Retention forestry in plantations: Synthesizing key findings of early studies from a long-term experiment in northern Japan

Retention forestry, the practice of intentionally leaving important forest structures at harvest time, is increasingly being implemented globally as an alternative to clearcut harvesting. However, there is little empirical information available for applying retention forestry in plantations, especially in Asia. Therefore, we launched a large-scale experiment called the Retention Experiment for Plantation Forestry in Sorachi, Hokkaido (REFRESH) in 2013 in native fir plantations in Hokkaido, northern Japan to establish a forest management system that balances the ecological, social, and economic values of plantations. The experimental design consists of six treatments (clearcutting, three levels of dispersed retention 10, 50, and 100 trees/ha, aggregated retention a 0.36-ha retained patch, and gap cutting) and two controls (unharvested plantations and natural forests) with three replicates each. In dispersed retention, naturally regenerated broad-leaved trees were retained to restore structures and elements of original natural forests. This paper provides a comprehensive review of the short-term (pre-harvest to six-year post-harvest) results of ecosystem services (harvesting cost, and public perception) and biodiversity (birds, understory plants, beetles, bats, and ectomycorrhizal fungi). In dispersed retention, the level of retention was positively associated with the beneficial effects of public perception, and diversity of forest species in all taxa except understory plants. The responses were mostly monotonic for the diversity of each taxon, suggesting that we cannot find clear threshold values for the optimal amount of retained trees. In contrast, harvesting costs increased slightly when more than 50 trees/ha (9% in volume) were retained. In aggregated retention, small aggregates maintained undisturbed forest floor and thus served as refugia for forest species of understory plants, carabid beetles, and ectomycorrhizal fungi but not for forest species of birds, and necrophagous silphid and dung beetles. Our results revealed that because dispersed and aggregated retention have different advantages, combinations of aggregated and dispersed retention are important. For dispersed retention, although retention of fewer than 50 trees/ha can minimize economic loss, higher retention levels are recommended to mitigate the negative impacts of harvesting in ecosystem services and biodiversity. In conclusion, short-term results from the REFRESH project demonstrate that retention forestry is a promising way to reconcile biodiversity conservation and multiple ecosystem services into productive plantations.