Drivers of soil organic carbon storage and vertical distribution in Eastern Australia
- 1.2k Downloads
Drivers of soil organic carbon (SOC) storage are likely to vary in importance in different regions and at different depths due to local factors influencing SOC dynamics. This paper explores the factors influencing SOC to a depth of 30 cm in eastern Australia.
We used a machine learning approach to identify the key drivers of SOC storage and vertical distribution at 1401 sites from New South Wales, Australia. We then assessed the influence of the identified factors using traditional statistical approaches.
Precipitation was important to and positively associated with SOC content, whereas temperature was important to and negatively associated with SOC vertical distribution. The importance of geology to SOC content increased with increasing soil depth. Land-use was important to both SOC content and its vertical distribution.
We attribute these results to the influence of precipitation on primary production controlling SOC content, and the stronger influence of temperature on microbial activity affecting SOC degradation and vertical distribution. Geology affects SOC retention below the surface. Land-use controls SOC via production, removal and vertical mixing. The factors driving SOC storage are not identical to those driving SOC vertical distribution. Changes to these drivers will have differential effects on SOC storage and depth distribution.
KeywordsMachine learning Vertical distribution Land-use Climate Geology
This project is supported by funding from the Australian Government Department of Agriculture. Thank you to Matt Tighe at the University of New England for pre-review comments and to the reviewers who helped to improve this work.
- Charman PEV, Murphy BW (eds) (2007) Soils their properties and management, 3rd edn. Oxford University Press, South MelbourneGoogle Scholar
- Cotching WE (2012) Carbon stocks in Tasmanian soils. Soil Res 50:83–90Google Scholar
- Isbell R (2002) The Australian soil classification. CSIRO Publishing, MelbourneGoogle Scholar
- Jackson RB, Banner JL, Jobbagy EG, Pockman WT, Wall DH (2002) Ecosystem carbon loss with woody plant invasion of grasslands. Nature 418:623–626Google Scholar
- Jenny H (1941) Factors of soil formation: a system of quantitative pedology. McGraw-Hill, New YorkGoogle Scholar
- R Core Team (2014) R: A language for statistical computing 3.1.1. R Foundation for Statistical Computing, ViennaGoogle Scholar
- Richards G (2001) The FullCAM Carbon Accounting Model: Development, Calibration and Implementation. Carbon accounting and emissions trading related to bioenergy, wood products and carbon sequestration: 1Google Scholar
- Sahukar R, Gallery C, Smart J, Mitchell P (2003) The bioregions of New South Wales. NSW National Parks and Wildlife Service, HurstvilleGoogle Scholar
- Soil and Landscape Grid National Soil Attribute Maps - Soil Depth (3″ resolution) - Release 1. v2. (2014) CSIRO. http://www.clw.csiro.au/aclep/soilandlandscapegrid/index.html
- State of NSW Environmental Protection Authority (2012) New South Wales state of the environment. Office of Environment and HeritageGoogle Scholar
- Statistics ABo (2013) 7121.0—Agricultural Commodities, Australia, 2012–2013Google Scholar