After publication of the article, the authors noticed a mistake in the concentration calculations of the two biomarker lipids IPSO<inf>25</inf> and dinosterol. The volume of the internal standards 7-hexylnonadecane and 5α-androstan-3-ol had incorrectly been calculated twice. Hence, the absolute concentrations of IPSO<inf>25</inf> are erroneously higher by factor 20 and the concentrations of dinosterol are erroneously higher by factor 60. The calculated P<inf>D</inf>IPSO<inf>25</inf> index is not affected by these changes. In 4.1.2. Unit B: Dynamic ice front The value for IPSO<inf>25</inf> was reported as 4–8 μg∗g OC⁻¹, but should now be corrected to 0.2–0.4 μg∗g OC⁻¹. The value for dinosterol was reported as 200–400 μg∗g OC⁻¹, but should now be corrected to ca 2–7 μg∗g OC⁻¹. In 4.1.3. Unit C: Reduced sea-ice cover followed by sea-ice re-expansion In subunit C1, the value for dinosterol was reported as 180 μg∗g OC⁻¹, but should be corrected to 3 μg∗g OC⁻¹. In subunit C2, the value for IPSO<inf>25</inf> was reported as 6–16 μg∗g OC⁻¹, but should be corrected to 0.3–0.8 μg∗g OC⁻¹. In subunit C2, the value for dinosterol was reported as 300–600 μg∗g OC⁻¹, but should be corrected to 5–10 μg∗g OC⁻¹. In 4.1.4. Unit D: Seasonal sea-ice cover environment In subunit D1, the value for IPSO<inf>25</inf> was reported as ca 1 μg∗g OC⁻¹, but should be corrected to ca 0.05 μg∗g OC⁻¹. In subunit D1, the value for dinosterol was reported as 100–400 μg∗g OC⁻¹, but should be corrected to ca 1–7 μg∗g OC⁻¹. In subunit D2, the value for IPSO<inf>25</inf> was reported as 0–12 μg∗g OC⁻¹, but should be corrected to 0–0.2 μg∗g OC⁻¹. In subunit D2, the value for dinosterol was reported as 200–500 μg∗g OC⁻¹, but should be corrected to ca 3–8 μg∗g OC⁻¹. Please note that the values of the IPSO<inf>25</inf> and dinosterol axes in Fig. 2. and the corresponding supplementary figure plotted versus age (printed below) are now corrected. Datasets uploaded to PANGAEA Data Publisher for Earth & Environmental Science have also been corrected (https://doi.org/10.1594/PANGAEA.904263). The changes have no impact on the scientific results and reasoning presented in the paper. [Figure presented] Fig. 2. Contents of P<inf>D</inf>IPSO<inf>25</inf> (a), dinosterol (b), IPSO<inf>25</inf> (c) and TOC (d) in sediment core PS69/274-1. AMS ¹⁴C age constraints in calib. a before present (BP) in dark grey; unreliable age given in italics (Hillenbrand et al., 2010). Interval highlighted by dashed grey line marks the Antarctic Cold Reversal (ACR; Jouzel et al., 1995). Biomarker concentrations and calculated P<inf>D</inf>IPSO<inf>25</inf> value of the surface sample (box core PS69/275-2) indicated by dots with black circle and label in respective color. Triangles in P<inf>D</inf>IPSO<inf>25</inf>-curve: thick ice cover, maximum value of 1 assigned to these samples. Core is divided into four units as indicated in the lowermost and topmost bar: Unit A: floating ice canopy (blue shading), Unit B: dynamic ice front, Unit C: reduced sea ice/sea-ice re-expansion (red shading), Unit D: seasonal sea ice. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)