Use of Digital Terrain Analysis and Classification Trees for Predictive Mapping of Soil Organic Carbon in Southern Denmark

R. Bou Kheir, M. H. Greve, P. K. Bocher, M. B. Greve
Department of Agroecology and Environment, Faculty of Agricultural Sciences (DJF), Aarhus University, Blichers Allé 20,
P.O. Box 50, DK-8830 Tjele, Denmark

Soil organic carbon (SOC) is a dynamic component of the terrestrial system, with both internal changes in the vertical and horizontal directions and external changes with the atmosphere and the biosphere. Changes in SOC are attributed to both natural processes and human activities, and reflect the balance between decomposition of organic matter and input from roots and litter (Turner and Lambert 2000). In recent years, the importance of human activities has been widely recognized. Land use changes, including deforestation, biomass burning, draining of wetlands, ploughing, use of fertilisers and other agricultural practices, are regarded as the main factors causing loss of SOC and the emission of CO2 into the atmosphere. These changes can be significant in grassland and cropland (Conant and Paustian 2002, Schuman et al. 2002) where intensive agricultural activities are carried out.

As part of international efforts to stabilize atmospheric greenhouse gas concentrations, Denmark is committed to establish inventories of the C stock in the frame of the Kyoto protocol. In this context, our study focuses on building a simple, realistic, practical and informative classification-tree model to predict the distribution of spatial patterns and changes in SOC across a study area in southern Denmark from mapped environmental variables.

AttachmentSizeHitsLast download
kheir2009geomorphometry.pdf696.34 KB15422 days 11 hours ago
Your rating: None Average: 2 (4 votes)