Quantifying the roles of historical versus contemporary constraints in determining species diversity is a central issue in island biogeography, and the phylogenetic beta diversity between islands is an essential measure specifying the influence of historical barriers on insular assemblages. In this study, using phylogenetic information for 513 tree species on 26 islands in the subtropical Ryukyu Archipelago, phylogenetic beta diversity between islands was calculated, and effects of historical factors (gaps as surrogate measures of historical barriers) and current ones (distance, area and elevation) on the phylogenetic structure of tree assemblages were examined. The pattern of phylogenetic beta diversity demonstrated that the Tokara Gap and geographical distance were consistently important for characterizing tree assemblages in the Ryukyus relative to other historical and current factors, which suggests that the Tokara Gap and distance‐limited dispersal from the two adjacent source islands have left a deep imprint on the phylogenetic structure of the current tree flora of the islands.

Abstract. The trait-based approach has received much research attention as it provides a heuristic framework for evaluating the ecological impacts of anthropogenic activities on communities and ecosystems. In this study, functional diversity (or structure) measures, such as functional richness, functional evenness, functional divergence, and functional composition, were used to examine management impacts on subtropical forests on the Ryukyu Islands of Japan. Functional indices were compared in tandem with taxonomic diversity indices between three forest types with different management histories: intact old-growth forests, secondary forests after clear-cutting, and abandoned Pinus luchuensis plantations. Species diversity indices were not significantly different among the three forest types. In contrast, functional diversity indices were significantly different among intact forests and managed forests. Functional richness and functional evenness were significantly lower in secondary forests than in intact forests and P. luchuensis plantations. Functional divergence was significantly higher in secondary forests and P. luchuensis plantations than in intact forests. These differences suggest that management activities affected niche space and the patterns of niche differentiation among component species in the functional space of managed forests. Community weighted means for each functional trait were also different among the forest types. The managed forests had greater leaf thickness, leaf dry matter content and maximum height, and lower specific leaf area and leaf nitrogen concentration than intact forests. These differences in functional composition of traits suggested potential functional impacts. This study demonstrated the utilization of species functional traits and community functional structure as a tool of natural experiment for assessing impacts of forest management practices on woodland ecosystems. It was also suggested that logging activities that include large-scale clear-cutting or establishment of P. luchuensis plantations may be incompatible with the conservation of natural ecosystem properties in subtropical forests.

Subtropical forests in the Ryukyu Islands have been degraded by silvicultural practices, and thus their structural attributes are being shifted to other states dominated by a few tree species. This study clarified the mechanisms of the change, and examined the effect of clear logging on the resilience of a subtropical forest. Sprouting regeneration and typhoon disturbance were introduced into an individual-based model, SEIB-DGVM, for describing stand development and succession. The regeneration dynamics from young secondary to oldgrowth stands were reproduced fairly well with the model. Sprouting recruitment produced high stem density at the beginning of stand development, which caused a selfthinning trajectory following the -3/2 power law. In the late development stage after 70 years, tree species diversity fluctuated because of the regenerative response of sprouting species and the facilitatory effect of typhoon disturbance on the coexistence of subordinate species. The death of canopy trees because of typhoon disturbances reduced the dominance of Castanopsis sieboldii, and depressed its dominance in the understory. Consequently, the understory species could establish by virtue of fallen canopy trees, and tree species diversity increased at the stand level. Clear logging experiments in the model revealed that species diversity deteriorated, especially in the stand dominated by sprouting species. Resilience of subtropical forests was determined by initial species composition before clear logging. Our simulation results suggest that repeated logging drives subtropical forests with high species diversity to a stand monopolized by C. sieboldii.

Defining the spatial arrangement and length of the cutting cycle in a logged area is crucial for reconciling potential conflicts between timber yields and maintenance of ecosystem services in natural forests. In this study, we investigated long-term impacts of clear-fell logging on timber production and tree species diversity in a subtropical forest on the Ryukyu Islands, using an individual-based simulation model. We assumed six logging scenarios defined by combinations of forest type and regeneration processes, which acted as surrogates for spatial scales of clear-fell logging. These scenarios were simulated under cutting cycles ranging from 20 to 150 years. Short-cutting cycles resulted in dominance by the sprouting species Castanopsis sieboldii. The compositional shift was accelerated by the lack of seed dispersal from surrounding forest areas. The simulations demonstrated that a sustainable logging regime maintaining both yield and tree species diversity requires a cutting cycle longer than 50 years. The simulation results also suggest that the trade-off between the recovery of tree species diversity and timber production is favored more in stands surrounded by mature forest than in isolated stands or stands surrounded by immature forest. Ecological risk assessments based on model simulations provide an alternative to current forest management practices that rely on empirical knowledge.

Based on the simulation of a spatially explicit individual-based model, this study examined the long-term impacts of understory thinning on stand structure, tree growth and number of tree species in a subtropical forest on the Ryukyu Islands, Japan. The treatment options for understory thinning were defined by combinations of thinning cycle and critical size of thinning trees. The results of the simulation demonstrated that understory thinning had no major influence on the growth of canopy trees, and conversely reduced the stand biomass over time. Moreover, tree species diversity drastically decreased with thinning of understory trees. Frequent thinning of smaller trees prevented the regeneration of understory species. Our predictions suggested that the prescription of understory thinning has a negative impact on the economic efficiency of timber production and the resilience of tree species diversity in a subtropical forest in southern Japan. This is contrary to the expectation that understory thinning, regarded as being analogous to the traditional prescription for coppice woodland, fosters a productive forest with species diversity.