| As a largest grassland type in China,alpine meadow plays an important role in soil and water conservation,biodiversity conservation,maintaining balance of ecosystem and animal husbandry development.However,the community structure of alpine meadow ecosystem is relative simple,and is vulnerable to disturbance.In recent years,as a result of global climate changing,excessive exploitation and utilization of human beings,the alpine meadow is generally degraded.It is generally believed that the species diversity can not only increase the productivity of ecosystem,but also improve the stability of ecosystem.The long-term overgrazing and blind fertilizing often lead to the loss of species diversity,which may reduce the community stability.The impact of environmental changed and loss of biodiversity caused by human disturbance onecosystem stability is a major concern around the world.But up to now,the relationship between diversity and stability,as well as the mechanism of control stability are mostly studied in the artificial community.In order to study the effects of human activities on the stability of alpine meadow and to explore the relationship between the loss of species diversity and the community stability,it is necessary by using methods of simulated grazing and nutrient addition to study the effects of human disturbance on the stability of community in alpine meadow.This research was carried out in Kobresia humilis meadow of the Haibei Alpine Meadow Ecosystem Research Station from 2007 to 2016.The experiment used a split-plot design with clipping treatment in the whole plot(three clipping levels,i.e.heavy clipping(HC):stubbled 1 cm,moderate clipping(MC):stubbled 3 cm and non-clipping(NC))and fertilizer(two fertilizing levels,i.e.fertilized and not fertilized)treatment assigned to subplots.We studied firstly the effects of both clipping and fertilizing disturbance on the temporal stability of ANPP of the meadow,and secondly explained the reason of changes in temporal stability of different communities’ ANPP by testing the potential mechanisms of temporal stability in order to further explore the relationship between species diversity and community temporal stability.Based on the available literatures,we proposed four speculations remain to be test:Ⅰ:According to the intermediate disturbance hypothesis,moderate clipping will significantly increase the temporal stability of the meadow but heavy clipping will decrease the temporal stability of the community compared with the unclipped community.Although fertilizing can reduce the species diversity of community,it can also increase the community dominance;therefore,the temporal stability of the community results from selection effect that driving by fertilization should not be lower than that of non-fertilizing community.Ⅱ:There are differences in species diversity and community dominance among six communities exposed different treatments,so various temporal stability mechanisms may coexist in a certain community,but the main temporal stability mechanism of different communities should be different.Ⅲ:Whether the selective effect can improve the temporal stability of a community depends mainly on the stability of dominant species of the community,and ultimately should be independent of species richness or negative correlated with species richness.Regardless of fertilizing or not,therefore,there should be obvious selection effect in unclipped communities,and significant complementary effect in clipping communities(including asynchrony effect,portfolio effect,facilitation interactions and weak interactions).Ⅳ:Because of clipping decrease interspecific competition,fertilizing increase interspecific competition,therefore,with the increasing of clipping intensity,fluctuation of species abundance should become more randomization,the portfolio effect will play a major role on the maintenance of temporal stability in clipped communities,and the selection effect or asynchronous effect will play an important role in fertilizing communities.In order to test the above speculations,we analyzed these observational data by employing the statistical program,such as RMANOVA,univariate ANOVA,single factor ANOVA,Regression Analysis and simplified Path Analysis,and get the following results:(1)Compared with no clipping and no fertilizing,clipping and fertilizing significantly increased the temporal stability of alpine meadow community,which partially confirmed speculation I.Both heavy clipping and fertilizing not only did not decrease temporal stability of community but also increase it.(2)The asynchronous effect,the portfolio effect and the facilitation interactions coexisted in six different communities;moreover,the selection effect was also found in fertilizing communities.This result showed the coexistence of different mechanisms in a community,and partially confirmed speculation Ⅱ.(3)Simplified Path analysis showed that there was no significant difference in the effect of these mechanisms on temporal stability in both no clipping without fertilizing(NC-NF)and moderate clipping without fertilizing(MC-NF)community;the portfolio effect was the main mechanism in heavy clipping without fertilizing community(HC-NF);the selection effect played an important role in both no clipping with fertilizing(NC-F)and heavy clipping with fertilizing(HC-F)community,and it had the more significant relative contribution to the temporal stability of the community.Despite the portfolio effect and the facilitation interactions played a major role in moderate clipping with fertilizing community(MC-F),the facilitation interactions had a larger contribution to the temporal stability of the community.Because the selection effect was the main mechanism of maintaining stability of heavy clipping community(HC-F),therefore,the results of this study partially confirmed speculation Ⅲ.(4)In non-fertilizing communities,the main mechanism of temporal stability was portfolio effect with increasing clipping intensity.Under fertilizing condition,the main mechanism was selection effect for both no clipping community(NC-F)and heavy clipping community(HC-F),whereas,the main mechanism were portfolio effect and facilitation interactions.The results partially confirmed speculation Ⅳ.Our research suggested that the main mechanism controlling temporal stability ofthe alpine meadow was driven by disturbing factors and disturbing intensity,namely,themain mechanism maintaining community stability was disturbance-dependent;under the condition of non-fertilizing,clipping was main factor that driving portfolio effect became a main mechanism with increasing clipping intensity.However,the main mechanism may be selection effect,portfolio effect and/or facilitation interactions as a result of altering clipping intensity.Thus we concluded that the portfolio effect was mainly driven by clipping disturbance,while fertilizing can change the driving effect resulting from clipping disturbance on the mechanisms,weaken the portfolio effect and touch off portfolio effect and/or facilitation interactions becoming a main temporal stability maintenance mechanism.The results of the study confirmed simultaneously the stabilizing effect of species diversity on ecosystem functioning because clipping treatment not only improved the species diversity,but also increased the community stability mainly through the portfolio effect.However,this stabilizing effect may be achieved only in the community where the species abundance fluctuated randomly resulting from clipping plus non-fertilizing,whereas,the species diversity decreased significantly after fertilizing,the temporal stability of fertilizing community was increased mainly by the selection effect.This result can explain the phenomena that there may be a negative correlation between the species diversity and the ANPP stability of community.Therefore,disturbance is an external driving factors impacting correlation between the species diversity and the ANPP stability of community,clipping and fertilizing disturbance can induce directly positive and negative correlation,respectively,thus ANPP stability of community will be maintained mainly through portfolio effect and selection effect,respectively. |