【摘要】 在植物生长期 ,采用夹日法对放牧扰动下的祁连山北坡山地荒漠草地小型哺乳动物群落多样性空间分布格局进行测度 ,结果表明 :在不同的放牧强度条件下 ,小型哺乳动物群落物种组成、结构和多样性指数受相应影响 ,物种丰富度和多样性指数随放牧压力增加而减小 ,经 Whittaker的相似性指数的聚类分析 ,在相似性指数大于 0 .4的水平上 ,可把研究区小型哺乳动物群落分为被家畜啃食和未被家畜啃食 2个类群。研究结果进一步验证了大型食草动物放牧干扰对改变小型哺乳动物群落结构具有重要作用的假设更多还原

【Abstract】 Selective grazing and high stocking rates of domestic animals have marked effects on the composition and shaping of plant communities in arid and semiarid ecosystems. Accordingly, vegetation changes superimposed by grazing might therefore also influence species diversity and the composition of small mammal communities. The composition of small mammals in many vegetation types in the northwestern arid regions of China is incompletely known. We also do not know what effect habitat changes caused by grazing have on small mammal communities. We investigated the species distribution and abundance of small mammals in different grazing gradients in a montane desert-steppe and used them as indicators for monitoring environmental changes by grazing, and to explore how these changes would affect spatial distribution patterns and local biodiversity in a more general sense. The study was conducted in the Xishui Protection Station of Qilianshan Nature Reserve and its surrounding regions (N (38°33′)~38°36′, E 100°19′~100°33′), which is lying at 1950~2290 m elevation in the midsection of the northern slope of Range Qilianshan, located about 76 km south to Zhangye City, in northwest China. Temperatures range from a mean January minimum of -5.0~-8.5℃ to a mean July maximum 19.0~23.0℃. Average annual precipitation in this region is 160 mm~230 mm, with about half occurring among June and September. The dominant plant formation is arid shrubland and herbaceous, shrubland dominated by the Kalidium foliatum、Salsola passerina、Reaumuria soongorica、Potentilla parvifolia, and the herbaceous layer is mostly comprised of perennial plant in the following species: Sympegma regelii, Stipa breviflora, Suaeda glauca, Stellera chamaejasme. Grazing intensity along the gradient was quantified by dung collection from 10 1×1 m~2 plots at each of the six focal sites. Dung were dried for 48 h at 80℃ and weighted. Accumulated dung data for all plots and collection dates were pooled for each focal sites, this indicated that the frequency of animal visitation decreased with distance from the village. To develop an effective and sufficient sampling protocol, the field study was conducted in July - August 2002 during the middle of the wet season when food for small mammal was plentiful and no species were estivated. Small mammals were trapped using snap-traps at six sites along a grazing gradient. At each site, we set two trap lines separated by about 100 m. Within each trap line, 100 snap-traps (165mm × 70mm) were each set about 5 m apart for a total of 200 traps per site. Traps were baited with oil-fried peanuts. Traps were examined and rebaited each morning and late afternoon. Each site was surveyed for three days, for a total of 600 trap-nights per site, 3600 trap-nights in the study as a whole. Trap success was defined as the number of captures per 100 trap-nights. In order to uncover the responses of species diversity of site to grazing pressure, a measurement on species richness and α-diversity in each of these sites was conducted.During the study 351 individuals of eight species of small mammals were captured. Allactaga sibirica (37.32%) was the most commonly trapped dessert rodent, and next was Dipus sagitta (27.35%) Meriones meridianus(19.37%). Ochotona erythrotis and Cardioranus paradoxus were rarely encountered, the trapping success was 0.57% and 0.28%, respectively. The result showed that the species composition, structure, and diversity of the small mammals were changes with the intensity of grazing pressure. The species richness and diversity of the small mammals were highest under un-grazing site and declined with increased grazing pressure and reached minimum under over-grazing pressure. The species richness and diversity of the small mammals showed significantly negatively correlations to grazing intensity, respectively. Based on similarity indices, six grazing sites can be clustered at the level of 0.2 with systematic clustering. When a threshold of 0.4 for the similarity index was considered, the six grazing sizes could be cate更多还原