The Southern Ocean is thought to play an important role in the context of global warming and anthropogenic emissions of CO 2 due to its high sensitivity to both climate change and changes in the carbon cycle. Assessing the penetration of anthropogenic CO 2 (C ant) into the Southern Ocean is therefore highly relevant to reduce the uncertainties attached to both the present knowledge of anthropogenic carbon inventories and predictions made by current ocean carbon models. This study compares different databased approaches for estimating the distribution of C ant in the ocean: a recently developed method based on the composite Tracer Combining Oxygen, Inorganic Carbon, and Total Alkalinity (TrOCA) and the ''historical'' back-calculation methods (the so-called DC* and preformed dissolved inorganic carbon methods). Note that the back-calculation technique requires special care when used in the Southern Ocean, where surface oxygen can significantly deviate from equilibrium with the atmosphere. All three methods were applied to data collected at the Indian-Atlantic boundary (WOCE line I6), where significant transient tracer concentrations were observed in deep and bottom waters. North of 50°S, distribution and inventories of C ant are coherent with previous data-based and model estimates, but we found larger storage of C ant south of 50°S as compared to the midlatitude region. In that, our results disagree with most previous estimates and suggest that the global inventory of anthropogenic CO 2 in the Southern Ocean could be much larger than what is currently believed.