RESPIRATION FROM SOIL AND LITTER IN A SOWN PERENNIAL GRASS PASTURE

The severe nitrogen (N) deficiency which occurs in many sown grass pas tures in Queensland is believed to be exacerbated by large and continu ous inputs of carbon (C) from decomposing plant residues. In this stud y we attempted to quantify the importance of surface litter, roots and soil organic matter as sources of respiration in an established green panic (Panicum maximum var. trichoglume) pasture in south-east and Qu eensland. Intact pasture cores were taken from the field and the surfa ce litter was removed before applying the following treatments: (i) su rface litter added, (ii) roots pruned to kill approximately 60% of roo ts but not kill the plant, and (iii) plant shoots removed. Cores from bare soil between green panic plants were also included. The cores wer e kept in a glasshouse and CO2 evolution measured continuously for 117 days using an alkali absorption method. Respiration from the various components of the system was estimated. Evolution of CO2 from the core s was increased by litter addition and decreased by shoot removal. Roo t pruning stimulated CO2 evolution in litter-removed treatments but ha d no effect in litter-added treatments. Root respiration and microbial respiration of root-derived C accounted for an average of 53% of the total evolved CO2. Surface litter, soil organic matter and dead roots accounted for an average of 40%, 4% and 3% respectively. The importanc e of a particular C source to microorganisms varied depending on the a vailability of other C sources. Cores were destructively sampled on fi ve occasions and the soils incubated at 25 degrees C for 10 days to me asure CO2 evolution with surface litter and roots removed. Evolution o f CO2 in incubated soils was increased by litter and, to a lesser exte nt, by live roots, demonstrating that some of the labile C from these plant components was soluble or well incorporated into the soil.