Recent studies have shown that patterns of plant species richness and community biomass are best understood in a multivariate context. The objective of this study was to develop and evaluate a multivariate hypothesis about how herbaceous biomass and richness relate to gradients in soil conditions and woody plant cover in blackland prairies. Structural equation modeling was used to investigate how soil characteristics and shade by scattered Juniperus virginiana trees relate to standing biomass and species richness in 99 0.25 m2 quadrats collected in eastern Mississippi, USA. Analysis proceeded in two stages. In the first stage, we evaluated the hypothesis that correlations among soil parameters could be represented by two underlying (latent) soil factors, mineral content and organic content. In the second stage, we evaluated the hypothesis that richness and biomass were related to (1) soil properties, (2) tree canopy extent, and (3) each other (i.e. reciprocal effects between richness and biomass). With some modification to the details of the original model, it was found that soil properties could be represented as two latent variables. In the overall model, 51% and 53% of the observed variation in richness and biomass were explained. The order of importance for variables explaining variations in richness was (1) soil organic content, (2) soil mineral content, (3) community biomass, and (4) tree canopy extent. The order of importance for variables explaining biomass was (1) tree canopy and (2) soil organic content, with neither soil mineral content nor species richness explaining significant variation in biomass. Based on these findings, we conclude that variations in richness are uniquely related to both variations in soil conditions and variations in herbaceous biomass. We further conclude that there is no evidence in these data for effects of species richness on biomass.
Additional Publication Details
Multivariate control of plant species richness and community biomass in blackland prairie