The manuscript entitled ‘Marked recent declines in boron in Baltic Sea cod otoliths – a bellwether of incipient acidification in a vast hypoxic system?’ is an interesting study focused on temporal trends in otolith boron concentrations and how they relate to water chemistry, salinity, and pH.

Overall, the manuscript has clear objectives and the subject matter is quite relevant. However, I have several issues with the manuscript that I feel need to be addressed and therefore I recommend this manuscript be revised and resubmitted for review. Specifically, there several instances in which statements are made that are beyond the scope of the study. There are also several instances in which an analysis is reported in the results section yet no description of it was provided in the methods section.

My detailed suggestions are listed below.  

Abstract:

1. Line 14: remove the following sentence ‘B:Ca is positively proportional to pH in carbonates, as B in the form of borate is taken up in the CaCO3 matrix’
2. Line 19: remove ‘healthy’.
3. Line 21: What does anti-correlated mean? Does it mean negatively correlated or no correlation?
4. Line 23: Making a statement about the ‘decomposition of large algal blooms’ is beyond the scope of this study and should be removed.

Introduction:

1. Line 42: Sentence is too long. I suggest revising it into 2 sentences. For example, ‘Despite buffering from alkalinity sources, pH is highly variable in space and time with its recent tendency to decline at greater depths in many parts of the Baltic Sea (citation). This decline in pH at greater depths is associated with the ongoing eutrophication and higher vertical export of organic matter that leads also to worsening deoxygenation (Kuliński et al., 2022).’
2. Line 46: Remove ‘past’ from this sentence.
3. Line 49: Can you provide more detail about ‘worsened [body] condition’? For example, was it length to weight ratio?
4. The 2^nd paragraph could use some background information before launching into otolith chemistry and hypoxia. Below I provide a suggestion. ‘The population of the Eastern Baltic cod (Gadus morhua, hereafter referred to as EBC) has been severely impacted by a number of factors that include past overfishing, hypoxia, parasite infections, and seal predation (Eero et al., 2020), leading to reduced growth and age at maturity (Eero et al., 2016; ICES, 2019). For example, EBC exposed to hypoxia were shown to have lower somatic growth rates and poorer body condition (Limburg & Casini, 2018, 2019). Maybe a sentence or two citing other studies that have shown negative links between hypoxia and fish health (e.g., somatic growth/body condition), then the third paragraph could focus on hypoxia and its measurement in otoliths. 3^rd paragraph: Recent studies that reported links between hypoxia and fish body condition have relied on the measurement of manganese concentrations in otoliths (ear-stones) (Limburg et al., 2015). Briefly, otoliths, the calcified structures that form part of the hearing/balance system in teleost fishes, are sectioned and then analyzed by ablating micro-transects along the major growth axis and analyzing by mass spectrometry (see Methods). grow continually by the deposition of layers of calcium carbonate and protein, and as these layers form, ions are deposited onto the otoliths’ growing surface often at concentrations that reflect those in the environment. The resulting elemental data are lifetime concentration histories, being incorporated at the time of exposure. The use of otolith manganese to track hypoxia exposure is one of the emerging biomarkers in fisheries ecology (Reis-Santos et al., 2022).’ 
5. Line 50: This sentence has a lot of information and is somewhat confusing, especially ‘ratio to magnesium which corrects to some extent for growth influences on Mn uptake’. How does it correct for growth?
6. Line 60: I suggest adding your preliminary findings after this sentence. ‘Recently, with acquisition of more sensitive instrumentation, we began to experiment with quantifying elements having sub-ppm concentrations in cod otoliths, including the trace element boron. Interestingly, preliminary results from this work showed elevated levels of boron in 2000 but not in 2019. Boron is noteworthy because it is an indicator of salinity. Specifically, Boron in seawater generally correlates with salinity (Kuliński et al., 2017) predominantly in the form of weak boric acid (H3BO3) at standard seawater salinity (35 PSU) and pH of 8…. ’. Following the above text I would then put a new paragraph that describes your study. ‘In this study, we explored the extent to which otolith B:Ca varied through time and whether its values were correlated with pH, salinity, or other otolith derived values that are proxies for environmental factors.’
7. Line 74: ‘It was therefore surprising when we began to analyze otoliths of EBC captured in 2019 and discovered greatly reduced boron concentrations (Figure 1(b)). The measurements were repeated for verification.’ Figures are not typically referenced in the introduction. If the journal permits it, great, otherwise I suggest removing the figure and just describe that you observed elevated levels of B:Ca in 2000 but not in 2019.
8. Line 84 & 86: No justification is given as to why Mn:Mg and P:Ca are used. I suggest moving lines 146-153 from the methods to the introduction at line 80.

Methods:

1. Line 91: When were the otoliths collected? Were fish of a certain size/age targeted?
2. Line 91: Break into 2 sentences: ‘Otoliths of Baltic cod (N = 156) were obtained from both fishery-independent and fishery-dependent surveys conducted by the Swedish Fisheries Board and its successor, the Department of Aquatic Resources, Swedish University of Agricultural Sciences. Fish were collected in the summers of 2000 to 2019 from ICES sub-divisions (SD) 24, 25, 27, and 28 (Figure 2). Otoliths were extracted, measured, weighed, and embedded in epoxy…’
3. Line 127: Additional details are needed about fish age and ageing techniques used; this text could be placed after the mention of otolith polishing. Also, was age and the location of annuli (ie otolith radii) determined by multiple readers?
   1. It sounds like that the ablation derived trace element data was pooled within each annulus and then an average for each trace element was generated. To do this you would first need to know the otolith radius that coincides with beginning and end of each annulus for every fish, and then bin the ablation data based on those otolith radii.
   2. Ablations were run across the entire otolith, so was the ablation data from both halves of the otolith (ventral and dorsal) averaged? In other words, was the trace element data corresponding to the first annulus pooled from the transect running through both the dorsal and ventral portion of the first annulus?
4. Line 131-135: Sentence too long.
5. Line 138: More otolith growth occurs in the summer rather than the winter, and therefore trace element ratios averaged within an annulus correspond to summer more so than winter. If that is correct, shouldn’t the water data also focus more so on summer time than winter? ‘Annual mean water values were computed and matched to corresponding otolith chemistry data (annual means of Element:Ca data parsed to calendar years as described above).’
6. Line 142: what R packages were used?
7. Line 143: ‘Statistical analyses were separated into examination of relationships of B:Ca values to potential environmental drivers (primarily pH and S, but also AT, DO, and T) and to trace element ratios the internal variables P:Ca and Mn:Mg.’
8. Line 145: This sentence describes a result and should be moved to the Results section.
9. Line 146-153: Move these sentences to the introduction to justify your reasoning for using P:Ca and Mn:Mg. Line 149-153 is too long of a sentence and it should be broken up into at least two sentences.
10. Line 154: define what you mean by ‘graphical analysis’. Also, specify that PCA was performed on water data and why was it used. Will the resulting PC factors be used in subsequent analyses? Describe why you used repeated measures regressions and on what data you used.
11. Line 154: clarify what is the dependent and independent variables used in the mixed models.
12. Line 158: Is there a reason why trends among decades were examined and not at an annual basis? Maybe it has to do with sample sizes, but please clarify.

Results:

1. Line 165: specify that the first two components explained 91% of the variation in the water dataset (71% and 20% on component 1 and 2, respectively).
2. Figure S2 should be placed on a separate page in landscape orientation to increase its size.
3. Figure S3: I suggest overlaying the datapoints in each plot. Also, there isn’t any mention of how otoliths were aged and by whom.
4. Line 174: remove words ‘box plots of’.
5. Figure 4: remove ‘within year’ from y-axis label.
   1. Define what the thick line, box, and whisker correspond to.
   2. For any given year of B:Ca in this plot, does the box and whisker correspond to the B:Ca from several individuals and potentially different annuli?
   3. B:Ca is plotted twice. I think B:Ca should be plotted once and both salinity and pH included as a line plot using the second y-axis.
6. Table 1: Why was the PCA performed if the resulting PC factors, which summarize the water data into two varaibles, weren’t used in subsequent analysis? I suggest removing the PCA from this paper.
7. Line 193: I’m confused. What kind of two factor tests were performed.
8. Line 195: clarify what you mean by the ‘overall model’. I don’t see this analysis described in the methods section. Please clarify.
9. Line 211: What are the r-squared of figure 5 plots? ‘Annual average P:Ca was highly positively correlated with B:Ca in Icelandic cod otoliths, but negatively correlated in Baltic cod with a great deal of scatter’
10. Line 217: Remove ‘Box plots’ from this sentence.
11. Line 218: I don’t see an ANOVA described in methods. Please add a description of how this analysis was used to the methods section.
12. Line 225: ‘We used a similar analysis of individual fish regressions of Mn:Mg, our hypoxia exposure proxy, on B:Ca, as was performed with P:Ca.’ This should be described in the methods section.

Discussion:

1. Line 247-249: remove this paragraph because it is irrelevant to the study.
2. Line 257: I suggest more details: ‘In this study, univariate regressions of otolith B:Ca on five abiotic water variables all were highly significant, with dissolved oxygen and pH showing positive relationships while negative relationships with alkinity, temperature and salinity (Table 1).’
3. Figure S2 is referenced quite a bit in the discussion and thus I think it should be moved from supplementary section to the main body of the manuscript.
4. Line 294: ‘….we can observe a regime shift and a clear decoupling…’. These are substantial statements about the environment changing. Conclusions about regimes shifts are not within the scope of the study and thus I don’t feel they are needed. I would just rephrase these sentences to talk about trends and/or variability.