The turnover time of terrestrial carbon was estimated using a multiobjective parameterization method that combined data sets of plant production, biomass, litter and soil-C observations in the calibration of a C-cycle model for the Australian continent (VAST1.1; Vegetation and Soil carbon Transfer). The method employed a genetic algorithm to minimize model-data deviations and maximize consistency between estimated model parameters and all available data. Based on the parameterization, the turnover time of biosphere C for Australia was estimated to be 78 years which is longer than global C-turnover estimates (of 26--60 years) due entirely to slower turnover of C in the upper 20 cm of soil. Turnover times of litter and deeper soil-C were similar to global values. By splitting total C in the upper 20 cm between labile and nonlabile fractions (based on published data) the turnover time of the labile pool was at least 44 years which is still longer than global estimates (9--25 years). Longer C-turnover in Australian surface soils was attributed to (1) limited soil moisture slowing decomposition more than net primary production, (2) frequent fires leading to a large fraction of nonlabile charcoal C in soil, and (3) strong adsorbing capacity for organic-C in these highly weathered soils. It was found that >89% of the C flux to the atmosphere from decomposition of organic matter originated from fine litter, coarse woody debris and the upper 20 cm of soil in all biomes.