General Comments:

Thank you to the authors for this important data set, which addresses an expansive deficiency in strontium isotope mapping in Northern Australia.  The methodology of this study is robust, and it will provide an important tool to facilitate future research in this area.

Several aspects of this study are of particular note.  The analysis of the relationship between the "top outlet sediment" and "bottom outlet sediment" is novel and expands the interpretability of this data set.  The consideration of strontium isotope values compared to mineralisation is also a very interesting inclusion.

Specific Comments:

I suggest the authors make explicit that food tracing, provenance and archaeological applications of strontium isotope tracing (referred to specifically in lines 64-68) rely on bioavailable rather than "whole rock" values and so don't benefit directly from this data set.

The decision to analyse catchment outlet sediments is understandable given the availability of these samples from the National Geochemical Survey of Australia, however this methodology differs from most other strontium isoscape studies and has important implications for interpretation of these results.  I suggest the authors elaborate on these issues, perhaps as a new section in the discussion.

The data set presented by the authors in this paper is extremely useful and interesting however it could be considerably enhanced by including the bedrock geology, age and stratigraphic unit (as summarised in Figure 6), notwithstanding that these samples have been collected from fluvial systems.

Technical Corrections:

Figure 1 is extremely small in the manuscript and the sample locations use the same colour as the Towns and Places, can I suggest this is revised to make it easier to view?

General comments

The manuscript “A strontium isoscape of northern Australia” by Caritat et al. presents a new large-scale 87Sr/86Sr isotope dataset for Northern Australia based on archival fluvial sediment samples. The scope of the study is very impressive, and the data are going to be very useful for the science community. The analytical methodology and statistical evaluation of the data are appropriate and provide confidence in the provided results. Specifically, the use of archived sediment samples that should represent a mixture of the varied geology is very interesting and overcomes the lack of samples from Northern Australia. Overall, I believe this is a very important contribution and it is great to see that the authors have chosen to make this data available.

Specific comments

• I think it would benefit the reader to make it very clear that the data presented here are total Sr and not bioavailable Sr. This does get mentioned in the manuscript, but I think it should be front and center to avoid any potential confusion and incorrect use of the data. I am also interested in why the bulk Sr extraction was chosen rather than bioavailable? I think elaborating on this in the discussion would be useful and interesting. Is there any use in getting bioavailable Sr out of these samples?
• In the quality assessment section (Line 222-228): To confirm the data are interpreted based on the 3^rd decimal place because the within sample variability was high? Why is that so? Is that a result of the total Sr extraction method, the type of these sediments, or another process? This should be elaborated on in the discussion. For example, would you recommend analyzing more sediment samples to get a better handle on this? Or a different extraction method? Or a different sample type?

Technical corrections

Abstract

I focused on the abstract as this is the most widely read part of the paper. The following comments are all suggestions that I hope improve clarity.

Line 8: Maybe add “useful as a tracer” or something similar. Just useful seems not very informative.

Line 10: I think it should be “archived”.

Line 11: Suggest removing the reference style (last accessed) from the abstract (here and at the end).

Line 12-14: This is a lot of method detail for an abstract. Since the paper is open access and the method section will be available to all to read I suggest to condense this to a single sentence. “Total Sr was extracted and measured…”

Line 16: Why preliminary?

Line 21: Can you add a range to the carbonate units. Also, the word “signature” at least to my ears implies some type of unique value so maybe the word “range” or “ratio” is better here.

Line 27: How should these data be used in archaeological, forensic, and ecological studies? The sentence from Line 76 might actually be a good fit here. “The present study affords an opportunity to further redress this deficiency and will reduce the northern-hemisphere bias in future global 87Sr/86Sr models.”

Introduction

Line 40-70: I don’t believe all these references are needed or useful for the reader. Maybe focusing on 2-3 key references per statement would improve this intro.

Figure 1: Great figure but small in the pdf and symbology is hard to distinguish (maybe different symbols could improve this). Actually all figures could be larger (full width).

Data analysis

Line 230: Why choose Excel rather than R or python for the data analysis? One of the main benefits of using a scripting language would be that others could reproduce the results (and figures) directly.

Figure 4: Remove one of the legends to save space and as before make the figure larger

Table 1: Really useful compilation! Thank you.

Conclusions

Line 373-381: This could be removed to provide more focus in the conclusion. The number of samples and the range could be incorporated into the following sentences.

Also again be very specific about how these data could be applied in provenance studies in archaeology/forensics.

General comments:

Congratulations to the authors. This paper contains a great quality, highly valuable large-scale bulk soil ^87/86Sr dataset that will be applicable to many areas of scientific research, for an area of Australia where there was previously very little information available. Within the paper the bulk soil ^87/86Sr dataset is compared with catchment specific information including bedrock age, lithology and mineralisation.

Specific comments:

Soil samples were taken at the bottom of the catchment which is a unique and pragmatic approach for maximising coverage. How does this compare with previous approaches and effect interpretation?

It would be interesting to see how bulk soil ^87/86Sr compares to ^87/86Sr in other environmental materials within each catchment. I am interested in how it might compare to bioavailable ^87/86Sr leachates from soils for provenance studies. This is an idea for future work, rather than for inclusion in the current paper.

Catchments with high ^87/86Sr were compared to lithologies. It would be useful to see this comparison across all catchments.

Technical comments:

• Well structured, well written, easy to follow.
• Figure 2a: Valuable diagram, but hard to make out bedrock units. Maybe adjust colour scheme and increase size?
• Figure 4: May be worth including some of the major landmarks that you use in the discussion, for example, Cape York.
• Lines: 262 – 265: good points.
• Table 1: Whilst digestion does hint at whether analyses were bulk soil versus bioavailable, I think it would be worthwhile making a clear distinction in the table which are which.
• Line 283: There is a good correlation between TOS and BOS ^87/86Sr. Will you please explain why you would want to recalculate TOS for BOS?
• Line 302-305: It would be great to compare this with all catchments and graph it similarly to Figure 6.