Theme 1 Using near-infra-red spectroscopy (NIR) for characterisation of plant residue quality - A new approach for prediction of decomposition and nitrogen release in agricultural soils

Introduction Return of organic materials such as crop residues, catch crops and green manures has a bearing influence on carbon (C) and nitrogen (N) turnover in agricultural soils. In recent works, Mueller et al. (1998), Henriksen & Breland (1999), and Trinsoutrot et al. (2000) found that results of stepwise chemical digestion (Goering and Van Soest 1970) constituted an adequate basis for describing à priori the degradability of both C and N in crop residues within a wide range of qualities. This may be used for organic input parameterisation in many simulation models widely used to predict the turnover of N from plant materials returned to the soil (Breland 1997; Mueller et al. 1998; Henriksen & Breland 1999; Nicolardot et al. 2001). However, the stepwise chemical digestion and the associated C and N analyses are time-consuming, costly and will hardly be used routinely. More timeand cost-effective methods should therefore be sought. Near-infrared reflectance spectroscopy (NIR) is a quick and cost-effective technique, which has been successfully used for various quality characterisations, e.g., organic matter quality in agricultural soils (Börjesson et al. 1999), decomposition stage of forest litter (Gillon et al. 1993), and digestibility of animal feeds (Coleman & Murray 1993). NIR has also proven to predict the results of stepwise chemical digestion with great precision (McLellan et al. 1991; Joffre et al. 1992; Bruno-Soares et al. 1998). Consequently, NIR may hold great promise as a criterion for distributing plant C and N into fractions of predictive value for residue degradation. Hence, a joint, collaborative project was initiated in year 2000 between research groups in all the Scandinavian countries, with the major objective of developing NIR calibrations as a cost-effective and reliable tool for characterising degradability of crop residues in a model of C and N turnover in soil. The first partial objective was to make NIR calibrations with data on C and N fractionation according to the Van Soest analysis for a broad range of materials. Results from this first objective will be presented at the workshop. Another partial objective was to test the NIR and decomposition models against independent data sets (validation). Some very promising, preliminary data obtained from independent French and Norwegian samples are presented in this abstract. Further information about the project can be found at www.agsci.kvl.dk/~lastjej5/project/nkj/