Study On The Integration And Adjustment Of Plant Characters In Alpine Meadow Community By Cutting And Fertilization

Understanding how communities assembly is a key challenge in community ecology. Conflicting hypotheses suggest that plant traits within communities should show divergence to reflect strategies to reduce competition and convergence under strong selection of environmental filters. There is also the hypotheses that community assemble under neutral process, and variation of species abundance is random rather than deterministic, while the number of species in the community is controlled by species extinction and immigration (emigration) and speciation of new species, and the lack of any pattern of convergence or divergence in any particular trait means that this trait is not of importance in structuring the community. However, there is a view that plant traits related to productivity show convergence, while those related to disturbance show divergence. Many studies have demonstrated links between certain traits and certain ecosystem process, and understanding whether traits show convergence or divergence may be important. Stress and disturbance have long been recognized as two major environmental gradients affecting the fitness of plants and therefore the structure of plant communities. However, studies interested in trait dispersion pattern are scare, and most of them do not relate the traits to explicitly measured environmental gradients.A field manipulation experiment was conducted in alpine meadow from2007to2012. In the experiment, a split-plot design was adopted. Three clipping levels (stubbled1cm,3cm and unclipped) of clipping treatment were used on the whole plot and subplots were treated with or without fertilizing. Data on functional diversity (FDvar) of10traits (growth form, plant inclination, reproductive modes, life cycle, nitrogen fixation, mature plant biomass, canopy height, leaf area, specific leaf area and leaf dry weight) from108quadrats (under different treatment combinations of mowing and fertilizing) were employed to test trait convergence/divergence using randomizations tests. The null model was that calculated values of FDvar of certain traits would higher or lower than values of FDvar calculated from randomly assembled communities. A trait was considered to show convergence if the average rank of its FDvar fell in the lowest2.5%of values produced by the randomization, and for divergence in the highest2.5%. Rank correlation and regression analysis based on Akaike information criterion were employed to test the relationship between environmental factors and variation of plant traits. The results showed that:1) A two-tailed test was employed and result showed that some traits showed a pattern of convergence/divergence at a certain degree, but the probability of FDvar of all the traits tested greater or lower than expected FDvar were greater than0.05. It demonstrates that no traits showed consistent significant convergence or divergence pattern in integral gradients of fertilizing and mowing treatments.2) Three basic patterns were seen in the relationship of the community weighted mean of each trait,,FDVar and the effect size of each trait. The first was exhibited by life cycle, in that it showed no relationship to any of the environmental variables tested. The second type of pattern was shown by mature plant biomass. It was responsive to the environment and had relatively invariant FDvar. There was a rise in ES with increased productivity. CWM was negative correlated with biomass loss, but positive correlated with vegetation height. The third pattern was shown by the remanent8traits. They all showed clear correlation with the environment; ratio of plants which carried out sexual reproduction and free living nitrogen fixation would be higher as disturbance increased, plant were inclined to extended to lower space, variation of specific leaf area, leaf area and canopy height were significantly decreased, growth of bunched and closely bunched plants were significantly restrained. The available space would be more as productivity increased, variation of leaf dry weight would be higher. Among which, the variation of ES of4traits showed a similar trend with FDvar, ES of growth form, canopy height, leaf area and leaf dry weight showed no significant patterns.3) Both mature plant biomass and leaf dry weight produced weakly supported relationships between the FDvar and the productivity. Overall, productivity cannot well predict the variation of plant traits.4) Conversely, the evidence supported disturbance as the major driver of FDvar for reproduction modes, canopy height and specific leaf area. Furthermore, FDvar of plant inclination and nitrogen fixation was affected by integral treatment of disturbance and productivity. It shows that FDvar was best predicted by solely disturbance measures and a combination of disturbance and productivity measures.5) The commonest pattern across the7traits response for disturbance was for a fall in.FDvar with increased disturbance. It indicated a general pattern of trait convergence, this may not be true for all traits, or the variance in a trait may impact on the ecosystem process. Weak convergence may still be enough to make a system predictable.It shows that the assembly of alpine meadow(Kobresia humilis) plant community was a neutral process, impact of resource intimidation on trait assembly was not significant, and mowing showed a significantly effect and would lead to trait convergence. The result may be help on management of alpine meadow plant community under different level of productivity and disturbance and achieve the aid of sustainable development. Also, it benefits for management and restoration of degraded grassland ecosystem.