Single-species plantations of exotic trees, biological deserts or islands of plant diver- sity? Contrasting biotic and abiotic situa- tions in northern Kivu (Democratic Repu- blic of the Congo)

While forest plantations are a topic attracting growing interest in Central Africa, their undergrowth has been little studied. In order to assess the potential contribution of plantations of exotic species to the restoration of plant cover on anthropised land in Kivu (Democratic Republic of the Congo), the plant diversity of their understorey was characterised and compared in a selection of single-species plantations of Eucalyptus saligna and Grevillea robusta on two sites (Sake and Kirumba) with different altitudinal and pedoclimatic conditions. The variations in plant composition found by the surveys was then analysed according to certain ecological parameters. Finally, indicator species of the altitudinal and pedoclimatic conditions at each site were identified. The data collected in 2018 and 2020 in 12 plantations, using the minimum area method, showed a high degree of heterogeneity in the plant composition of the understorey. Although herbaceous species predominated, some woody and liana species were also found in these understoreys. The species richness did not differ significantly between sites or between the tree species planted. Species richness varied from one survey to another depending on latitude, soil pH and bioavailable aluminium and magnesium concentrations. Thirteen indicator species for the altitudinal and pedoclimatic conditions of the sites were identified altogether, nine for Sake and four for Kirumba. Repetition of the surveys increased the sampling effort even though the minimum area did not change significantly over two years. This study highlighted the contribution of exotic species plantations to the recovery of plant cover on potentially degraded sites in Kivu and suggested management measures suited to the physico-chemical properties of the soils. The use of indicator species of pedoclimatic conditions as predictors of site productivity deserves to be tested in the plantations studied, in order to verify whether their presence could be used as a tool for selecting sites for reforestation.