牦牛放牧率对小嵩草高寒草甸暖季草场植物群落组成和植物多样性的影响

两年的牦牛放牧试验结果表明:在暖季草场,随着放牧率的提高,禾草和莎草类功能群的盖度、生物量及其组成与高度减小,它们与放牧率呈显著的负相关 P<0.05 可食杂草和毒杂草类功能群的盖度、生物量及其组成与高度增加,它们与放牧率呈显著的正相关 P<0.05 各功能群的盖度与生物量之间是显著的正相关关系 P<0.05 .对照草地由于没有牦牛的采食,群落由少数优势种植物所统治,群落结构趋于简单,物种组成贫乏,物种多样性和均匀度指数最小;轻度放牧牦牛选择采食对植物群落的影响较小,群落的物种丰富度指数、均匀度指数和多样性指数均不高;中度放牧提高了资源的利用效率,增加了群落结构的复杂性,草地的物种丰富度指数、均匀度指数和多样性指数均最高,该结果支持 中度干扰理论 ;重度放牧由于牦牛采食过于频繁,改变了植物的竞争能力,导致植物种的均匀度下降,多样性减少.不同放牧率草地群落的物种数 S 、丰富度指数 Ma 、多样性指数 Shan-non-Wiener指数H′和Simpson指数D 、均匀度指数 Pielow指数J′ 的排序为:对照<轻度放牧<重度放牧<中度放牧,优势度指数 Berger-Parke指数 的变化趋势则与之相反.     

封育对高寒草甸植物群落构成及生态位特征的影响

封育是天然草地管理的一种有效措施,利于草地生产力提高和退化草地恢复。该研究通过对青藏高原高寒草甸放牧和封育草地物种多样性、植被构成、植物种生态位特征、草地演替度等分析,解析高寒草甸植物种间关系及草地演替方向。结果表明:(1)封育显著降低了草地群落植物物种数以及α和β多样性指数,显著增加了草地1年生植物种数、地面芽植物种数和地上生物量。(2)放牧草地地上生物量以莎草科(59.7%)和禾本科(23.9%)为主,封育草地地上生物量以禾本科(85.0%)为主;放牧草地优势种为矮嵩草(Kobresia humilis)和线叶嵩草(K.capillifolia),封育草地优势种为垂穗披碱草(Elymus nutans)和早熟禾(Poasp.)。(3)放牧和封育草地的群落植物种重要值均与各自的生态位宽度变化规律一致。(4)草地植物种间竞争主要发生在不同科属植物种之间,封育增加了群落植物种整体生态位重叠值和植物种间竞争。(5)草地群落演替度为封育地放牧地,封育群落处于较稳定状态。研究认为,封育促进了高寒草甸由莎草+杂类草群落向禾草+杂类草群落演替。     

半干旱黄土区不同管理措施下草地群落结构对短期氮、水添加的响应

氮沉降和降水变异显著影响草地群落结构和功能,但缺乏对不同管理措施下草地群落结构对氮沉降和降水变异响应的研究。为模拟不同管理措施下草地群落结构对氮沉降和降水变异的响应特征,以半干旱黄土区云雾山国家自然保护区典型草原为研究对象,系统分析了在封育、刈割和火烧三种管理措施下,氮添加和水添加对群落地上生物量、功能群组成和群落多样性的影响。结果表明,氮添加和水添加对地上生物量、功能群组成和群落多样性指数的影响因管理措施不同有所差异。（1）在封育草地上,氮添加显著降低物种多样性,对地上生物量影响较小;水添加显著增加物种多样性指数,氮添加和水添加的交互作用显著增加地上生物量、禾本科所占比例和莎草科所占比例;物种多样性指数均与地上生物量无显著相关,与不同功能群所占比例显著相关。（2）在刈割草地上,氮添加和水添加显著提高草地群落地上生物量,氮添加和水添加交互作用尤为显著;氮添加和水添加显著增加物种丰富度指数,对物种均匀度影响较小;杂草类所占比例和地上生物量对Shannon-Weiner多样性指数的贡献率较大。（3）在火烧草地上,氮添加和水添加显著提高群落地上生物量,对物种多样性的影响因年份不同有所差异,氮添加和水添加交互作用具有累加效应;Shannon-Weiner多样性指数与地上生物量呈显著负相关,与莎草科所占比例呈显著正相关。研究表明管理措施显著影响群落结构对氮添加和水添加的响应特征,亦改变生产力和物种多样性的关系模式,为更好地应对全球变化进行草地管理提供数据支撑。     

不同干扰类型对高寒草甸群落结构和植物多样性的影响

对高寒草甸天然草地进行了施肥、围栏和放牧 中牧和重牧 处理,研究不同干扰类型对草地植物多样性的影响.结果表明,施肥使草地植物群落物种组成贫乏,群落结构趋于简单,物种多样性减少;中等程度放牧增加了群落结构的复杂性,丰富度指数和多样性指数均最高,支持 中度干扰理论 ;重度放牧,由于干扰过于剧烈而减少了物种优势度和多样性;而轻度干扰的围栏草地,群落由少数优势种所统治,多样性也不高.物种数 S 、丰富度指数 Ma 、Shannon-Wiener指数 H' 、Simpson指数 D 的排列顺序均为:施肥草地<围栏草地<重牧草地<中牧草地;均匀度指数 Jsw 的变化趋势与上述各指数相同;优势度指数的变化趋势则相反,为施肥草地>围栏草地>重牧草地>中牧草地.4种干扰类型草地群落的生活型功能群基本一致,均由多年生禾草、多年生杂类草和莎草类组成,但各功能群在群落中所占比重及各功能群内所含物种数则大不相同.说明不同干扰类型对草地植物群落的物种组成、多样性格局及系统功能等方面产生不同的影响.     

长期封育演替下典型草原植物群落生产力与多样性关系

封育是退化草地的重要恢复措施, 理解长期封育过程中草地群落生产力和植物多样性变化特征及两者间关系, 有助于草地植被的恢复管理与利用。该研究依托宁夏云雾山国家级自然保护区典型草原长期封育演替梯度, 选择持续放牧、封育9年、26年和34年的草地群落作为研究对象, 分析其地上生产力、物种多样性和功能多样性的变化特征及内在联系。结果表明, 封育显著提高典型草原植物群落的地上生产力、凋落物生物量、功能丰富度和功能离散度, 未改变草地群落的物种丰富度、Shannon-Wiener指数和功能均匀度, 但Simpson优势度指数和Pielou均匀度指数在长期封育(34年)后显著下降。此外, 封育对不同植物群落加权平均功能性状的影响存在差异。随机森林模型和方差分解结果显示, 群落加权平均功能性状对封育草地群落地上生产力变异的解释度高达70.70%, 其中植株高度是最主要的解释因子; 功能多样性的解释度为36.86%, 主要由功能丰富度贡献; 而物种多样性的解释度仅为14.72%。由此可见, 植物功能性状和功能多样性对草地群落地上生产力的贡献远高于物种多样性, 建议将其纳入植物群落恢复演替动态研究, 以便全面了解植物多样性与生态系统功能的关系, 为更好地实现生态恢复目标奠定基础。     

不同干扰方式对内蒙古典型草原植物种组成和功能群特征的影响

研究休牧、补播和自由放牧3种人为干扰方式对内蒙古典型草原群落植物种组成、功能群组成、群落物种多样性和地上生物量的影响.结果表明: 3种干扰方式对植物功能群结构及其多样性有显著影响,休牧方式下群落植物种数量、地上生物量、多样性指数和丰富度指数最大,分别为22种、171.32 g·m^-2、1.46和5.7,灌木半灌木和多年生禾草的重要值、物种所占比例、地上生物量、多样性指数、丰富度指数和均匀度指数显著高于其他干扰方式,多年生杂类草的重要值、物种所占比例、地上生物量、多样性指数和均匀度指数显著低于其他干扰方式,中生植物、C3和C4植物的地上生物量最大,分别为22.22、143.35和27.97 g·m^-2.补播方式下,中旱生植物地上生物量最大,为13.60 g·m^-2,旱生植物物种所占比例最小(48.5%),C4植物物种所占比例最大(28.8%).自由放牧方式下,一、二年生草本植物物种所占比例最大,达12.3%,中生植物和C4植物所占比例最小,分别为17.0%和20.9%.春季休牧措施有利于退化草原的恢复.     

不同干扰形式对砂生槐群落结构特征的影响

在雅鲁藏布江中游流域以放牧、封育和水淹3种不同干扰形式下的砂生槐群落为研究对象,从林分特征、物种组成、优势种种群发育状况和物种多样性等对群落特征进行对比研究,以揭示不同干扰形式对砂生槐群落结构的影响.结果显示:(1)砂生槐群落以群落A(放牧)的林分状况最好,其次为群落B(封育),群落C(水淹)最差;(2)不同干扰形式下群落中植物种类分别为群落A(13种)＞群落C(9种)＞群落B(6种);(3)优势种群生长状况说明放牧和封育形式下砂生槐生长情况基本一致,高度级和幅度级小的个体均较多,而水淹干扰下群落仅有少数生长旺盛的个体存在,种群呈衰退趋势;(4)物种多样性指数为群落A＞群落B＞群落C.研究表明:一定程度放牧干扰能够促进砂生槐种群更新,增加群落物种丰富度和物种多样性,封育可促进砂生槐种群萌蘖更新,而水淹则促使群落衰退.     

青藏高原高寒灌丛植被对长期放牧强度试验的响应特征

在青藏高原中国科学院海北高寒草甸生态系统定位研究站对金露梅高寒灌丛草场植被开展了长期不同放牧强度试验,分别在短期(4年)、中期(11年)和长期(18年)放牧阶段研究不同放牧干扰强度对草地植物物种多样性、群落结构、地上生物量和草场质量的影响.研究表明,在不同放牧阶段,随着放牧强度增加植物群落的高度和盖度都降低.在中期放牧干扰阶段,物种多样性数和均匀度指数随着放牧强度增加呈现典型的单峰曲线模式;在长期放牧干扰阶段,随着放牧强度增加,占优势地位的灌木和禾草被典型杂类草替代,其中的重度放牧干扰简化了高寒灌丛植被群落结构,减少了地上现存生物量,特别是可食优良牧草生物量.植被对放牧的响应除了与放牧强度和放牧时间阶段密切相关外,还与该地区水热条件的变化有一定的相关性.针对长期放牧干扰的反应特性可将金露梅灌丛草场中植物划分为增加型、敏感型、忍耐型和无反应型4种类型.除了丰富度指数、多样性指数和均匀度指数外,其它一些特征参数并不支持著名的中度干扰假说.本研究发现,长期重度放牧促进了青藏高原高寒草地退化,适度放牧有利于高寒灌丛草场的生物多样性保护和牧草利用;"取半留半"的放牧原则在青藏高原草场放牧管理实践中值得推荐,它将有利于防止草场退化,提高牧草利用率和维持较高的生物多样性.     

北京东灵山地区植物群落多样性的研究Ⅱ丰富度、均匀度和物种多样性指数

本文以１２５块植物群落调查样地资料为基础，从不同类型、层次的丰富度、均匀度和物种多样性指数及其与海拔的关系等方面对东灵山地区植物群落多样性进行了分析。本区亚高山草甸植物群落多样性沿海拔梯度的变化规律是：物种丰富度和物种多样性指数随海拔升高而下降；物种均匀度则随海拔升高而增加。植物生长型与群落多样性指数的关系表现为“乔木层、灌木层物种丰富度指数相近且明显低于草本层；灌木层和草本层的均匀度指数相近，群落间变异幅度较小，乔本层则变异幅度很大；物种多样性指教则表现出草本层＞乔本层＞灌木层的规律。物种盖度和地上生物量作为测度指标计算群落多样性所得结果相近，且优于以株数作为测度指际计算的结果。     

退化土地植被恢复群落物种多样性与生物量分配格局

研究了湘西北女儿寨小流域7种典型植被恢复群落生物量的空间分配格局、物种多样性特征及其相互关系。结果表明:1)润楠次生林乔木层生物量最高,马尾松天然林灌木层生物量最大,油桐人工林草本层生物量最高,乔木层最低,毛竹-杉木林灌木层与草本层生物量最小,群落总生物量以马尾松天然林群落最高,荒草灌丛群落最低。2)乔木层物种以润楠次生林最丰富,多样性指数最高,物种分布最均匀;灌木层物种丰富度以毛竹-杉木混交林最高,油桐人工林多样性与均匀度指数最大,优势度最小,荒草灌丛群落的物种丰富度、多样性及均匀度最低;草本层物种丰富度以油桐人工林最高,马尾松天然林最低,杉木人工林草本层多样性指数与均匀度指数最大,润楠次生林优势度最高;群落总体物种丰富度以油桐人工林最高,润楠次生林群落的多样性指数最高,种群分布最均匀,荒草灌丛群落的多样性、均匀度与优势度都最低。3)女儿寨小流域退化土地物种丰富度与群落生物量之间的关系可用"S"曲线较好地描述,双曲线能较好地描述群落多样性指数、群落均匀度指数与生物量之间的关系,而生态优势度与生物量之间为非性线关系。     

围栏禁牧对内蒙古典型草原群落特征的影响

采用样方法对内蒙古典型草原区围封26 a(E26)7、a(E7)2、a(E2)和未围封(G)样地的植物多样性、群落高度、盖度、地上与地下生物量进行实地调查分析,以研究围栏禁牧对草原植物群落特征的影响.结果表明:围封26a样地具有最高的丰富度指数和多样性指数,围封相对于放牧在一定程度上可以维持和保护植物多样性和丰富度.围封7 a样地的多样性指数与未围封样地(G)以及围封2 a样地(E2)无显著差异,说明放牧导致了该区植物多样性的降低,但围封并未显示出使其恢复的趋势;从自由放牧样地到围封样地,退化指示植物糙隐子草(Cleistogenessquarrosa)以及一年生植物猪毛菜(Salsola collina)、刺穗藜(Chenopodium aristatum)等的比重和出现频率呈减少的趋势;围封2 a样地的地上现存量显著低于围封7 a样地,而围封26 a样地相对于围封7 a样地又表现出显著减小的结果.说明围封在一定时期内可显著提高草地生产力,但围封时间过长将不利于其维持较高的生产力.     

青海湖区针茅草原植物群落特征及群落多样性研究

根据样方调查资料,采用聚类分析法,研究了青海湖区针茅草原围栏封育后植物群落特征及群落多样性变化。结果表明,草原经过长期围栏后,群落内优势种发生了不同程度的分异,由围栏外的紫花针茅(Stipa purpurea) 青海苔草(Carex ivanovae)草原演变成围栏内的冷地早熟禾(Poa crymophila) 猪毛蒿(Artemisia scoparia)草原群落,导致了群落结构特征的变化;长期的围栏活动对提高草原群落的盖度和生产能力是有益的,但却降低了群落的物种丰富度和多样性;草原开垦后,对周围草原有明显的影响,所形成土垄的群落特征和多样性变化都发生了极大的差异,成为赖草(Leymus secalinus) 大籽蒿(Artemisia sieversiana)群落类型,群落的物种丰富度和多样性均降低。     

Linking Community and Ecosystem Processes: The Role of Minor Species

The link between species and ecosystem functioning is a central issue in ecology. In natural plant communities, the dominant species determine most of the productivity-related processes but what is the function of minor species? A recent hypothesis suggests that after disturbance, minor species facilitate the recruitment and abundance of dominants during re-colonization, thus indirectly determining ecosystem function. We tested this hypothesis using a long-term dataset of annual plant communities in a semiarid shrubland by comparing plant density and biomass from plots in which all vegetation had been removed; plots from which only the dominant (the annual grass Stipa capensis) had been removed, and control plots. In the absence of vegetation, the dominant failed to re-establish during the following growing season. After being removed the dominant re-established similarly to the controls. An ant exclosure experiment excluded the possibility that this was due to seed predation. In an experiment with individual dispersal units of S. capensis, we demonstrated the mechanism by which minor species can control the dominant’s abundance. Minor species indirectly govern ecosystem processes by providing structures facilitating seed soil penetration and thus recruitment of the dominant.     

Understory cover and biomass indices predictions for forest ecosystems of the Northwestern United States

The understory community is a critical component of many processes of forest ecosystems. Cover and biomass indices of shrubs and herbs of forested ecosystems of Northwestern United States are presented. Various forest data were recorded for 10,895 plots during a Current Vegetation Survey, over the National Forest lands of entire Pacific Northwest. No significant relationships between the percent canopy cover and understory percent cover and biomass indices were found for the 129 ecoclasses analyzed. Disturbance time and type, and the soil characteristics significantly influenced the shrub biomass indices (p-values of <0.001, <0.001, and 0.01, respectively). Only disturbance time and type significantly influenced the shrub percent cover (p-values <0.001). There were no significant interactions between these variables. No significant differences were found for herb biomass indices and cover. Climate variables are reasonable predictors of understory cover and biomass indices. Elevation and slope are also influential: understory cover decreases with altitude, while understory biomass increases with slope. Most models showed weak predictive power (adjusted R-squared ≤ 0.27). However, robust models for the maximum/potential understory biomass indices for the forested areas in the Northwestern United States are reported (adjusted R-squared of 0.76 and 0.51 for shrubs and herbs, respectively). Overall, our study provides conceptual and statistical models for the understory of the National Forest lands of the Pacific Northwest. The results are comparable with other models for the area, suggesting that the predictions regarding understory vegetation are inherently difficult.     

Vegetation recovery and plant facilitation in a human-disturbed lava field in a megacity: searching tools for ecosystem restoration

Unplanned urban development threatens natural ecosystems. Assessing ecosystem recovery after anthropogenic disturbances and identifying plant species that may facilitate vegetation regeneration are critical for the conservation of biodiversity and ecosystem services in urban areas. At the periphery of Mexico City, illegal human settlements produced different levels of disturbance on natural plant communities developed on a lava field near the Ajusco mountain range. We assessed natural regeneration of plant communities 20 years after the abandonment of the settlements, in sites that received low (manual harvesting of non-timber forest products), medium (removal of aboveground vegetation), and high (removal of substrate and whole vegetation) disturbance levels. We also tested the potential facilitative role played by dominant tree and shrub species. Plant diversity and vegetation biomass decreased as disturbance level increased. Sites with high disturbance level showed poor regeneration and the lowest species similarity compared to the least disturbed sites. Six dominant species (i.e., those with the highest abundance, frequency, and/or basal area) were common to all sites. Among them, three species (the tree Buddleja cordata, and two shrubs, Ageratina glabrata and Sedum oxypetalum) were identified as potential facilitators of community regeneration, because plant density and species richness were significantly higher under their canopies than at open sites. We propose that analyzing community structural traits of the successional vegetation (such as species diversity and biomass) and identifying potential facilitator species are useful steps in assessing the recovery ability of plant communities to anthropogenic disturbances, and in designing restoration strategies.     

The effect of earthworms on plant diversity and soil properties under different landuses

Earthworms have significant effect on vegetation and soil physical and chemical properties. The relationship between earthworms and biodiversity is important in ecological researches. Therefore, the effects of earthworm on plant diversity and various indices of diversity and richness, as well as physical and chemical properties of soil in shrub, grass and bare vegetation types on exclosure and grazing sites of Miankaleh Mazandaran biosphere reserve is investigated in this work. Accordingly, sampling was carried out randomly-systematically in three different vegetative types (shrubs, grass and bare soil) at 0 to 20 cm depth. Each vegetative types was placed on three transects of 100 m, 10 plots at a distance of 30 m. Earthworm and soil were sampled from the center of each plot of 25 cm × 25 cm. The Simpson and Shannon-Wiener indexes and Margalef and Menhinick richness indices were used to evaluate diversity and richness of species at different grazing intensities. Stepwise regression was used to study the relationship between diversity and richness indices with different earthworm parameters. The results showed that there was significant difference of potassium and nitrogen in vegetation types of both exclosure and grazing sites at P ≤ .01. Furthermore, the organic carbon, organic matter and potassium in the exclosure site and nitrogen and soil moisture in the grazing site are increased by earthworm abundance.     

Ungulate herbivory reduces abundance and fluctuations of herbivorous insects in a boreal old-growth forest

Ecological theory predicts the strongest ecosystem effects of herbivory when dominant and ecologically important species are consumed. Bilberry, Vaccinium myrtillus, is such a key plant species, attractive to many other species in the boreal forests, for example ungulate and invertebrate herbivores. Large herbivores may remove substantial biomass and alter plant quality and therefore affect abundance and populations of invertebrate animals sharing the same food plant. We combined experimental exclusion of ungulates with a browsing intensity gradient to investigate the 15-year effect of ungulate (Cervus elaphus and Ovis aries) browsing on bilberry plant size and on bilberry-feeding herbivorous larvae (Lepidoptera and Symphyta), in a Norwegian old growth boreal forest ecosystem. Bilberry ramets in exclosure plots had nearly nine times higher dry mass and three times higher abundance of invertebrates feeding on them than in ungulate-access plots. Sweep-netting data verified these findings as larval numbers were twice as high in exclosure plots. The pattern in the large herbivore effects on bilberry size and abundance of herbivorous larvae were identical along the browsing gradient. Differences in larval abundance between treatments, as indicated by leaf-chewing, increased during the 15-year study period, and the community fluctuations were larger when ungulate herbivores were excluded. The browsing effect was moderated by plant quality as larval densities were lowest on both heavily-browsed and non-browsed plants, and highest on ramets that had 50–74% of annual shoots browsed. Our study supports previous findings in that bilberry is relatively disturbance tolerant and may recover quickly, but that ungulates may compete with herbivorous larvae for food biomass. Additionally, our results strongly indicates that population insect community peaks and fluctuations are dampened by ungulate consumption. Our findings add to the understanding on how ungulates may structure forest ecosystems directly and indirectly.     

Effect of grazing on community structure and productivity of a Uruguayan grassland

Grasslands and their grazers provide some of the most compelling examples for studying the relationship between diversity, productivity, and disturbance. In this study, we analyzed the impact of grazing-induced changes in species composition and community structure upon the productivity of a grassland in the Campos region, Uruguay. We compared three treatments: a continuously grazed area, a 9-year old exclosure to domestic herbivores, and grazing-simulated plots inside the exclosure, which were clipped so that their standing biomass resembled that of the grazed area. We studied the community composition of the grazed and ungrazed situations, and determined biomass and above-ground net primary production (ANPP) of the three treatments during 1 year. Grazed plots had higher species richness and diversity than the exclosure. Grazing resulted in the replacement of some cool-season, tussock grasses by warm-season, prostrate grasses. ANPP was 51% higher under grazing than in the exclosure, but the grazing-simulated plots inside the exclosure were the most productive treatment, 29% higher than the grazed plots. Thus, two components of grazing effect may be postulated for this grassland. The structural component resulted in higher ANPP, probably due to the elimination of standing dead biomass. The species composition component resulted in lower ANPP once the structural component was controlled, probably due to the shift to warm-season phenology and prostrate habit. Our findings contrast with a similar experiment carried out in the neighbouring Flooding Pampa region, which suggests that the relationship between grazing and community structure and function is difficult to generalize.     

模拟N沉降对太岳山油松人工林和天然林草本群落的影响

由于人类活动氮沉降呈逐年增加的趋势,进而增加了陆地生态系统氮的输入,从而影响陆地生态系统多样性、物种组成和功能。为揭示氮沉降增加对油松林草本群落的影响,于2009年7月在太岳山油松人工林和天然林,设计4个施氮水平:对照(CK,0 kg N hm-2a-1),低氮(LN,50 kg N hm-2a-1),中氮(MN,100 kg N hm-2a-1)和高氮(HN,150 kg N hm-2a-1),研究草本群落的生物多样性、生物量以及草本元素含量对模拟N沉降的响应。研究结果表明:模拟N沉降未能显著影响人工林草本群落的生物多样性(P0.05),而中氮、高氮显著降低了天然林草本群落的生物多样性(P0.05);从Jaccard指数和Sorensen指数分析得出人工林不同氮水平之间草本群落差异性较小,而天然林不同氮水平之间草本群落差异性较大。模拟N沉降没有显著改变人工林草本群落生物量(P0.05),而高氮明显促进天然林草本群落生物量的增加(P0.05)。与对照相比,模拟N沉降提高了人工林和天然林羊胡子苔草叶根中的全N含量(P0.05),而降低了全Mg的含量(P0.05),并且根部元素含量变化与土壤养分含量变化较为一致。施氮提高了N/K、N/Ca、N/Mg(P0.05)的比值。说明油松林下草本群落对氮沉降的响应因林分土壤N饱和程度以及林地利用历史的不同而产生差异,其中天然林响应最为敏感。     

乡土植物生活型构成对川渝地区边坡植被水土保持效益的影响

乡土植物灌木化建植是高速公路边坡防护的重要生态模式。为了探讨生态恢复过程中乡土植物的生活型构成对边坡植被水土保持效益的影响,揭示乡土植物生活型-物种多样性-生态系统功能间的偶合关系,借助3个物种配置试验,于2009年4月构建了以草本、灌木或乔木为主体的草本型、灌木型、乔木型绿化配置及草-灌-乔混合型试验区。自建植次年(2010年)起,对试验区进行持续5a生态监测。结果表明:1)边坡植被的物种丰富度与乡土植物的生活型有关,呈现乔木型灌木型草-灌-乔混合型草本型趋势;2)植被的物种多样性(Shannon-Wiener指数、Pielou指数)与乡土植物生活型构成及建植年限有关:建植后第1、第2年,多样性水平呈草本型草-灌-乔混合型灌木型乔木型变化趋势;自建植后第三起(2012—2014年),呈草-灌-乔混合型草本型灌木型乔木型波动;3)植被的水土保持性能(径流系数、侵蚀模数)与群落的物种多样性密切相关:多样性水平越高,水土保持性能越强。可见,乡土植物的生活型构成对提高边坡植被的物种多样性、改善生态性能至关重要。