Soil organic carbon and floristic composition relationships: A simple model using principle factor analysis and regression analysis

Abstract

Future carbon storage in tropical forests will be influenced strongly by species composition. Huge knowledge gap has arisen on how floristic richness correlates or contributes to soil carbon storage. This study examines correlations among floristic parameters, soil carbon and relevant environmental variables. We collected soil data (carbon fractions, total C, total N, soil pH, and fine root biomass) from at the Sigiriya forest sanctuary, and Popham Arboretum in the dry zone and Udawattakelle forest sanctuary in the wet zone. The data were pooled and biomass/carbon stocks (e.g. above-ground biomass stock, above-stock, and below-ground C stocks) were calculated using standard equations. The correlation analysis between biomass/C stock and soil parameters was performed using principle factor analysis (PFA) and multiple regression of SAS version 6.12. PFA resulted in five clusters. One cluster was formed by soil C, soil N, and Mackintosh Distance (U) with high correlation (R2 = 082). Multiple regression model showed that there is a positive relationship between the biomass stocks and species richness, while Mackintosh evenness has a negative relationship with total carbon of the soils. Basal area of the forests was recognized as a neutral parameter, because it acts as both positively and negatively. These results indicated that the above floristic composition parameters acted as determinants of total soil carbon stock. Thus, changing floristic composition to maximize storage of soil carbon will provide opportunity to enhance soil carbon sequestration.