Effects of selective logging on tree species diversity and composition of Bornean tropical rain forests at different spatial scales

Reduced-impact logging (RIL) is known to be beneficial in biodiversity conservation, but its effects on tree diversity remain unknown. Pattern of tree diversity following disturbance usually varies with spatial scale of sampling (i.e., plot size). We examined the impacts of RIL on species richness and community composition of tree species at different spatial scales, and the scale (plot size) dependency of the two metrics; species richness versus community similarity. One 2-ha and three to four 0.2-ha plots were established in each of primary, RIL, and conventionally logged (CL) forest in Sabah, Malaysia. Species richness (the number of species per unit number of stems) was higher in the RIL than in the CL forest at both scales. The relationship between species richness and logging intensity varied with plot size. Species richness was greater in the RIL than in the primary forest at the 2-ha scale, while it was similar between the two forests at 0.2-ha scale. Similarly, species richness in the CL forest demonstrated a greater value at the 2-ha scale than at the 0.2-ha scale. Greater species richness in the two logged forests at the 2-ha scale is attributable to a greater probability of encountering the species-rich, small patches that are distributed heterogeneously. Community composition of the RIL forest more resembled that of the primary forest than that of the CL forest, regardless of plot size. Accordingly, species richness is a scale-dependent metric, while community similarity is a more robust metric to indicate the response of tree assemblage to anthropogenic disturbance.