Sources of CO2 efflux from soil and review of partitioning methods

Five main biogenic sources of CO2 efflux from soils have been distinguished and described according to their turnover rates and the mean residence time of carbon. They are root respiration, rhizomicrobial respiration, decomposition of plant residues, the priming effect induced by root exudation or by addition of plant residues, and basal respiration by microbial decomposition of soil organic matter (SOM). These sources can be grouped in several combinations to summarize CO2 efflux from the soil including: root-derived CO2, plant-derived CO2, SOM-derived CO2, rhizosphere respiration, heterotrophic microbial respiration (respiration by heterotrophs), and respiration by autotrophs. These distinctions are important because without separation of SOM-derived CO2 from plant-derived CO2, measurements of total soil respiration have very limited value for evaluation of the soil as a source or sink of atmospheric CO2 and for interpreting the sources of CO2 and the fate of carbon within soils and ecosystems. Additionally, the processes linked to the five sources of CO2 efflux from soil have various responses to environmental variables and consequently to global warming. This review describes the basic principles and assumptions of the following methods which allow SOM-derived and root-derived CO2 efflux to be separated under laboratory and field conditions: root exclusion techniques, shading and clipping, tree girdling, regression, component integration, excised roots and insitu root respiration; continuous and pulse labeling, 13C natural abundance and FACE, and radiocarbon dating and bomb-14C. A short sections cover the separation of the respiration of autotrophs and that of heterotrophs, i.e. the separation of actual root respiration from microbial respiration, as well as methods allowing the amount of CO2 evolved by decomposition of plant residues and by priming effects to be estimated. All these methods have been evaluated according to their inherent disturbance of the ecosystem and C fluxes, and their versatility under various conditions. The shortfalls of existing approaches and the need for further development and standardization of methods are highlighted.