Distribution of carabid beetles among regenerating and natural forest types in Southwestern China Xiao-Dong Yu a,b , Tian-Hong Luo a,b , Hong-Zhang Zhou a, * a Institute of Zoology, Chinese Academy of Sciences, Beijing 100080, PR China b Graduate School, Chinese Academy of Sciences, Beijing 100039, PR China Received 10 October 2005; received in revised form 10 May 2006; accepted 17 May 2006 Abstract Heterogeneity in species assemblages of carabid beetles was studied between regenerating larch plantations of different forest ages and natural deciduous broad-leaved forests in Southwestern China. Larch plantations that were dominated by the non-native tree species (Larix kaempferi) interspersed with a small proportion of native tree species (L. mastersian) were established after the clear-cutting of the natural deciduous broad- leaved forests. Using pitfall traps, we compared the distribution of carabid beetles in recently planted (5 years old), young (15 years old), mature plantation (45 years old) and adjacent natural deciduous broad-leaved forests (ca. 100 years old). Four replicated plots for each plantation and five replicated transects in natural broad-leaved forest were chosen, and each plot or transect was investigated with four trap locations twice each month during the growing season (May–October) in 2004. Our results showed that species richness, abundance and Shannon diversity were significantly higher in the natural forest than in the regenerating plantations. In addition, these indices were not significantly different in 15- and 45-year-old plantations but were higher than those in 5-year-old plantation. Principal coordinate analysis (PCoA) ordination grouped the carabid assemblages according to forest type and forest age, suggesting that there is a general pattern of recovery after logging. Moreover, indicator species analysis detected significant species associated with forest types and ages. Based on the specificity and fidelity, carabids can be distinguished into six species groups: habitat generalists, generalists in forests with closed canopy, specialists in natural broad-leaved forest, generalists in plantations with closed canopy, young closed canopy larch plantation-associated species and open canopy larch plantation-associated species. Multiple linear regression showed that the depth of leaf litter, the cover of canopy and shrub layer had significant effects in determining richness, abundance and diversity of carabid beetles. Therefore, the regenerating larch plantations, even mature plantations, could not replace the role of the natural deciduous broad-leaved forests in sustaining the diversity of carabid assemblages. Furthermore, establishments of large areas of relatively homogeneous larch plantations may have detrimental long-term effects on the composition and structure of carabid assemblages. Thus, it is important for us to understand the subtle variations in habitat and take appropriate actions for maintaining the invertebrate diversity before the logging of natural forests and establishments of plantations. # 2006 Elsevier B.V. All rights reserved. Keywords: Carabid beetles; Diversity; Larch plantation; Natural deciduous broad-leaved forest; Forest regeneration; Indicator species 1. Introduction Forestry practices, mainly the clear-cutting and establish- ment of young plantations, are causing increasing fragmenta- tion and changing forest structure in China. Since the 1960s, as an important strategy for reforestation, extensive plantations of economic coniferous tree species such as larch (Larix), pine (Pinus), arborvitae (Platycladus), spruce (Picea) and fir (Abies) have been established in several mountain ranges in China. Forestry has significantly contributed to the loss of natural habitats (MacKinnon and Xie, 2001). Much effort has been put into reforestation, but few studies have been conducted on the general response of forest communities to the logging or of their recovery with reforestation. Ground beetles (Coleoptera: Carabidae) are an appropriate group for such studies as they have a well-studied ecology and systematics, and they are sensitive (Eyre et al., 1996; Lo ¨vei and Sunderland, 1996; Niemela ¨ et al., 2000; Werner and Raffa, 2000) and respond rapidly to habitat changes (Lenski, 1982; Jennings et al., 1986; Davies and Margules, 1998; Duchesne et al., 1999; Kotze and Samways, 1999; Niemela ¨, 2001). Previous research has been conducted on the effects of conifer plantations and forest succession on the composition www.elsevier.com/locate/foreco Forest Ecology and Management 231 (2006) 169–177 * Corresponding author. Tel.: +86 10 6256 2794; fax: +86 10 6256 5689. E-mail address: zhouhz@ioz.ac.cn (H.-Z. Zhou). 0378-1127/$ – see front matter # 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.foreco.2006.05.043