Don’t miss the forest for the trees! Evidence for vertical differences in the response of plant diversity to disturbance in a tropical rain forest

Ecological studies in tropical rain forests traditionally focus on trees above a threshold diameter at breast height (dbh), since ignoring plant species of the other structural compartments is believed to be an acceptable tradeoff between exhaustiveness and effectiveness. However, the consequences of missing species below a threshold dbh value have been largely neglected so far. We evaluated whether the response of species diversity of ≥10-cm dbh trees was similar to the response of other structural ensembles (namely treelets, saplings, and terricolous herbs) in a lowland tropical rain forest, to three disturbance regimes: natural gap dynamics (control), and selective logging with and without additional thinning. We studied forest vegetation composition and diversity in a 20-yr replicated field experiment comprising nine 1 ha permanent plots established in a semi-deciduous rain forest of the Congo Basin and equally distributed among the three treatments. Once corrected by stem density, species richness was similar between logged (20 years since logging) and untouched old-growth forest stands with respect of trees, but higher with respect of treelets. As disturbance intensity increased, species richness increased within sapling layers but decreased within herb layers, while species spatial turnover (beta diversity) increased in both cases. Regarding the parameters of the partitioned rarefaction curves and relative abundance distribution curves, no correlation was found between trees and any of the other structural compartments. Whilst tree and treelet species composition was similar among treatments, the understories still reflected past disturbance intensity, with a strong response of the sapling and herb layers. These results show that ecological studies based solely on tree layers (dbh ≥ 10 cm) are misleading because their response to disturbance cannot be used as a surrogate for the response of other structural ensembles. Long-lasting effects of anthropogenic disturbance on the sapling bank and the herb layer may durably influence the long-term forest dynamics. Since overstory but not understory plant communities have recovered from human disturbances 20 years after silvicultural operations, African tropical rain forest ecosystems may not be as resilient to selective logging as previously thought.