摘要
以黄河下游典型农区封丘县为研究区,在林地景观中进行地表节肢动物的观测。用物种丰富度和香农多样性指数代表物种多样性,选择代表景观背景的5个竞争模型:生境特性(H1,2012)、基质特性(H2,2012)、生境变化(H3,1984—2012)、基质变化(H4,1984—2012)和土壤-环境条件(H5,2012)从4个空间尺度上(100,250,350和500 m)进行分析,通过运用基于赤池信息量准则(Akaike information criterion,AIC)的多模型推理(Multi-model Inference,MMI)方法,在R软件里用广义线性模型(Generalized Linear Models,GLM)探究了研究区近30年(1984—2012年)景观背景变化对林地地表节肢动物多样性的影响。研究表明,不同景观背景模型对地表节肢动物多样性的影响具有尺度依赖性。在100 m的尺度下,生境特性(H1)最能够解释香农多样性和物种丰富度,但是随着尺度的增加,生境特性变化(H3)在较大(250、350 m和500 m)的尺度对物种丰富度和香农多样性影响最大,而基质特性和土壤-环境条件(H2和H5)的作用不显著。景观背景对地表节肢动物多样性的解释量达到40%。在研究区域,生境特性是表征香农多样性指数和物种丰富度的指标。
Biodiversity is one of the most limiting factors for ecosystem services. It is well known that there are very close connections between agricultural ecosystems and species diversity. Large numbers of non-agricultural wildlife rely on the food and habitats provided by the agricultural landscape. Natural and semi-natural habitats （such as farmland boundaries, woodland, hedges, and creeks ） in agro-landscapes have very important effects on ground arthropods inhabiting these microenvironments. The lower reaches of the Yellow River, China, is a major grain producing areas in China, and is important for the security of the national food supply. It is crucial to study the relationships between regional agro-landscape heterogeneity and biodiversity to maximize ecosystem services and to maintain ecosystem stability. In this study, Fengqiu County in Henan Province, which is a typical agricultural region in the lower reaches of the Yellow River, was selected to investigate the effects of agricultural landscape dynamics on the species diversity of ground arthropods in woodlands. The Shannon diversity index and species richness were applied to explore the relationships between landscape changes and arthropod diversity from 1984 to 2012. The role of the landscape context on ground arthropod diversity in woodlands was evaluated by a multi-model inference （MMI） based on the Akaike information criterion （AIC）. Five competing models of landscape context were used to analyze how it impacts arthropods. These models included habitat quality （ HI, 2012）, matrix quality （H2, 2012）, habitat change （H3, 1984--2012）, matrix quality change （H4, 1984--2012）, and soil- environmental conditions （H5, 2012）. All of these models were measured respectively at four spatial scales （100, 250, 350, and 500 m） to examine how spatial scales impact ground arthropod diversity. The relationship between the response variables and the predictor variables was analyzed by Generalized Linear Models （GLM） in R statistical software. From 1984 to 2013 in the study area, woodland area increased significantly, while agricultural landscape heterogeneity increased substantially. The dominant groups of ground arthropods included Hymenoptera, Coleoptera, Amphipoda, and Araneae. The impacts of landscape context on ground arthropod diversity in woodlands were significantly scale-dependent. Habitat quality （ H1 ） was the highest supported model for explaining Shannon diversity index and species richness of the ground arthropod community at a small scale （ 100 m）. As the spatial scale increased, the Shannon diversity index and species richness responded mostly to habitat change quality （ H3 ） in the broader surroundings （ 250, 350, and 500 m）. No significant correlation was observed between matrix quality and soil-environmental conditions （H2 and H5 ）. The responses to landscape context of ground arthropod species richness and Shannon diversity index were fairly consistent, significantly responding to habitat quality. Approximately 40% of variation in ground arthropod diversity was explained by the landscape context. In the study area, habitat quality （H1） may represent a good indicator for the Shannon diversity index and species richness. The analytical results showed that historic landscape change also has a predominant influence on the present features of ground arthropod diversity in woodlands ; thus, the temporal dynamics of the landscape context may help towards improving our understanding about the interactions between landscape context change and ground arthropod diversity in woodlands. To enhance the explanation for the effects of landscape context on ground arthropods, further studies should consider additional landscape competing models and multi-scale spatiotemporal analysis.
出处
《生态学报》
CAS
CSCD
北大核心
2015年第23期7659-7668,共10页
Acta Ecologica Sinica
基金
国家自然科学基金资助项目(41071118
41371195)
关键词
多模型推理
赤池信息量准则
地表节肢动物
生物多样性
林地
农业景观
multi-model inference
Akaike information criterion
ground arthropods
biodiversity
woodlands
agricultural landscape
