Multivariate analysis of grassland in the three rivers area, natal
D. B. Woods, Eugene Moll
Abstract
D. B. Woods, Eugene Moll
Abstract
Twenty grassland sites in the Three Rivers area, Natal, were sampled for presence of grass species in 20 8 ft. square quadrats placed in a restricted random manner at each site. The data were analysed vising Wisconsin ordination, principal components ordination and normal association analysis. Comparable results were obtained from ordination and association analyses. They reveal that highland grassland differs markedly from lowland grassland, and there also appears to be subdivisions within the latter. The results suggest that a more comprehensive analysis could result in a useful vegetation map with objectively defined boundaries.
OpenAlex reports 2 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Twenty grassland sites in the Three Rivers area, Natal, were sampled for presence of grass species in 20 8 ft. square quadrats placed in a restricted random manner at each site. The data were analysed vising Wisconsin ordination, principal components ordination and normal association analysis. Comparable results were obtained from ordination and association analyses. They reveal that highland grassland differs markedly from lowland grassland, and there also appears to be subdivisions within the latter. The results suggest that a more comprehensive analysis could result in a useful vegetation map with objectively defined boundaries.
Key concepts: Ordination, Grassland, Quadrat, Geography, Vegetation (pathology), Principal component analysis, Multivariate statistics, Correspondence analysis