Short and medium term plant litter decomposition in a tropical Ultisol elucidated by physical fractionation in a dual 13 C and 14 C isotope study J. Magid a, * , G. Cadisch b , K.E. Giller c a Soil, Water and Plant Nutrition, Department of Agricultural Sciences, Royal Veterinary and Agricultural University, Thorvaldsensvej 40, Copenhagen DK 1871, Frederiksberg, C, Denmark b Department of Biological Sciences, Wye College, University of London, Wye, Ashford, Kent TN25 5AH, UK c Department of Plant Sciences, Wageningen University, P.O. Box 430, 6700 AK Wageningen, The Netherlands Received 11 June 2001; received in revised form 14 March 2002; accepted 25 March 2002 Abstract Size – density fractionation in combination with the use of 13 C analysis yielded detailed information on soil organic matter (SOM) dynamics and stabilisation in a Brachiaria humidicola pasture (C4), established after rain-forest (C3) by slash and burn 16 year prior to soil sampling. The medium term 13 C marker was supplemented with a short-term 14 C marker by incubating the soil over 75 days with homogeneously-labelled Lolium perenne. The use of the medium-term marker ( 13 C) showed a potential for density fractionation that would not have become apparent using the short-term marker ( 14 C) on its own. To a large extent this disparity has to do with limiting characteristics of the used 14 C-labelled material (small particle size and high content of water soluble carbon). The 14 C-labelled material was immediately present in all fractions but mainly corroborated the results from the 13 C determination. Both 14 C and 13 C measurements indicated the transfer of carbon from large light fractions to fine heavy organomineral fractions. The ‘youngest’ C4 fractions floated on water, whereas the ‘intermediately aged’ C4/C3 fractions were intimately associated with the mineral phase through biological processes, and came out as heavy. The ‘oldest’ C3 fraction consisted of stabilised free organic material that had not attained high mineral affinity, and therefore had intermediate density. The changes in 14 C specific activity of the isolated fractions in the short term indicated that these fractions were not homogeneous. This suggests that there will be limited scope for tying the size – density fractions and the conceptual homogenous pools used in SOM models together without further characterisation of these pools or by using alternative decay models. q 2002 Published by Elsevier Science Ltd. Keywords: Size and density fractionation; Charcoal; Soil organic matter pools; Decomposition; 13 C; 14 C 1. Introduction Our understanding of and ability to predict changes in the dynamics of organic matter decomposition in soil is mainly based on a single exponential decay function (Jenny et al., 1949). The concept of organic matter consisting of one or more pools, each decomposing at their specific rate, k, forms the basis of virtually all simulation models of decomposition in soil. This is reflected in a number of computer simulation models, that describe decomposition of organic materials as the first order kinetic decay of several discrete pools, (Hansen et al., 1990; Jenkinson et al., 1987; Parton et al., 1988). Recent research on soil organic matter (SOM) dynamics has been directed towards finding measures of biologically meaningful pools and fluxes. The need for such measure- ments was identified by Jansson and Persson (1968) who highlighted the gap between the biological concepts and the application of chemical fractionation methods to SOM studies. A large body of literature has developed in which physical methods based on particle size (and to a lesser extent density) have been used to isolate fractions prior to chemical characterisation. Christensen (1992) stated that although the concept of non-complexed organic matter as a discrete pool of relevance to SOM turnover is attractive theoretically, the potential of density fractionation is less clear than that of size separation. In earlier work (Magid et al., 1996) we examined different size –density fractionation approaches for tracing the fate of homogeneously 14 C-labelled plant materials. Fractionation by size and density was shown to be a more powerful approach for separating SOM fractions than fractionation based on density alone. In a subsequent 0038-0717/02/$ - see front matter q 2002 Published by Elsevier Science Ltd. PII: S0038-0717(02)00069-X Soil Biology & Biochemistry 34 (2002) 1273–1281 www.elsevier.com/locate/soilbio * Corresponding author. Tel.: þ 45-3528-3491; fax: þ 45-3528-3098. E-mail address: jma@kvl.dk (J. Magid).