncertainty increased the variation of both standardized NRI and NTI in each quadrat. These impacts were independent between both indices in either spatial pattern or the degree of impact. NRI was more sensitive than NTI to the uncertainty. At community scale, phylogenetic uncertainty also affected the variation of the mean standardized NRI and NTI of all quadrats, with mean standard deviation of 0.37 and 0.077, respectively. Such a result suggests that mean standardized NRI at community level was more vulnerable to the phylogenetic uncertainty, which is consistent with the result at the sample level. Our findings showed that phylogenetic uncertainty could add different variation into the NRI and NTI series indices and might increase biases in the quantification of community phylogenetic structure and its underlying ecological processes. Our results implied that non-random community phylogenetic structure was probably overestimated in the previous studies which ignored phylogenetic uncertainty.