Metacommunity patterns of ground-beetle assemblages in two mixed broad-leaved Korean pine (Pinus koraiensis) forests in the Xiao Xing'an Mountains
A metacommunity is defined as a set of local communities that are linked by the dispersal of multiple potentially interacting species. Metacommunity theory framework provides a powerful method for the understanding of spatio-temporal patterns and explaining the underlying processes of a community structuring. Metacommunity framework has been used in various organism communities (including plant, bat, arbuscular mycorrhizal fungi, bird, and invertebrate communities). However, few studies have been used to recognize the metacommunity patterns and their underlying processes for soil animal communities at multiple scales, especially at small scales. In 2015, an experiment was conducted in two permanent dynamic monitoring plots in two typical mixed broad-leaved Korean pine (Pinus koraiensis) forests in the Liangshui (LS) and Fenglin (FL) National Nature Reserves. Each plot consisted of a 9-hm² area with 300 m × 300 m and was equally divided into 225 squares by 20 m. Based on a spatially explicitly investigating method, ground beetles were collected by pitfall traps in July, August, and October in the LS plot and in August and October in the FL plot. The topographic variables, that is elevation, slope convexity, slope steepness and slope aspect, were measured for each square both in LS and FL plots. The soil variables, including bulk density, soil organic carbon content, total nitrogen, available phosphorus, total phosphorus, pH, available potassium, hydrolyzable nitrogen and soil moisture, were measured for each square in LS plot. The metacommunity patterns of ground beetles were determined by elements of metacommunity structure (EMS). We evaluated if the Clementsian pattern is common for ground beetle metacommunities at such small scales. Then we explained the association between the site scores along the first axis of the reciprocal averaging and geographic and/or soil variables by Spearman rank correlation. Results of the EMS showed that nestedness and random patterns were common for ground beetle metacommunities at small scales, such as in the two studied typical mixed broad-leaved Korean pine forests. The Clementsian pattern was not detected for any ground beetle metacommunity. Moreover, metacommunity patterns of ground beetles were group-dependent. Metacommunity patterns were nestedness for Carabid and Silphidae beetles, while metacommunity patterns were random for Staphylinidae beetles. At the same time, metacommunity patterns of ground beetles were seasonal dynamic. The nestedness patterns were formed in warm seasons (i.e. July and August), and the random patterns were formed in a relatively cool season (October). Metacommunity patterns between the LS and FL plots were not significantly different. The site scores along the first axis of the reciprocal averaging were correlated with topographic and/or soil variables for Carabid, Silphidae, and total beetle communities. Important factors included mean elevation, aspect, slope, total nitrogen, pH, and soil organic matter content. These variables were suggested influences on ground beetle metacommunity patterns. However, the influences were relatively complicated and were group-dependent and season-dependent. This study suggests that nestedness and random patterns are common for ground beetle metacommunities at small scales in the typical mixed broad-leaved Korean pine forests in the Xiao Xing'an Mountains. Additionally, topographic and soil variables may be important factors for formatting these metacommunity patterns.