We examined soil organic carbon (SOC) concentration (mg g⁻¹) and total organic carbon (TOC) stock (Mg ha-1 to 30 cm soil depth) in three pasture systems in northern New South Wales: improved pasture, native pasture, and lightly wooded pasture, at two sampling times (2009 and 2011). No significant difference was found in SOC or TOC between sample times, suggesting that under the conditions we examined, neither 2 years nor an intervening significant rainfall event was sufficient to change the quantity or our capacity to detect SOC, and neither represented a barrier to soil carbon accounting. Low fertility, lightly wooded pastures had a slightly but significantly lower SOC concentration, particularly in the surface soil layers. However, no significant differences in TOC were detected between the three pasture systems studied, and from a carbon estimation perspective, they represent one, single dataset. A wide range in TOC values existed within the dataset that could not be explained by environmental factors. The TOC was weakly but significantly correlated with soil nitrogen and phosphorus, but a more significant pattern seemed to be the association of high TOC with proportionally larger subsoil (0.1-0.3 m) organic carbon storage. This we attribute to historical, long-term rather than contemporary management. Of the SOC fractions, particulate organic carbon (POC) dominated in the surface layers but diminished with depth, whereas the proportion of humic carbon (HUM) and resistant organic carbon (ROC) increased with soil depth. The POC did not differ between the pasture systems but native pasture had larger quantities of HUM and ROC, particularly in the surface soil layers, suggesting that this pasture system tends to accumulate organic carbon in more resistant forms, presumably because of litter input quality and historical management.