Search for a command to run...
Abstract Aim To maintain biodiversity in plantation forest ecosystems it is important to understand the community assembly process. Canopy thinning effectively increases understorey diversity in plantation forests. However, the process of understorey community assembly after canopy thinning remains unclear. We compared the relative importance of two mechanisms that could affect the assembly of understorey woody communities after thinning: (1) environmental filtering due to competition from canopy trees, and (2) stochastic immigration. If the latter is dominant, the effectiveness of local treatment might be unpredictable because of the variable probability of immigration at each location. Location Yamagata Prefecture, Japan. Methods We censused the understorey tree communities in 18 quadrats (each 1 m × 1 m) in each of 49 stands of Japanese cedar plantations. The mean species richness of the 18 quadrats (alpha‐diversity), variation in community composition among them (beta‐diversity), deviation of beta‐diversity from that of a random community (beta‐deviation) and total species richness of the plots (gamma‐diversity) were compared among censuses (before thinning, 2 and 5 yr after thinning). The relative contributions of two explanatory variables to the various diversities were examined: (1) basal area of canopy trees (overstorey BA, an index of environmental filtering), and (2) available seed source in the surrounding landscape (landscape effect; an index of the probability of immigration). Results Although neither beta‐diversity nor gamma‐diversity changed throughout the censuses, the alpha‐diversity, stem abundance and beta‐deviation were affected by thinning. beta‐deviation increased after thinning. Overstorey BA limited alpha‐ and gamma‐diversity throughout the censuses. The landscape effect was more significant in terms of beta‐deviation after than before thinning. In addition, the effective scale of the landscape effect increased after thinning. Conclusion Our results indicate that both processes are important: thinning brought about an increase in the relative importance of stochastic immigration as well as a reduction in competitive pressure from canopy trees. These results suggest that while local treatment can increase understorey diversity, stochastic immigration from seed sources is the direct driver after thinning. Although local treatment is effective, its efficiency depends on the seed source availability in the surrounding landscape.