The relationships between land uses, soil management practices, and soil carbon fractions in South Eastern Australia

Abstract

This project aimed to identify land uses and soil management practices that have significant associations with soil organic carbon (SOC) stocks (0-0.3 m) in New South Wales (NSW), Australia. The work presented in this paper is based on a one-off survey targeting key land uses and management practices of eastern NSW. Because of the nature of the work, the land uses and management combinations surveyed in different soils and climatic conditions were significantly unbalanced, and separately analyzing associations after breaking the dataset into different land uses may lead to significant increases in Type errors. Therefore, redundancy analysis (RDA) was undertaken to explore the association between explanatory variables (i.e., land uses, soil management, soil properties and environmental variables) and the variation in stocks (mass per unit area) of particulate organic carbon (POC), humic organic carbon (HOC) and resistant organic carbon (ROC) across 780 sites in eastern NSW, south eastern Australia. Results indicated that soil properties, land uses, soil management and environmental variables together could explain 52% of total variation in stocks of the SOC fractions. Specifically soil properties and environmental variables explained 42.8%, whereas land uses and management practices together explained 9.2% of the total variation in SOC fractions. A forward selection RDA was also undertaken considering soil properties and environmental variables as covariates to assess the statistical significance of land uses and management practices on stocks of POC, HOC and ROC. We found that pasture had significant positive associations on stocks of carbon fractions. Among the soil properties and environmental variables rainfall, longitude and elevation had a significant positive influence while pH and bulk density had a significantly negative influence on the HOC, POC and ROC stocks. Using a novel multivariate technique, the current work identified the land uses and soil management that had significant impact on carbon stocks in soil after accounting for influences soil properties and environmental variables.