This is a well-written manuscript that presents analyses of biomarkers and stable isotopes of specific compounds. The analyses are state of the art and have not previously been used in the south-western Baltic. It is important to use new methods to improve our understanding of past environmental changes. However, the interpretations of the data are not straightforward and after reading the manuscript I did not feel that I learned much new about the history of the Baltic Sea. I am sorry but I don’t feel that the study “provided new insights into changes … in the vegetation in the western Baltic region throughout the Holocene” (line 210).

We thank the reviewer for their praise and constructive comments on our manuscript. We will restructure the manuscript to focus more on the hydrological shifts noted in the isotope data and how our results confirm diversification of vegetation in the region after ~10.5 ka.

The interpretations are hampered by the fact that the geography of the Arkona Basin changed over time. During the Yoldia Sea stage it was a bay of the Baltic Basin, which was connected to the Kattegat via straits in south-central Sweden. During the early part of the Ancylus Lake stage the outflow shifted from south-central Sweden to what became the Danish straits. During the Littorina Sea stage it was part of the Baltic Sea, with outflowing and inflowing water masses. Water level varied a lot over time due to glacioisostatic rebound and global eustatic sea level rise.

We agree that the geography of the basin changed a lot during the Holocene. Our main interest lies in the terrestrial catchment and linking plant biomarker isotopes to changes in vegetation and regional hydrology. The main impact of the geographical changes occurring in the basin will be on the distance from the continent to the core site (i.e., decreasing with decreased sea level). In addition, increased sea level can cause enhanced coastal erosion, mobilizing soil and plant matter. We will add more discussion on the impact of sea-level changes in the revised manuscript.

Another issue concerns erosion, reworking and redeposition. Erosion is particularly important at the transition from the Ancylus Lake to the Littorina Sea. Could it be that reworking explains the low δ¹³C values of bulk sediment samples from the Transgression phase?

During the transgression, there is a shift from low δ¹³C_bulk (characteristic of C₃ plants and freshwater algae, Fig. 2) to higher values typical for marine environments. It is possible that, at the beginning of the transgression, enhanced erosion of coastal soils explains the sharp peak in TOC (Fig. S1) although it is not reflected in the δ¹³C_bulk.

It is mentioned that some of the molecules analyzed in the study can be wind transported (line 69). I wonder if the molecules in the sediment formed in the Arkona Basin and the surrounding land area? Or could it be that the molecules come from the Kattegat and were transported by inflowing water to the Arkona Basin? Or did they come from northern parts of the Baltic Sea or from the whole drainage area of the Baltic Sea?

Long-chain alkanes were likely produced by vegetation in the whole drainage area of the Baltic Sea. The short-chain alkanes could have been produced in the Arkona basin or transported into the basin from surrounding lakes and rivers.

The authors discuss the results “by Baltic Sea phase”(s) (line 212). The phases are Yoldia Sea, Ancylus Lake and Littorina Sea. However, the discussion is divided into: Yoldia Sea, Ancylus Lake, Ancylus Regression, Ancylus Lake vegetation and hydrological change, Marine transgression and finally Littorina Sea and Modern Baltic phases. I suggest that the authors use only three headings. What they term Marine Transgression is part of the Littorina Sea stage.

We will change the division of the discussion to follow this suggestion.

Modern palaeostudies relies on well-dated records and high temporal resolution. However, I get the feeling that the chronology of the studied record is poorly constrained. The age-depth model needs to be described in detail in the paper. It is mentioned that the age-depth model is described in detail in Weiss et al. (2020) and that is was created by combining ¹⁴C-ages of mollusk shells and … (line 99). According to Weiss et al. (2020) three radiocarbon ages were obtained. I could find no information on the ages or which species was used for dating in the 2020 paper or in the supplementary material. Also, I saw no information about which calibration curve was used. It should also be explained in the paper that only calibrated ages are discussed. In particular, I wonder how the older non-marine part of the core was dated.

The core was dated by correlating XRF data to two nearby cores described in Warden et al. (2016). The ¹⁴C ages were corrected using local marine reservoir values from Lougheed et al. (2013; doi:10.5194/cp-9-1015-2013). We will add a more detailed discussion of this information to the revised manuscript.

With respect to resolution, I note that four samples were analysed for the time period from 6 ka to the present. The resolution is higher in the older part of the record.

The focus of our study is on the Ancylus Lake, hence the difference in sampling resolution. We will make this more explicit in the revised version of the manuscript.

The paper must also provide information about the sediments in the core, based on a visual core description.

We will add a supplementary figure with the core description.

One of the main scientific questions in the development of the Arkona Basin and the Baltic Basin is the dating of the transition from the Ancylus Lake to the Littorina Sea. Andrén et al . (2000) dated it to ca. 10.1 cal. ka BP and Berglund et al. (2005) dated it to ca. 9.8 cal. ka BP, whereas other studies have dated it to 7-8 cal. ka BP. I am surprised to see that this question is not mentioned in the manuscript. Did the study provide any information on this issue?

Our results suggest a later transition from the Ancylus Lake into the Littorina Sea. We will add more detail about this in the revised manuscript. 

I think one of the most interesting outcomes of the study it that the Ancylus Lake stage was highly dynamic with large isotopic shifts. But the interpretation is difficult. It was partly caused by decreasing influence of melt water. But I suggest that the shift in drainage may also play a role. I am not sure about the importance of the short-lived cold events during this time.

We agree with the reviewer that the Ancylus Lake phase was dynamic. We will add discussion about the shift in drainage as another possible contributing factor to the complexity of the phase in the revised version. However, we note that such depleted hydrogen isotope values of lipids followed by a large, positive shift is difficult to explain by a change in drainage alone.

Other comments

Line 15. According to the first sentence in the abstract: “The Baltic Sea experienced a number of marine transgressions and regressions throughout the Holocene”. However, only one marine transgression is discussed in the manuscript. The northern part of the Baltic Sea experienced regression throughout the Holocene, whereas the southern part experienced one marine transgression during the Holocene.

We will revise this statement to: “The Baltic Sea experienced changes in marine input throughout the Holocene.”

Line 16. According to the next sentence: “These fluctuations in sea level coupled with substantial regional ice melt led to isostatic adjustment and periodic isolation from the North Sea”. Isostatic adjustment was caused by ice retreat, not by the “fluctuations in sea level”.

Thanks for catching this. We will revise it to: “The substantial regional ice melt led to isostatic adjustment and periodic isolation from the North Sea.”

Line 28-29. “that promote … diverse phytoplankton communities”. A reference is needed. Is the diversity larger than in other areas?

We will add the following references to the end of this sentence: e.g., Wasmund and Uhlig (2003) doi: 10.1016/S1054-3139(02)00280-1; Golubkov et al. (2020), doi: 10.1016/j.oceano.2020.11.002.

Line 31, 33. Retreat was caused by melting and calving.

We will add “calving” to the reason for the retreat of the SIS.

Line 33. “the Scandinavian Ice Sheet covered large swaths of Europe”. Change to: the Scandinavian Ice Sheet covered large parts of northern Europe.

We will make this change in the revised manuscript.

Line 35. The Yoldia Sea stage began in the earliest part of the Holocene. Therefore it was not caused by global and regional temperatures that continued to rise during the Holocene.

We will remove “during the Holocene” in the revised version. 

Line 38. Sea level fluctuations. It should be water level fluctuations, because lake stages are also involved.

We will correct accordingly in the revised manuscript.

Line 38. The Ancylus Lake was a freshwater lake ­- I find it a bit strange to call it a low salinity phase.

We respectfully disagree. Some brackish diatom species have been found in Ancylus Lake sediments (Alhonen, 1972) and Winterhalter (1992) noted that saline water flowed into the basin during the Ancylus Lake phase. It seems most likely that the surface water layer was fresh, but there was brackish / saline water at depth.

Line 41. The freshening was probably caused by land uplift, increased precipitation and decreased evaporation, not by lack of large marine transgressions.

We will omit “lack of large marine transgression” in the revised manuscript.

Line 42. until now the authors have discussed salinity changes, but now they state that “The complex climate dynamics caused substantial shifts in the salinity of the Baltic Sea during the Holocene, indicated by changes in the phytoplankton population”. However, the main salinity changes were not caused by climatic changes.

We respectfully disagree and think the main salinity changes are linked with climatic changes. The melting of the SIS followed by large freshwater input and isostatic changes are all linked to climate. These factors contributed to the changes in the regional environment (i.e., salinity of the basin) which in turn led to shifts in the phytoplankton (and higher plant) populations.

Line 57. recalcitrant – is that the same as resistant?

Yes. 

Line 60, 61. Are there any relevant C4 and CAM plants in the region?

To the best of our knowledge, there are no C₄ or CAM plants in the region.

Line 71. It is mentioned that wind-transported n-alkanes are generally deposited within weeks. Does this mean that part of the n-alkanes could have their origin in North America?

Alkanes can be transported long distances. For example, alkanes thought to be from the Sahara desert were found in Atlantic sediments, but in low abundance (Schreuder et al., 2018, 10.1016/j.orggeochem.2017.10.010). It seems unlikely that n-alkanes in our record were transported all the way from North America.

Line 74. preserve information – change to can provide information.

We will make this change in the revised manuscript.

Line 77. convolutes – is this the correct word?

We will change this to “complicates” to avoid confusion.

Line 96. what was the diameter of the piston core?

The piston core has a diameter of 10 cm.

Line 97. what was the water depth at the core site?

The water depth at the core site was around 45 m.

Line 98. Arkona Basin not Arkona basin.

We will change into “Basin” in the revised manuscript.

Line 110. I don’t think the authors extracted sediment samples, it should be lipids.

Organics were extracted from the sediments. We will change the sentence in the revised manuscript.

Line 161. The authors note large difference between the piston core described in the manuscript and nearby cores. I wonder if these differences could be explained if the other cores are gravity cores?

It seems our wording was not clear here. The nearby cores also recorded large changes in sedimentation rate. We will revise the sentence to make this more explicit.  

Line 199. The temperature reconstructions for the early part of the record is not similar to those of Kotthoff et al. (2017). The record by Kotthoff only went back to ca. 7.4 ka. I don’t think that 18.5°C for the Yoldia Sea phase, 11.5°C for the Ancylus Lake phase and 24° fit with other temperature reconstructions from the region.

It is true that the record of Kothoff et al. (2017) only goes back to 7.4 ka. As the only other record of temperatures reconstructed using LDI in the area, we felt it should be included. Our record has similar reconstructed temperatures as presented in Kothoff et al. (2017). Because our main focus was not on temperature, we originally chose to omit more detailed information about this to avoid adding to the complexity of the discussion. We will add more discussion about the utility of this proxy in the Baltic and the similarity with other Baltic temperature reconstructions in the revised version.

Line 212. change Baltic Sea phase to Baltic Sea phases.

We will change accordingly in the revised manuscript.

Line 215. The salinity of the Yoldia Sea phase was first discussed from the presence of Yoldia arctica (now Portlandia arctica, a marine bivalve). However, marine species are only recorded from the Baltic proper (including the Gotland Deep discussed by Sohlenius et al.). Not sure if any marine species have been recorded from the southern part of the Baltic Basin.

Alkenone distributions characteristic of marine alkenone producing species were observed in these samples (described in Weiss et al. (2020)). 

Line 218. Yoldia Sea sediments in the Arkona Basin are usually considered to be barren of fossils (except for reworked pollen and spores). It is interesting to see that the authors found diols that are produced by freshwater eustigmatophytes. However, I wonder if these algae lived in the Arkona Basin, or in rivers and pools in the catchment?

Eustigmatophytes have soft tissues only, but the organic compounds they produce (i.e., diols) do preserve well in sediments and are colloquially known as “chemical fossils”. Eustigmatophytes producing the C32 1,15-diol were likely living in the rivers and ponds of the catchment as suggested by previous studies (Lattaud et al., 2017).

Line 234. The pollen records referred to are not nearby.

We agree with the reviewer that the records are not just next to the core site, however we consider southern/central Sweden to be within the Baltic Sea region.

Line 235. Strictly speaking, temperatures can be low or high, not cold.

We will change this in the revised manuscript.

Line 241. Moros et al. (2002) did not suggest that the Baltic Sea experienced “a large regression” at 10.2 ka – they only suggested a regression. The evidence for this regression was weak. It is currently debated if the Ancylus Lake stage ended with a large regression, a small regression or no regression. If there was a indeed a regression in the Arkona Basin, it was definitely not caused by “a continental uplift”.

We will revise the sentence to state simply that Moros et al. (2002) suggest a regression occurred at 10.2 ka, and then discuss how our data confirms the idea of a regression at this time.

Line 247. The authors suggest that a meltwater pulse occurred at 10.2 ka. However, some studies indicate that the Scandinavian Ice Sheet expanded at ca. 10.2 ka (the Erdalen event).

With our results, and those of Weiss et al. (2020), a meltwater pulse is a possible explanation for the negative alkane and alkenone hydrogen isotope values noted before 10.2 ka. The Erdalen event mainly concerned the western Norwegian glacier, which might explain why the Arkona Basin sediments did not record such an event. Previous studies suggest a decrease of the Scandinavian Ice Sheet at that time (Muschitiello et al., 2015; Cuzzone et al., 2016). The decreased amount of ice melt as source water for plants is reflected in the shift to higher hydrogen isotope values. We will make sure it is clear in the revised manuscript that the meltwater pulse was before 10.2 ka.

Lines 258-259. did you observe a thin layer rich in remains of terrestrial plants at the same level? Such a layer is seen in many cores from the Arkona Basin.

We will check the core for this thin layer.

Line 266 Pinaceae is a family name, it should not be in italics

We will correct accordingly in the revised manuscript.

Line 267 “can be tentatively attributed to Juniperus shrub extension” (should be expansion?). To my knowledge, no pollen records from the region show a Juniperus peak at ca. 9.2 ka. However, the 9.2 event was short-lived and you need extremely high-resolution pollen analyses and high sedimentation rates or varves) to be able to detect possible influence on the vegetation.

We will revise the sentence to say “expansion” rather than “extension”.

Line 268. I don’t understand why the authors chose to compare their record with pollen records from far away (northernmost Finland and Bohuslän in south-central Sweden). Why not compare with nearby pollen records? Anyway, to my knowledge no maximum occurrence of Pinus and Juniperus at 9.2 ka have been reported in pollen diagrams from the Arkona Basin region. See for example the detailed and well dated pollen diagram from Krageholmssjön in Scania in southernmost Sweden (Berglund et al. 2008, Veget Hist Archaeobot).

We will add the two Berglund et al. (2008) findings into our discussion of regional vegetation. These two pollen records show an increase in Pinus between 9.2 and 9.5 ka (Berglund et al. 2008a – doi: 10.1016/j.quaint.2007.09.018, and Berglund et al., 2008b – doi: 10.1007/s00334-007-0094-x) and revise the discussion to emphasize the dominance of woody species between ~11 and 3 ka.

Line 269. what is a regional lake?

By “regional lakes” we mean lakes from the Scandinavian region. We will make this clear in the revised manuscript.

Line 289. Not sure what you mean by this: “The global transition from a glacial to an interglacial climate state across the Holocene, was punctuated by a few abrupt cold events”. The cold events mentioned in the following happened long after the glacial-interglacial transition.

We agree with the reviewer, the actual transition between the glacial/interglacial period happened before the Holocene. The cold events mentioned here occurred during the Holocene. We will correct this sentence in the revised manuscript.

Line 293. It is unclear to me whether the authors see evidence of the 9.2 ka event in their data.

At 9.2 ka we see a peak in δ²H_alkane, which may have been caused by the 9.2 ka event.

Line 297. Moros et al. did not give an age of 7.7 ka for the re-establishment of the connection between the Baltic Sea and the North Sea. From where did you get this age?

Moros et al. gave an age of 6.475 ± 50 ¹⁴C yr BP for foraminifera from the transgression. Warden et al. 2016 gave the age of 7.7 ka for the start of transgression which reached a maximum at around 7.2 ka, this reference will be corrected accordingly.

Line 298. Do you mean that the onset of the transgression lasted from 7.7 to 7.2 ka. Or do you mean that the transgression lasted from 7.7 to 7.2 ka? I believe that the marine transgression of the Arkona Basin began somewhere between 7 and 8 ka and continued for the rest of the Holocene, although the transgression rate slowed down after ca. 6 ka.

Based on the discussion in Warden et al. (2016), the start of the transgression occurred around 7.7 ka and reached a maximum at around 7.2 ka. We will clarify this in the revised manuscript. 

Line 300. regional warming began already in the earliest part of the Holocene, although interrupted by some short-lasting cold events.

We will change this in the revised manuscript. 

Line 304. change n-alkanes were to n-alkane values were.

We are referring to the group of mid-chain n-alkanes so the “s” is needed. 

Line 341. Again, to my knowledge, no “large fluctuations in the extent of gymnosperm cover” have been reported by pollen studies from the region.

We will change this in the revised manuscript.

Line 345. “lack of diversification of terrestrial vegetation noted for this period”. Which period? The Ancylus Lake stage? The pollen records from the region show that many species arrived during this period.

Between ~11 and 6 ka, woody species dominate pollen assemblages in the region and there is a gradual increase in species diversification. We will revise this statement accordingly in the revised manuscript.   

Line 347. “regional warming which continued into the Late Holocene”. Warming certainly did not continue into the Late Holocene.

We will remove “which continued into the Late Holocene.”

Line 351. I don’t think that we can ever “fully understand the complexity of paleoenvironments” as stated by the authors – not even if we use multiple proxies.

Perhaps this is true, but multiple proxies can certainly provide a more holistic understanding. 

Figures and tables

The authors have Age (ka), Age (Ka) and Age (kyr). I am not sure if the journal has a style to follow, but it should be consistent.

We agree that Ka and ka should be consistent. Kyr and ka mean different things: ka means years ago (e.g., 10.2 ka event) and kyr refers to kilo years (e.g., the event lasted for 5 kyr).

Table 2 and 3 can go to supplementary.

We will add these tables to the SI.

The article by Weiss et al. is well written as well as well illustrated and the record is based on a core from a particularly interesting region of the Baltic Sea for which new biogeochemical data is to be appreciated. I am not an expert on this kind of data myself, but think that this publication very well and precisely documents the used approaches, and I think that the results and the discussion are worth being published to broaden the view on the Holocene in the southern Baltic Sea, even though the relatively low temporal resolution of the record makes detailed comparison with other records difficult (see below). I also think that the age model could be explained in more detail since one has not only to visit the supplementary material of Weiss et al. 2020 but ideally also Warden et al. 2016 to get a precise idea how it was generated. Another aspect I see a little sceptical are the results concerning the terrestrial ecosystem structure since I think that with the used proxies, it can only be well estimated concerning a few certain plant types (such as Sphagnum indicated by the C₂₃ n-alkane). In case of several statements concerning the vegetation/climate development, I think the literature cited for comparison is only partly suited to support the results based on the biogeochemical proxies – probably better references for direct comparison can be found (see below).

We thank the reviewer for these positive comments. As mentioned in response to reviewer one we will include more detail on the age model as well as the vegetation changes in the region in the revised manuscript. See the answer to reviewer one: “The core was dated by correlating XRF data to two nearby cores described in Warden et al. (2016). The ¹⁴C ages were corrected using local marine reservoir values from Lougheed et al. (2013; doi:10.5194/cp-9-1015-2013). We will add the two Berglund et al. (2008) findings into our discussion of regional vegetation. These two pollen records show an increase in Pinus between 9.2 and 9.5 ka (Berglund et al. 2008a – doi: 10.1016/j.quaint.2007.09.018, and Berglund et al., 2008b – doi: 10.1007/s00334-007-0094-x) and revise the discussion to emphasize the dominance of woody species between ~11 and 3 ka.” The dynamics of Sphagnum specifically can be difficult to determine in a mixed (sub)Arctic and temperate drainage basin as the dominant chain length of their n-alkanes has been reported to shift from C₂₃ in temperate environments to C₃₁ Scandinavian (sub)Arctic environments (see Vonk and Gustavsson, 2009).

Detailed remarks 

Title and Abstract: I personally think that “structure” in the title implies more than the paper can provide in the end. Consider that in the abstract only a water source shift and a suggested vegetation diversification are mentioned concerning terrestrial ecosystem structure. General, the abstract could give less methodology and more own environment-related results.

We are sorry the reviewer feels this way, but we respectfully disagree. The abstract should reflect the work presented in the manuscript and this type of work is strongly based on analytical chemistry and advances therein. Hence the methodology in the abstract. 

 Sections 1 to 3.2

I have almost no remarks to sections 1 to 3.2, because they are generally very well written and are of appropriate length and focus. One remark only: In line 33 the expression “the SIS melted, exposed the land …” sounds a little odd to me.  I cannot say very much concerning the method sections since this is out of my expertise, but as far as I can tell this is also well done.

We will rephrase the sentence on line 33 in the revised manuscript.

Section 3.3

I am not sure if there is an inconsistency with the discussion here – maybe it should be mentioned here already that the C28 1,14-diol was present in high amounts during the MB phase as mentioned in section 4.4 - the expression “only present” does not imply high amounts in my opinion.

We will add that the 1,14-diols (including the C₂₈ 1,14-diol) were present in a relatively high amount in the MB phase to section 3.3 of the revised manuscript.

Line 199: Checking Kotthoff et al. 2017, it appears to me that the Yoldia Sea (YS) is not reflected in the record described there, a comparison should not start before ca. 8000 yr BP.

As mentioned in our response to reviewer one, “it is true that the record of Kothoff et al. (2017) only goes back to 7.4 ka. As the only other record of temperatures reconstructed using LDI in the area, we felt it should be included. Our record has similar reconstructed temperatures to those presented in Kothoff et al. (2017).” We will make it clear in the revised manuscript that the comparisons of the two records are from 7.4 ka and younger.

Section 4

Section 4.2 line 234 and following: it is implied here that the pollen records mentioned in line 236 are nearby, but I think the records used un Seppä and Birks 2001 are quite far away (some pollen records from northern Germany, Southern Sweden, Denmark or Poland would be nearer).

Section 4.2.2

First paragraph: Again, I am not sure if Seppä and Birks 2001 is well fitting here.

See previous answer and our response to reviewer 1.

In line 269 in the same paragraph it is stated that “the maximum extension of Pinus and Juniperus was recorded at 9.2 ka in these regional lakes”… Checking the cited literature, I can only partly agree: There are only minor increases of Juniperus pollen in the related time interval in Seppä et al. 2005/Digerfeldt 1977 and the Pinus peak is earlier. In Antonsson and Seppä 2007, there is a peak in Pinus percentages at 9.2 ka, but Juniperus percentages are significantly higher during the late Holocene and one could not speak of a maximum extension of this taxon around 9.2 ka. The sentence in line 270 and the following lines is correct concerning Alnus, but the pollen diagram of Lake Trehörningen depicted in Antonsson and Seppä 2007 does much more imply a decrease in Pinus (this one is very clear) than in Juniperus percentages (which does not seem to be consistently present between 11 and 5 ka anyway). I would think in this context that the attribution of the C isotope shifts to Juniperus shrub extension is not supported by the cited pollen records.

A time lag can be expected between the pollen deposited in lakes (usually from nearby catchment) and alkanes deposited in the Baltic Sea sediments (after being transported over relatively long distances). We are talking about each Baltic Sea phase independently, hence when we talk about the “maximum extension of Juniperus” we refer to the relatively higher peak for the specific phase (here the Ancylus lake). Pinus do not produce enough alkanes to have a real impact on the total alkane concentration found in the Baltic Sea (Diefendorf et al., 2011, 2015; Diefendrof and Freimuth, 2017), hence other gymnosperms such as Juniperus are likely the main driver of the carbon isotope composition of the n-alkane. As Juniperus produce a large amount of alkanes, even small shifts will have an impact, especially if the population of angiosperm trees such as Alnus are shifting as well. We do not claim that one type of vegetation is driving the carbon isotope signature of the alkanes but rather the relative increase/decrease of angiosperm versus gymnosperm vegetation. We will rephrase in the revised manuscript and add more comparison sites.

It would be really nice to have closer pollen records for comparison in which coeval Juniperus increases were present to support the interpretation concerning the C isotope shifts (e.g. the record shown by Yu et al. 2005 is quite near, but does not reveal such a signal), or if possible to see pollen data from core 64PE410-S7 itself.

Unfortunately, we do not have a pollen record for this core; however, we will add more comparison sites as noted above and in response to reviewer one.

Line 294: A citation for the 8.2 ka event would be good, particularly if one could be found for the research area.

We cite Fleitmann et al. (2008) for the multiple cold events recorded in the northern hemisphere. We will check for a more regional citation to add to the revised manuscript.

Conclusions:

Here, the 8.2 ka event is mentioned again (and marked in the figures, too) while it was said in 4.2.2 that no such event could be found in the own record and no citation was given in the discussions for such an event. If mentioned in the conclusions, this should be discussed in more detail and with a citation that there was indeed a cold event around that time in the research area. Generally, it would be good if conclusions based on own results would be better indicated.

The 8.2 ka event (as well as 9.2 ka and 10.2 ka) is marked in the figures as a reference point since it can be found in multiple northern hemisphere records. Our conclusions are derived from the data presented in this paper and compared with previous studies.

This is a review for the manuscript “Co-evolution of terrestrial and aquatic ecosystem structure with hydrological change in the Holocene Baltic Sea” by Weiss et al. The authors use a large suite of organic biomarker proxies to analyze environmental change in the Arkona Basin for the Holocene. As expected, large changes in the different proxies indicate large changes in hydrology and possibly climate in the region fitting with the well-known different phases of marine conditions in the Baltic. However, I do have some problems with the current structure and missing discussion. The resolution of the records is very low, in several cases there is only two or three datapoints for a phase. Along with the absence of any information on the age model, this makes a discussion on trends in phases and at their transition and the presence or absence of events like 8.2 or 9.2 kind of useless. The discussion basically reads as a long list, i.e. “this proxy changed in this direction meaning that” without hardly including anything on the many studies in the area itself; I think the ms is missing a big opportunity to make this a much better story (see also the comments below).

I am not an organic biomarker specialist so I cannot judge on the suitability of the methods, although the description of the analyses and their background in the introduction is very elaborately done.

We thank the reviewer for their constructive comments. The two other reviewers also asked for more information about the age model. As mentioned in those responses, “the core was dated by correlating XRF data to two nearby cores described in Warden et al. (2016). The ¹⁴C ages were corrected using local marine reservoir values from Lougheed et al. (2013; doi:10.5194/cp-9-1015-2013). We will add a more detailed discussion of this information to the revised manuscript.”

I’m missing a discussion that is involving the enormous amount of studies that have been performed in the Baltic already. Most references are only related to biomarker records, some of them from non-Baltic locations. Because the resolution and age control are low and you are focusing on different phases, my suggestion would be to restructure the ms by starting to identify your different phases and what they are based on (i.e. existing literature) and then pool the samples you have for those phases into a specific signal for that phase so that you are basically creating snapshots of those phases. In a next step these can then be compared with studies that are especially concentrated around the southern Baltic like the Arkona/Bornholm area (IODP expedition 347 – Site M0065; Heinrich et al., 2018; Anjar et al., 2012; Jensen et al., 2016); Belt seas and Kattegat (e.g. Kotthoff et al., 2017, Ni et al., 2020; Hyttinnen et al., 2020), or lake/terrestrial records from northern Germany and southern Sweden (e.g. Dräger et al., 2017; Hannon et al., 2018). If such studies can be linked with the biomarker results it would make the study much more valuable in identifying the processes behind the signals.

Thank you for this suggestion. We will add a more extensive discussion of the existing literature on the Baltic Sea throughout the Holocene.

Line 31: The Baltic Basin existed long before the deglaciation

We will note this in the revised manuscript.

Intro: A very detailed background on organic proxies, but nothing about other salinity proxies in the Baltic. Many studies have attempted to reconstruct salinity changes in the Baltic, e.g. Gustafsson and Westman, 2002; Emeis et al., 2003; Mertens et al., 2012; Ning et al., 2017, Ni et al., 2020 and others.

We will add a discussion of salinity reconstructions and these references to the first part of the introduction.

First paragraph of the intro could use more referencing, it’s very well studied!

We will add more references to the introduction.

Terrestrial vs marine….how does that work in the Baltic? How do you define the brackish environments with this regard?

The discussion of terrestrial versus marine should be changed to terrestrial versus aquatic (which would then be lacustrine versus marine). Brackish is the mixing of lacustrine and marine.  

Age model: Just a short reference to previous papers explaining the age model is not enough. Sedimentation rates and variations can be extreme in the Baltic especially when changing between the different settings. It is essential that this is explained and shown in the ms. The Arkona Basin is located at an interesting point just after where the saline inflows enter the present Baltic. Timing in this area does not necessarily have to be the same as in the Baltic Proper or the Straits/Kattegat. Simply assuming that this is the case is unlikely to be true. Same goes for comparison with the lake records in Finland.

Other reviewers have suggested that we add more discussion of the age model as well, so we plan to add that into the revised manuscript.

I would not call the first 9.30 meters the top of the core when the whole core is 12 m. A 100 cm resolution is very low; is there any particular reason why this is so slow when downcore records were going to be reconstructed?

Our focus was primarily on the Ancylus Lake phase, hence the higher resolution sampling in that part of the core.

Sediments in the Baltic are notorious for transport of material. How does this affect the different proxies? Radiocarbon studies in the central Baltic have shown that organic matter is continuously re-deposited and can result in large temporal differences. Could this the reason some of your changes are not aligning with commonly accepted events?

Since we do not have radiocarbon ages on organic matter, we cannot rule out redeposition as an explanation for the smoothing out of some major events. However, the proxies used here should be affected by deposition in a similar manner to micropaleontological and other proxies used to establish such events in the Baltic. Those proxies record commonly accepted events, thus we have reason to believe our proxies should as well.  

Line 241: With continental uplift I assume you mean isostatic rebound?

Yes, we do mean isostatic rebound.

Lines 246-248: “The SIS was retreating at this time (Muschitiello et al., 2015; Cuzzone et al., 2016), thus it is plausible that a meltwater pulse transported a higher concentration of n-alkanes from the north into the basin just after 10.2 ka.” This reads like kind of a loose statement. This would require things like age control, and are there signs of a meltwater pulse (e.g. sedimentological)? Has this been shown before, then cite it, and if not you need to bring more explanation.

The sign of the meltwater pulse comes from the hydrogen isotope values and the statement will be revised to note that the meltwater pulse was before 10.2 ka. The increase in n-alkane hydrogen isotope values suggests less contribution from a ²H-depleted (lower hydrogen isotope composition) meltwater contribution to the system.

Section 4.2.1: First you conclude that the low concentration of alkanes indicates less continental runoff, but then following the other proxies you conclude that more continental runoff occurred. What does the literature say about this? Which pathways, e.g. rivers, were in the area, maybe climate was actually changing becoming drier or wetter.

In the early part of the Ancylus Lake, the n-alkane concentrations are low (likely due to ice cover) but then they increase by a large amount. We will make this more explicit in the revised manuscript. An increase in precipitation has been noted for other records, which we discuss in section 4.2.2.

Lines 293-295: “While our record is of insufficient resolution to capture this rapid event, the increase of δ2H alkane values noted at 9.2 ka is presumably also influenced by the environmental conditions present at that time. Another rapid cold event occurred at 8.2 ka, which is not observed in our record, but may be elucidated with higher resolution sampling at this time interval.” Indeed, as the age control is lacking and the resolution low your 9.2 event may well be the 8.2 one.

Yes, that is possible. However, both are cold events and would influence the proxies in the same way.

Several of the curves in the figures have no error bars on them. It would be good to add them. Add other relevant study sites to the map and include them into the discussion, e.g. Bornholm, northern Germany, southern Swedish lakes, Little Belt.

The error bars seem to be missing because they are small in some cases. The errors are listed in the tables (which will be moved to the SI in the revised version). We will add more locations to Figure 1. 

The Pangaea link is still missing.

It will be added in the revised version.

To conclude, I think this dataset definitely has the potential to make an interesting manuscript after re-structuring and expanding the discussion. So currently I recommend major revisions.

Thank you for the constructive comments. We will add the suggested information into the revised version of the manuscript.