Separating microbial respiration of exudates from root respiration in non-sterile soils: a comparison of four methods

Partitioning the root-derived CO2 efflux from the soil into actual root respiration (RR) and microbial respiration of exudates and root residues is very important for determining the carbon (C) and energy balance of soils. Studies based on artificial root environments like hydroponics or sterile soils give unrealistic figures for C partitioning and are unsuitable for predicting C flows under natural conditions. To date, only four methods have been suggested to separate RR and rhizomicrobial respiration in non-sterile soils: (1) the isotope dilution method, (2) the model rhizodeposition technique, (3) modeling of (CO2)-C-14 efflux dynamics, and (4) the exudate elation procedure. All four methods are based on the pulse labeling of shoots in a (CO2)-C-14 atmosphere and subsequent monitoring of (CO2)-C-14 efflux from the soil. However, the basic assumptions and principles of these methods, as well as the results observed in the original papers, all differ from one another. This study describes the separation of RR of Lolium perenne grown on a loamy Haplic Luvisol from microbial respiration of rhizodeposits by means of all four methods under the same experimental conditions. In spite of alternative principles, the isotope dilution and the (CO2)-C-14 dynamics methods show a similar level of RR: accordingly 39 and 45% of total root-derived CO2 efflux were accounted for by RR. The remainder is rhizomicrobial respiration. The exudate elution method, which underestimates the total rhizodeposition, shows that at least 19% of root-derived CO2 is produced by exudate decomposition. The microbial respiration of rhizodeposits calculated using the model rhizodeposition technique is also underestimated. The exudate elution method is the only procedure allowing physical separation of both C flows. The assumptions and principles of all four methods are reviewed and the effects of possible shortcomings on the separation results are discussed. In conclusion, RR contributes about 40-50% to the root-derived CO2 efflux. The remaining 50-60% comprise the microbial decomposition of root exudates and other rhizodeposits. The longer the period of monitoring the CO2 efflux after the pulse labeling is, the higher the contribution of rhizomicrobial respiration to the total root-derived CO2 efflux from soil. (C) 2002 Elsevier Science Ltd. All rights reserved.