Tue, Aug 16, 2022: 4:00 PM-4:15 PM
513F
Background/Question/MethodsTropical forests are hyperdiverse, and although broad-scale variation in diversity can be reasonably well explained, local-scale variation in diversity is harder to understand. It has been revealed that species may show species-habitats associations. However, how the assembled species are associated with local habitats and how community-habitat associations vary according to species abundance and tree size class remain unclear. Here, we analyzed the community-habitat associations by applying multivariate regression trees technique on data of 22,064 trees distributed and topography and soil factors within 30 1-ha plots of a semi-deciduous tropical rainforest of Cameroon. The most important species for structuring species assemblages were determined using indicator species analysis.
Results/ConclusionsThe Multivariate Regression Trees (MRT) divide the whole tree community and the tree class’s size except for small stems inventoried in the Doume Communal forest plots into four habitats types surprisingly based on soil parameters such as pH and sodium (Na) concentration. The habitat types for small stems size class were generated based on soil parameters (i.e., CEC concentration and MC) and topography parameters (i.e., A). The three nodes constituting the MRT explained 37.23 % of the total standardized species variance of the whole tree community species inventoried. However, only 17.71 of 37.45% of the total standardized species variance was explained by the second and third nodes, forming the four habitats in the DCF plots. For small, understorey, and large trees classes, the total species variance explained by the three nodes of the MRT were 26.76, 25.68, and 37.45 %, respectively, and CRVEs were 0.925, 0.890, and 0.795, respectively. Indicator species within the large trees class changed to either non-indicator or intermediate indicator species in the understorey trees class. This study highlight that soil variables adding to topographical habitat filtering were important in shaping local species composition.
Results/ConclusionsThe Multivariate Regression Trees (MRT) divide the whole tree community and the tree class’s size except for small stems inventoried in the Doume Communal forest plots into four habitats types surprisingly based on soil parameters such as pH and sodium (Na) concentration. The habitat types for small stems size class were generated based on soil parameters (i.e., CEC concentration and MC) and topography parameters (i.e., A). The three nodes constituting the MRT explained 37.23 % of the total standardized species variance of the whole tree community species inventoried. However, only 17.71 of 37.45% of the total standardized species variance was explained by the second and third nodes, forming the four habitats in the DCF plots. For small, understorey, and large trees classes, the total species variance explained by the three nodes of the MRT were 26.76, 25.68, and 37.45 %, respectively, and CRVEs were 0.925, 0.890, and 0.795, respectively. Indicator species within the large trees class changed to either non-indicator or intermediate indicator species in the understorey trees class. This study highlight that soil variables adding to topographical habitat filtering were important in shaping local species composition.