CSIRO PUBLISHING www.publish.csiro.au/journals/ajsr Australian Journal of Soil Research, 2006, 44, 233–244 Prediction and digital mapping of soil carbon storage in the Lower Namoi Valley Budiman Minasny A,D , Alex. B. McBratney A , M. L. Mendonc ¸a-Santos B , I. O. A. Odeh A , and Brice Guyon C A Faculty of Agriculture, Food & Natural Resources, The University of Sydney, JRA McMillan Building A05, NSW 2006, Australia. B EMBRAPA-Centro Nacional de Pesquisa de Solos, Rua Jardim Botˆ anico 1024, 22.460-000 Rio de Janeiro-RJ, Brazil. C Ecole Nationale d’Ingenieurs des Travaux Agricoles de Bordeaux, 1 cours du general de Gaulle, B.P. 201, 33175 Gradignan, Cedex, France. D Corresponding author. Email: b.minasny@usyd.edu.au Abstract. Estimation and mapping carbon storage in the soil is currently an important topic; thus, the knowledge of the distribution of carbon content with depth is essential. This paper examines the use of a negative exponential profile depth function to describe the soil carbon data at different depths, and its integral to represent the carbon storage. A novel method is then proposed for mapping the soil carbon storage in the Lower Namoi Valley, NSW. This involves deriving pedotransfer functions to predict soil organic carbon and bulk density, fitting the exponential depth function to the carbon profile data, deriving a neural network model to predict parameters of the exponential function from environmental data, and mapping the organic carbon storage. The exponential depth function is shown to fit the soil carbon data adequately, and the parameters also reflect the influence of soil order. The parameters of the exponential depth function were predicted from land use, radiometric K, and terrain attributes. Using the estimated parameters we map the carbon storage of the area from surface to a depth of 1 m. The organic carbon storage map shows the high influence of land use on the predicted storage. Values of 15–22kg/m 2 were predicted for the forested area and 2–6 kg/m 2 in the cultivated area in the plains. Additional keywords: soil information system, neural networks, carbon stock, carbon sequestration, organic carbon, Vertosol, digital soil mapping. Introduction Estimation and mapping of carbon storage in the soil is currently an important topic; carbon stored in the soil to a given depth has been estimated for the whole world (Batjes 1996), for the continent (Jones et al. 2005), countries (Bellamy et al. 2005; Bernoux et al. 2002; Mikhailova and Post 2006), and regionally (Knowles and Singh 2003). Soil can hold more than twice as much carbon as held in vegetation or the atmosphere (Batjes 1996). The amount of carbon stored in the soil per unit of land area is highly variable depending on the land use, annual input, soil type, and the degradation rate. A global average of 16–20 kg/m 2 is estimated for carbon stored up to 1 m in tropical forests and 11 kg/m 2 for cropping area (Jobb´ agy and Jackson 2000). The distribution of carbon content with depth is essential information for estimating soil carbon storage. Conventionally, carbon storage (also called the carbon stock, carbon pool or carbon density), i.e. carbon mass per unit area for a given depth, is calculated by summing the C density of soil layers 1, 2, ... , N: C I = N j=1 (C m × ρ j ) × thick j (1) where C I is carbon density (kg/m 2 ), C m is carbon content in mass basis (kg/kg), ρ is soil bulk density (kg/m 3 ), and thick is the thickness of the layer (m). Alternatively, a profile depth function can be defined and fitted to the soil carbon data, where carbon content at different depths can be estimated, and the integral of the function represents the carbon storage. This is useful where it is necessary to estimate the carbon storage down to certain depths. Expressing carbon content as a depth function is also advantageous when dealing with soil databases where the © CSIRO 2006 10.1071/SR05136 0004-9573/06/030233