Reply on RC1

RC1.1: I was wandering why authors decided to present (and discuss!) data only for specific (HBI triene Z and Brassicasterol) phytoplankton derived HBI indices in the main text while moving others into the supplementary. Do data in supplementary add anything to the study? Are there any key outcomes? If so which ones etc. I think it would be nice to comment on those additional data.It also implies that those outcomes presented in the main text (based on HBI Triene Z and Brassicasterol) have shown most promise (reflect environmental settings best) in previous calibrations and have been applied most extensively, while I`m not convinced that`s the case in Southern Ocean. While they have been utilised fairly extensively in the Arctic and subjected to several calibration studies, It`s not been the case in the Southern Ocean (as author also point out) and applicability of approach utilising any of these pelagic lipids is vastly unexplored.

RC1.1: I was wandering why authors decided to present (and discuss!) data only for specific (HBI triene Z and Brassicasterol) phytoplankton derived HBI indices in the main text while moving others into the supplementary. Do data in supplementary add anything to the study? Are there any key outcomes? If so which ones etc. I think it would be nice to comment on those additional data.It also implies that those outcomes presented in the main text (based on HBI Triene Z and Brassicasterol) have shown most promise (reflect environmental settings best) in previous calibrations and have been applied most extensively, while I`m not convinced that`s the case in Southern Ocean. While they have been utilised fairly extensively in the Arctic and subjected to several calibration studies, It`s not been the case in the Southern Ocean (as author also point out) and applicability of approach utilising any of these pelagic lipids is vastly unexplored.

Author´s response:
The dinosterol and HBI E-triene concentrations and thereof derived PIPSO 25 indices show very similar patterns when compared to the brassicasterol and Ztriene data presented in the main text. In order to avoid repetition while describing these results, we prefer to publish the data as supplement. This also allows other researchers to consider the applicability of dinosterol and/or HBI E-triene as phytoplankton markers for own studies. We now add a sentence in Sect. 4.1, commenting the similarity of the datasets.

RC1.2:
There seems to be weighting towards HBIs, which somewhat detracts from GDGT outcomes. It made me wonder if perhaps "less" could be more. Should authors concentrate either on GDGTs or on HBIs? While they carry out the evaluation between the individual indicies and satelite/modelling data I was missing an intercomarison between lipid derived proxies and where outcomes from one support/contradict those derived from other.
Author´s response: We recognize your point here, which has similarly been mentioned by Referee #2. We have now, based on the suggestion by Referee #2, used a different calibration for subsurface oceans by Kim et al. (2012) and compared those new temperature reconstructions to instrumental and model data. This enabled us to emphasize the GDGT data in the manuscript. In order to prevent repetitions when splitting the data into two papers, we prefer to keep both proxies (HBIs and GDGTs) in the manuscript and publish the data set as one. We note that we do not directly compare IPSO 25 and TEX L 86 as these proxies relate to different environmental variables (i.e. sea ice and subsurface ocean temperature). However, we now comment on the relation between WOA-derived sea surface temperatures and PIPSO 25 values.

RC1.3:
Have authors considered including any taxonomy work? It seems like biomarkers are depicting some regional differences (e.g. EAP vs WS or EAP vs WAP or even WS vs AS) and I was wandering to what extent these could be observed via differences in diatom distributions. Could taxonomy/diatom work also provide some indications about productivity or phytoplankton composition differences that authors refer to in text (e.g. lines 247-249, 282 etc.)?
Author´s response: We agree that taxonomy work would add to a more detailed assessment of the environmental conditions in the different regions and it would be interesting to see, whether diatom distributions follow a similar pattern than the biomarker reconstructions. Taxonomy work, however, is not within the scope of this manuscript but we now address this point in Sect. 7 regarding future work: "Further taxonomy work, the composition of the proxy´s source habitat (basal sea ice, platelet ice, brine channels) and its connection to platelet ice formation via in situ or laboratory measurements are required to better constrain the proxy´s potential for sea ice reconstructions." An important aspect that should be mentioned here as well concerns the preservation of diatoms. Particularly in coastal (often heavily sea ice covered) areas the application of diatoms as environmental proxies can be affected by opal dissolution. This would certainly impact comparisons and/or correlations with other environmental proxies and we suspect that such a study would benefit from a larger data set that also contains sample material from more distal ocean areas (off the continental shelf).