| Global climate change and human activities such as grazing and other disruptive actions are bound to have impacts on biological diversity and especially the ecosystem, but what kinds of feed back and response the biodiversity and ecosystem functioning will make to the influence is a major issue that related to environment. With the deepening of research on the relationship between biodiversity and ecosystem functioning in recent years, ecologists focus on the plant traits and functional diversity (FD) transferred from the species diversity (SD). However, there were no general conclusions about the relationship among species diversity, functional diversity and primary productivity (PP) as well as how they were affected by the combined environment. A field manipulation experiment aimed at the problem that how the FD affects the relationship between SD and ecosystem functioning was conducted in alpine meadow at the Haibei Research Station of the Chinese Academy of Sciences from2007to2010. This experiment used a split-plot design with clipping treatment in the whole plot using three clipping levels (stubbled1cm,3cm and unclipped). Subplots were treated with fertilizer (urea at7.5g·m-2·a-1＋ammonium phosphate1.8g·m-2·a-1and unfertilized) and watering (watered20.1kg·m-2·a-1and unwatered).19major species were chosen, and6plant traits such as leaf area (A), above ground weight (AGW), plant height (H), leaf dry weight (LDW), specific leaf area (SLA), specific leaf weight (SLW) were selected respectively and their combined form was also uesd to quantify FD of the communities, which means a starting point to the research on relationship patterns between FD, SD and PP in the community along the gradients of grazing and soil resources. Aim at revealing the relationship mechanism between SD and FD as well as PP. Research results showed that:(1) The habitat resource and community characteristics have changed to some extent after clipping, fertilizing and watering treatments. The total phosphorus (Pt) and water content of soil (SW) decreased significantly after clipping (p<0.05). Fertilizing had the content of nitrate nitrogen (Nen) in soil significantly improved (p<0.05). The content of Pt had significantly improved after watering in quadrats fertilized (p<0.05). The total coverage of community decreased after clipping. Various SD indices do not have a remarkable change in the different clipping intensity. All SD indices reduce obviously after fertilized (p<0.05). The watering only has a weakly increase to the species richness index (R)(0.05<p<0.1).(2) Clipping and fertilizing treatments have different effects on distinct FD indices. Clipping had the FDLDw and the FD6traits remarkably increased (p<0.05), and FDAGW weakly increased (0.05<p<0.1), suggesting that clipping has increased the inter-species difference in LDW, AGW and the average of the six traits. Whereas, clipping treatment has no effect on FDH, FDA, FDSLA and FDSLw (p>0.05), i.e., there were no difference among the discrepancy of plant traits H, A, SLA, SLW in distinct clipping treatments. Fertilizing had the FDSLW and FDSLA remarkable declined (p<0.05), and FD&traits weakly declined (0.05<p<0.1). All FD indices do not have remarkable response to water treatment (p>0.05), but the interaction between fertilizing and watering treatment have a weak influence on the FDAGW-In other words, there may be some opposite effect on the difference of functional traits between clipping and fertilizing treatments, i.e., clipping increased the FD by expand the interspecies difference, but the change was opposite when fertilized.(3) There is an extremely remarkable correlation between FD6traits (calculated by the average of six traits) and other FD indices calculated by every single plant trait (p<0.01), which indicated that FD6traits would be a good indicator to the functional difference among species in the community. Although there is a reciprocal relation between the2traits SLW and SLA, the FDSLW and FDSLA actually have a significantly positive correlation (p<0.01), indicating that the difference on SLW among the species in community can co-vary with that on SLA, that is to say, both of the relative difference increase or reduce together.(4) The PP has a remarkable increase along the clipping intensity weak to strong (p <0.05). Also, it has obviously increased after fertilized. Fertilizing could significantly increased the compensatory growth ability (biomass clipped and regenerated) in quadrats clipped; there are no obvious changes in community unclipped regardless fertilized or not. But watering treatment has no effect on the PP. As to the interactions among different treatments, all of them have no effect except that interaction between clipping and fertilizing has a remarkable influence on the PP (p<0.01).(5) Clipping and fertilizing treatments have remarkable effects on the plant traits. Plant traits such as AGW, H, LDW and A declined remarkably (p<0.01) along the clipping gradients weak to strong. H, A, LDW and SLA had obviously increased (p<0.01) after fertilized, while the SLW declined remarkably. These results showed that clipping and fertilizing treatments had opposite effect on most plant traits. Relevance analysis of the different plant traits showed that:there was a significantly positive correlation (p<0.01) between AGW, A, LDW and H, the same relationship exists in A as well as LDW, AGW. Two facets are relevant to this case. On the one hand, because the LDW of these species we choose such as Saussurea pulchra, Gentiana macrophylla, Kobresia humilis and so on occupy the AGW&major parts. And on the other hand, plant traits such A, H, LDW and so on response uniformly to the treatments clipping and fertilizing.(6) In integral gradients of clipping (including NH, H3and H1) and fertilizing (NF and F) treatments (n=54), there was a significantly positive correlation between SD and FD6traits, FDAGW, FDH respectively (p<0.05); the rest FD indices has no correlation with SD (p>0.05). These outcomes demonstrate that either increase or decrease in SD would cause the FD6traits (especially the FDAGW and FDH) increase or decrease in grassland community with a high degree of heterogeneity in utilization intensity and content of soil nutrient, i.e., the competitive feedback effects as a result of the increase in SD is first expressed in H and AGW, so PP has no correlation with SD (p>0.05). The FD calculated by the6single traits has no correlation with PP except that there is a weakly negative correlation between FDSLA and PP (0.05<p<0.1).(7) In combination gradients of clipping (including NH, H3and H1) and fertilizing (F) treatments (n=27), SD has a significantly positive correlation with FD6traits (p<0.05), and it is not related with the other FD indices (p>0.05); Besides, SD and FD6traits have weakly positive correlation with the PP (0.05<p<0.1). SD, FD and PP are not related (p>0.05) in gradients of clipping (including NH, H1and H3) and fertilizing (NF) treatments (n=27). Through comparison of two treatment combinations (NH, H3, H1and NF) and (NH, H3, H1and F), we could found that the relationship among the three shows positive correlation in high fertility, while none of them are related to each other in the low soil fertility. The FD has no correlation with SD and PP in gradients of clipping (NH) and fertilizing (NF and F) treatments (n=18)(p>0.05), but there was a weakly negative correlation (0.05<p<0.1) between SD and PP. There are significantly positive correlation (p<0.05) between SD and FDAGW, FDSLW in gradients of clipping (H3) and fertilizing (NF and F) treatments (n=18), but the SD is not related with the PP (p>0.05), meanwhile, there is a weakly negative correlation (0.05<p<0.1) between FDSLA, FD6traits and PP. SD, FD and PP are not related (p>0.05) in gradients of clipping (HI) and fertilizing (NF and F) treatments (n=18). These result confirmed that the complex and changeable relationship between FD, SD and PP showed dependence in disturbance, availability of soil resources and plant traits in natural community with a high degree of heterogeneity in utilization intensity and content of soil nutrient. The existing theoretical hypotheses about the relationship among SD, FD and PP such as null hypothesis, rivet hypothesis and so on may not have universal application. For example, the facts that SD, FD and PP are not related (p>0.05) in gradients of clipping (HI) and fertilizing (NF and F) treatments or in gradients of clipping (including NH, HI and H3) and fertilizing (NF) treatments may be means that the phenomena predicted in null hypothesis is a special case happened in condition with a poor resources or strong selection pressure.(8) RDA (Redundancy analysis, RDA), one of the Canonical correspondence analysis (Canonical correspondence analysis, CCA) methods was used to analyze the relationship among the plant traits, SD and FD as well as PP in all communities, which were viewed as the preselected variables, and the7soil nutrients viewed as explanatory variables. The results indicated that the greatest impact on preselected variables such as plant traits and so on were available phosphorus (Pe), followed by total nitrogen (Nt). There is a co-linear relationship among Pe and Nt and the rest soil factors, belonging to the redundancy of the explanatory variables. Sorted first-and second-order-axis by the Pe and the Nt together, explaining the28.1%of the total variation, which means that available phosphorus and nitrogen is an important environmental factors affecting plant community characteristics. What we can infer from the RDA chart is that influence of the soil nutrients on the pattern of biodiversity and PP is mainly achieved through shaping the community traits of plants.In summary, due to the effects of clipping or grazing intensity, the habitat resource as well as plant traits in alpine meadow ecosystem, the relationship between SD and FD displays a variability to a certain extent, which means that their relationships are not only traits-dependent, but also has something to do with the resource and interference in environment. Although our experimental results show that there is a positive correlation between SD and FD, this relationship is affected by the traits, disturbances and resource, and changes in PP in ecosystem associated with this phenomena. But there were different kinds of relationship forms among SD, FD and PP, which means that as a component of biodiversity, FD based on plant traits is different from the SD based on community level. There may be not an absolute substitution but a complementary relationship between them. It is too narrow to use only SD to interpret the biological diversity effects on ecosystem functioning in the mechanism. To reveal the effect of biodiversity on ecosystem functioning, we cannot ignore the study of plant functional traits. |
PDF Full Text Request