The recent update of key dosimetric data by the International Commission on Radiation Units and Measurements (ICRU) makes several changes to the computation of beam quality correction factors k Q with regard to, for example, the mean excitation energies, I, which enter the stopping power computation for water and air, the computation procedure itself, the average energy expended in the production of an ion pair in air, W/e, as well as chamber-specific factors for cobalt-60. With the new recommendations an accurate assessment of the water-to-air stopping-power ratio, [Formula: see text], in reference conditions is necessary to update the dosimetry protocols for carbon ion beams. The ICRU 90 key data were considered for computation of [Formula: see text] for carbon ion beams using Monte Carlo transport simulations for a number of reference conditions, namely monoenergetic carbon ion beams with a range in water from 3 to 30 cm and spread-out Bragg peaks (SOBPs) of different widths and depths in water. New recommendations for [Formula: see text] are presented, namely 1.1247 for the reference condition of depth 1 g cm-2 for monoenergetic carbon ion beams and 1.1273 at the center of physically optimized SOBPs. The recommendation of a constant value (1.126) represents the stopping-power ratio within a 0.3% variation of [Formula: see text] for all reference conditions considered. The impact of these new [Formula: see text] values and the updated key data on k Q for carbon ion beams was evaluated in a second step. Changes and the difference from experimental data were found to be non-significant, but larger discrepancies to measurements were observed for plane-parallel ionization chambers. The combined uncertainty for k Q in carbon ion beams decreased to 2.4%. In future, it could be further lowered by using chamber-specific Monte Carlo transport simulations, for which the implementation of ICRU 90 key data as done in this study is a prerequisite.