3846 R. Steinfeldt et al.: Anthropogenic carbon in the Atlantic Figure 3. (a–b) Zonal mean sections showing the 1/0 ratio inferred from simultaneous observations of different tracers. (c–d) Difference in zonal mean Cant concentrations calculated from a variable 1/0 ratio and from the constant ratio of 1/0 = 1. The Cant fields are based on tracer data from the whole period (1982–2021), and the reference year is 2020. (e–f) Zonal mean sections of 1tCanomant for a variable 1/0 ratio (Cant calculated for 2020 with data in 2020 minus Cant calculated for 2020 based on tracer data in 1990, i.e., Cant2020 ? Cant1990?2020). Contour lines are shown as in Fig. 2. For details, see the text. Note the uneven breaks in the color scale. waters, a pronounced Cant anomaly should only occur for a pronounced longtime change in the ocean circulation and/or ventilation. But even then, the Cant accumulation anomaly should be smaller than in the regions with high Cant concen- trations, like the water mass formation regions. Older waters contain a notable fraction with ages over 200 years (see the example in Table D1), i.e., Cant free waters, which cannot contribute to the Cant anomaly. We thus consider the strongly negative 1tCanomant values in LDW as an artifact of the TTD parameterization in the form of a single inverse Gaussian function. In the next section we show how a modification of the TTD parameterization by including an additional dilu- tion of young waters with old waters helps to overcome this artifact. 2.3.2 The modified TTD method with dilution Steinfeldt and Rhein (2004) presented the foundation of the TTD method applied here by focusing on the Deep Western Biogeosciences, 21, 3839–3867, 2024 https://doi.org/10.5194/bg-21-3839-2024