How does the nature of living and dead roots affect the residence time of carbon in the root litter continuum?

Root litter transformation is an important determinant of the carbon cycle in grassland ecosystems. Litter quality and rhizosphere activity are species-dependent factors which depend on the attributes of the dead and living roots respectively. These factors were tested, using non-disturbed soil monoliths of Dactylis glomerata L. and Loliam perenne L. monocultures. C-13-labelled root litter from these monoliths was obtained from a first stand of each crop, cultivated under very delta(13)C-depleted atmospheric CO2 (S1). In a factorial design. C-13-labelled root litter of each species was submitted to a second, non C-13-labelled, living stand of each species (S2). Carbon derived from S I and from S2 was measured during an 18-month incubation in the root phytomass and in three particulate organic matter fractions (POM). The decay rate of each particle size fraction was fitted to the experimental data in a mechanistic model of litter transformation, whose outputs were mineralisation and stabilisation of the litter-C. Few differences were found between species, in the amount and biochemical composition of the initial root litter. but Dactylis roots showed a greater C:N ratio, a lower mean root diameter and a greater specific root length compared to Lolium. A transient accumulation of litter residues arose successively in POM fractions of decreasing particle size. The litter-continuum hypothesis was validated, i.e. that the attributes of the compartments (C:N, chemical composition and residence time) depended mainly on their particle size. The S1 species influenced the rate of litter decay while the S2 species controlled the efficiency of litter-C stabilisation versus mineralisation: Dactylis liner decomposed faster and Lolium rhizosphere allowed a greater proportion of titter C stabilisation. Discussions focus on the processes responsible of species strategy in relation with the morphological root traits, and the implication of strategy diversity for rich grassland communities.