Premium
Responsiveness of performance and morphological traits to experimental submergence predicts field distribution pattern of wetland plants
Author(s) -
Luo FangLi,
Huang Lin,
Lei Ting,
Xue Wei,
Li HongLi,
Yu FeiHai,
Cornelissen Johannes H. C.
Publication year - 2016
Publication title -
journal of vegetation science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.1
H-Index - 115
eISSN - 1654-1103
pISSN - 1100-9233
DOI - 10.1111/jvs.12352
Subject(s) - wetland , trait , biology , biomass (ecology) , ecology , specific leaf area , shoot , agronomy , environmental science , botany , photosynthesis , computer science , programming language
Question Plant trait mean values and trait responsiveness to different environmental regimes are both important determinants of plant field distribution, but the degree to which plant trait means vs trait responsiveness predict plant distribution has rarely been compared quantitatively. Because hydrological regime is a key determinant of wetland plant distribution, we hypothesized that both plant trait means and trait responsiveness to experimental submergence could predict plant adaptation to a wet or a dry part of hydrological gradients in wetlands. Location Beijing, China. Methods We measured mean values for 14 plant traits by growing 30 wetland species both on land (control) and partially submerged in a greenhouse, and calculated log response ratios (Ln RR s) of these traits to submergence. A distribution pattern index of wetland plants along the moisture gradient (from the zone furthest from the wetland waterline to that nearest to the waterline) was developed based on plant survey data in 3988 field plots in 24 wetlands in Beijing, China. Results Ln RR s of performance traits (shoot biomass, root biomass, plant height and total root length; R 2 = 0.249, P = 0.005) and one out of five morphological traits (i.e. shoot elongation capacity; R 2 = 0.194, P = 0.015) between partially submerged and control treatments could predict the distribution pattern of the 30 wetland plant species. In contrast, means of plant traits in either control or submergence could not predict species distribution. The trait Ln RR s, increasing from very negative to very positive, corresponded positively with the distribution, ranging from the zone furthest from the wetland waterline to that nearest to the waterline. Surprisingly, physiological trait Ln RR s that had been expected to underpin performance trait Ln RR s did not themselves predict the distribution pattern of these species. Analyses at the level of multivariate trait groups (based on PCA ) showed that species Ln RR s of the morphological trait group were positively correlated with Ln RR s of the performance trait group. Conclusions Our findings demonstrate that screening wetland plant species for performance and morphological trait Ln RR s under experimentally flooded conditions is an effective approach to understand and predict their distribution pattern along moisture gradients in the field.