Ecological Studies in Four Contrasting Lowland Rain Forests in Gunung Mulu National Park, Sarawak: IV. Associations Between Tree Distribution and Soil Factors

Publisher Website
Google Scholar
Abstract
The position, dbh [diameter breast height] and species of trees (.gtoreq. 10 cm dbh) were recorded for 4 tropical rain forest sites in Gunung Mulu National Park, Sarawak. The sites, each 1 ha in area and divided into twenty-five 20 .times. 20 m plots, were: alluvial forest (AF); dipterocarp forest (DF); heath forest (HF); forest over limestone (LF). Three classification techniques were used to subdivide each site into 2 or 3 vegetation classes: polythetic agglomerative classification of 10 .times. 10 m subplots using presence or absence of species; monothetic divisive classification of trees based on the presence or absence of species among their nearest neighbors; and polythetic divisive classification of 10 .times. 10 m subplots using basal areas of species. The AF, DF and HF showed clear within-site floristic heterogeneity. One soil sample was collected at 0-10 cm and 10-30 cm depths at a mapped random location within each plot. The samples were analyzed for pH, total organic C, total N and P, exchangeable K, Na, Ca and Mg and cation exchange capacity. Principal components analysis reduced these variables to 2 components (approximately pH and Ca; and organic C and cation exchange capacity) for each site. Vegetation classes from the 3 methods of classification were related to the component scores of the soil principal components analysis for each soil sample location for the 2 depths separately. The vegetation classes on the AF were significantly associated with changes in the pH and calcium component, and classes on the HF were significantly associated with changes in the organic carbon and cation exchange capacity component. No associations between soils and vegetation were found for the DF and LF.