Biodiv Sci ›› 2006, Vol. 14 ›› Issue (6): 525-533.DOI: 10.1360/biodiv.060052

• Special Issue • Previous Articles     Next Articles

Rodent community structure of desert-oasis landscape in the middle reaches of the Heihe River

Lingying Shuai1, 2, Yanling Song1*,Junsheng Li3, Zhigao Zeng1, Jianquan Liu4   

  1. 1 Key Laboratory of Animal Ecology and Conservation Biology, Institute of Zoology, Chinese Academy of Sciences, Bei-jing 100080
    2 Graduate School of the Chinese Academy of Sciences, Beijing 100049
    3 Chinese Research Academy of Environmental Sciences, Beijing 100012
    4 Qilianshan National Nature Reserve, Zhangye, Gansu 734000
  • Received:2006-08-25 Revised:2006-09-26 Online:2006-11-20 Published:2006-11-20
  • Contact: Yanling Song

Abstract: By using living trapping, we studied the characteristics of rodent communities in desert-oasis landscape in the middle reaches of the Heihe River, Gansu. Field study was carried out from April to May, and September in 2005. A total of 4,800 living trappings were set in eight typical habitats, including sand dune, gravel-sandy desert, stabilized sandy desert, desert scrubland, shelter belt, irrigated grassland, irrigated crop field and riverine grassland. In total, 254 individuals representing 9 species, 8 genera, and 3 families were captured. We then calculated characteristic indices of community structure for the eight habitats. The species diversity index (Shannon-Wiener index) ranged from 0.6859 (irrigated grassland) to 1.7036 (stabilized sandy desert), while Pielou evenness index ranged from 0.6531 (sand dune) to 1.0000 (riverine grassland). Using Pearson correlation coefficients and hierarchical clustering, we could roughly classify the 8 ro-dent communities as 2 major types, i.e. desert type and oasis type. Desert-type habitats generally harbored higher species diversity and population density. Dipus sagitta showed a high density and was dominant in desert-type habitats, especially in sand dune. Cricetulus barabensis was common in habitats with relatively high vegetation coverage but also recorded in desert-type habitats (sand dune excluded). Among the four jerboa species, only Allactaga sibirica was found in oasis-type habitats. We did not capture any rodent in core region of shelter belt. However, we did record some individuals in the edge region. No significant rela-tionship was found between Shannon-Wiener diversity index and capture rate(r=0.240, P=0.566). Our results suggest that human disturbance in study area may have been beneficial to maintain rodent diversity.