the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 10 Jun 2025
A Bioavailable Strontium Isoscape of Australia
Abstract. Strontium isotope ratios (⁸⁷Sr/⁸⁶Sr) at the Earth’s surface offer powerful tools for geological, environmental, and archaeological applications. In minerals and biological materials, ⁸⁷Sr/⁸⁶Sr reflects the isotopic composition of the local bedrock and derived soils. In Australia, however, large regional-scale surveys of bioavailable ⁸⁷Sr/⁸⁶Sr remain scarce. Here, we present a new dataset of bioavailable ⁸⁷Sr/⁸⁶Sr ratios from 278 catchment outlet (floodplain) sediment samples, spanning inland southeastern Australia (South Australia, New South Wales, Victoria), northern Western Australia, the Northern Territory, Queensland (north of 21.5° S), and the Yilgarn Craton in southern Western Australia. Combined with more than 20,000 global Sr isotope measurements, this dataset was used to generate a high-resolution isoscape of Australia using random forest regression (Bataille et al., 2020).
Australian bioavailable ⁸⁷Sr/⁸⁶Sr values span a narrower range (0.70501–0.78121) compared to co-located bulk sediment values (0.70480–1.09089) (Caritat et al., 2022, 2023, 2025b), reflecting the influence of soluble and exchangeable mineral phases and atmospheric inputs such as rain and dust. The predicted isoscape reproduces major geological patterns, with higher values over ancient crustal provinces like the Yilgarn Craton and eastern Palaeozoic orogens, and lower values across younger sedimentary basins and coastal margins. Model uncertainty, assessed via prediction standard deviations, is lowest across well-sampled, geologically stable regions and highest in coastal and lithologically complex zones. Compared to existing global and regional isoscapes, our model offers significantly improved coverage and resolution for Australia. This isoscape provides a robust baseline for applications in provenance research, palaeoecology, and environmental geochemistry.
- Preprint
(1142 KB) - Metadata XML
-
Supplement
(40902 KB) - BibTeX
- EndNote
Status: open (until 24 Aug 2025)
-
RC1: 'Comment on essd-2025-277', Anonymous Referee #1, 16 Jul 2025
reply
A Bioavailable Strontium Isoscape of Australia
Anthony Dosseto, Florian Dux, Clément Bataille, Patrice de Caritat
Abstract
Can you better define the difference between the catchment outlet sediment samples used in this study, and the co-located bulk sediments to which results are compared.
Introduction
Line 49-50: also Sr content and more specifically the elemental Rb/Sr ratio. A high Rb content will move the 87Sr/86Sr ratio little if the Sr content is very high, but will significantly move this ratio if the Sr content is very low. So the Rb/Sr ratio is key. A little comment, but important to note.
Study Area
Material and Methods
Material
Some repetition with Study Area section, these could be combined?
Methods
The 87Sr/86Sr data were not referenced to a specific value for SRM987? If these were, then please provide this reference value to allow easy re-referencing in future to other data sets. Or at least the average for the SRM987 analyses during this project. That should also allow for comparison between datasets generated in different facilities.
Strontium Isoscape Calculation
Results
Bioavailable and Bulk ⁸⁷Sr/⁸⁶Sr Distributions
Fig 3. Colours of histogram differ from legend
Fig 4. Use same colours for regions as in Fig 2. Using consistent colours for the same subsets makes it easier for the reader to “move” between figures
Predicted Isoscape and Regional Patterns
Line 197: the colour range used in Fig 5 makes it very difficult to clearly discern the marine influence in the northern and northwestern Australia. It will take a keen eye to teel the difference between a value of 0.715 and 0.709 using the blue in the colour range.
Prediction Uncertainty
I agree that Fig 6 is critical to assess the model output, but again the colour range used makes it very difficult as most of the figure is variable shades of dark blue.
Model Performance and Variable Importance
Discussion
Line 235: again, it doesn’t matter to the 87Sr/86Sr ratio if the rock is rich in Rb if it is also rich in Sr, and the 87Sr/86Sr value can significantly change if the rock has average or low Rb and EVEN lower Sr. So, Rb/Sr elemental ratio trumps Rb concentration only
Line 245: yes, but it is difficult to use Fig 6 for the purpose of assess local uncertainty with the current colour scheme. Can alternative schemes be tested?
Conclusions
General comment
The Sr isoscape produced is impressive and will indeed be invaluable for future research. I realise this is not included at the moment, but a small test application as example would add greatly to the manuscript. If the authors have some 87Sr/86Sr data for plant material for a specific region or site in Australia at hand – say from some published project – this could be used to show that the isoscape can be used to provenance these to the right place.
Citation: https://doi.org/10.5194/essd-2025-277-RC1 -
AC1: 'Reply on RC1', Anthony Dosseto, 08 Aug 2025
reply
We thank the reviewer for their constructive comments, which have helped improve the clarity and impact of the manuscript. Below we address each point in turn.
Abstract
Comment: Can you better define the difference between the catchment outlet sediment samples used in this study, and the co-located bulk sediments to which results are compared.
Response:
We have clarified this in the Material and Methods sections. The bioavailable Sr was extracted from the <2 mm fraction of 'top outlet sediment' (TOS) samples using NH₄OAc leaching, whereas the bulk Sr values refer to total acid digestion of co-located sediments, inclusive of all mineral phases. This distinction is now explicitly stated in the revised manuscript.Introduction
Comment (Lines 49–50): Also Sr content and more specifically the elemental Rb/Sr ratio. A high Rb content will move the 87Sr/86Sr ratio little if the Sr content is very high...
Response:
We agree and have revised the relevant passage to clarify that the Rb/Sr ratio, rather than Rb content alone, governs the radiogenic evolution of ⁸⁷Sr/⁸⁶Sr. This change improves the accuracy of our geochemical context.Study Area / Material Section
Comment: Some repetition with Study Area section, these could be combined?
Response:
We acknowledge the overlap between the Study Area and Material sections and have revised both to reduce redundancy while maintaining their distinct roles.Methods – Referencing SRM987
Comment: The ⁸⁷Sr/⁸⁶Sr data were not referenced to a specific value for SRM987...
Response:
We now report the average ⁸⁷Sr/⁸⁶Sr value measured for SRM987 during the analytical campaign: 0.709810 ± 0.000044 (2SE, n = 90). As this is lower than the accepted value of 0.710252 ± 0.000013 from Weis et al. (2006), all sample ⁸⁷Sr/⁸⁶Sr ratios were normalised to the Weis et al. reference value using session-specific SRM987 means. This correction facilitates accurate inter-laboratory comparison and allows future users to re-reference the data if needed.Results – Fig 3 and Fig 4 Colour Consistency
Comment: Colours of histogram differ from legend
Comment: Use same colours for regions as in Fig 2...
Response:
Figure 3 compares bioavailable and bulk ⁸⁷Sr/⁸⁶Sr values rather than regional subsets, so regional colour consistency does not apply. However, we have corrected the legend to ensure colours match the plotted data. In Figure 4, regional colours have been updated to match those used in Figure 2 for consistency across region-based plots.Fig 5 – Colour Range
Comment: The colour range used in Fig 5 makes it very difficult to discern subtle differences...
Response:
We acknowledge that the colour scale in Figure 5 may limit visual differentiation of lower ⁸⁷Sr/⁸⁶Sr values, particularly in northern and northwestern Australia. To address this, we have added an alternative version of the isoscape to the Supplementary Material using a different colour ramp optimised to distinguish values in the 0.709–0.715 range.Fig 6 – Uncertainty Colour Scheme
Comment: ...most of the figure is variable shades of dark blue.
Response:
We acknowledge the reviewer’s concern that most of Figure 6 appears as variable shades of dark blue, making it difficult to interpret spatial patterns in uncertainty. To address this, we have retained the main figure for consistency with the other maps, but included an alternative version in the Supplementary Material (Figure S2) that uses a modified colour ramp to enhance visual contrast in the lower SD range. This supplementary figure allows better differentiation of uncertainty across the full prediction surface.Discussion – Line 235
Comment: ...Rb/Sr elemental ratio trumps Rb concentration only.
Response:
Agreed. The discussion has been updated to state that the Rb/Sr ratio is the key control on radiogenic evolution of ⁸⁷Sr/⁸⁶Sr, not absolute Rb concentration alone.Discussion – Line 245 (Fig 6 usability)
Comment: Can alternative schemes be tested?
Response:
Yes. While the main figure has been retained for consistency with the rest of the manuscript, we have included an alternative version in the Supplementary Material (Figure S2) that uses a modified colour scheme to enhance contrast across the range of prediction uncertainty. This version improves interpretability, particularly in areas with low standard deviation values.Conclusions / General Comment
Comment: The Sr isoscape produced is impressive... but a small test application would add greatly...
Response:
We agree and have included a short application example in the Discussion section comparing predicted values to published plant/water/soil data from Adams et al. (2019) in Cape York Peninsula. This demonstrates the model’s applicability to provenance studies and strengthens the practical relevance of the isoscape.Citation: https://doi.org/10.5194/essd-2025-277-AC1 -
RC3: 'Reply on AC1', Petrus le Roux, 08 Aug 2025
reply
I appreciate the positive replies and updates to the manuscript. I have no further comments
Great job!
Citation: https://doi.org/10.5194/essd-2025-277-RC3
-
RC3: 'Reply on AC1', Petrus le Roux, 08 Aug 2025
reply
-
AC1: 'Reply on RC1', Anthony Dosseto, 08 Aug 2025
reply
-
RC2: 'Comment on essd-2025-277', Ian Moffat, 04 Aug 2025
reply
Thanks to Dosseto et al for making another important contribution to the geochemical literature in Australia. This is an impressive piece of research that fill an important gap in providing bioavailable strontium isotope values for Australia.
The authors rightly contend (lines 34-39) that geochemistry can make important contributions to understanding ecological histories, the origin and movement of archaeological materials/human remains and provenancing agricultural products but, sadly, fail to offer any Australian references from previous statements to support these statements.
My principal concern with this research is the use of catchment outlet samples, rather than those associated with the local bedrock in a more immediate way. I wonder if the authors could comment on this choice and how it may have impacted the results of the analysis and modelling?
The authors are right to highlight the foundational work of Adams et al 2019 in Australia but should probably also mention the (much smaller number of) results from Rippon et al. 2020 https://www.sciencedirect.com/science/article/pii/S2352409X20304053
Citation: https://doi.org/10.5194/essd-2025-277-RC2 -
AC2: 'Reply on RC2', Anthony Dosseto, 08 Aug 2025
reply
We appreciate the reviewer’s positive feedback and helpful suggestions. Below we respond to each concern raised.
Introduction – Citations
Comment: ...fail to offer any Australian references to support these statements...
Response:
Thank you for this helpful suggestion — we were not previously aware of the study by Rippon et al. (2020), and have now included it. While the opening paragraph of the Introduction is intended to broadly frame the relevance of provenance tools across disciplines (without reference to specific isotopic systems), we have added citations to relevant Australian work at the point where strontium isotopes are first introduced. Specifically, we now cite Adams et al. (2019), Rippon et al. (2020), and Caritat et al. (2022, 2023, 2025b) to acknowledge existing contributions to Sr isotope provenancing and geochemical mapping in the Australian context.Main Concern – Use of Catchment Outlet Samples
Comment: ...rather than those associated with the local bedrock... how might this have impacted results?
Response:
We have expanded the Discussion to explain this methodological choice and its implications. Catchment outlet sediments were chosen for their integrative properties at regional scales. While this limits fine-scale resolution, they provide a robust first-order approximation of regional bioavailable Sr, suitable for broad-scale isoscape modelling. We acknowledge that point-sourced biological samples (e.g. plants) would be ideal for local-scale provenancing and encourage future work to incorporate such samples, especially for applications requiring finer spatial resolution.Additional Citation – Rippon et al. 2020
Comment: ...should probably also mention Rippon et al. 2020...
Response:
Thank you for bringing this study to our attention. We have now cited Rippon et al. (2020) in both the Introduction and Discussion. In the Introduction, it is referenced alongside Adams et al. (2019) as an example of Australian Sr isotope data relevant to provenance studies. In the Discussion, we highlight Rippon et al. (2020) as an important regional effort to characterise bioavailable Sr in the Adelaide region using low-mobility fauna, further supporting the value of baseline Sr data for archaeological applications.Citation: https://doi.org/10.5194/essd-2025-277-AC2
-
AC2: 'Reply on RC2', Anthony Dosseto, 08 Aug 2025
reply
-
RC4: 'Comment on essd-2025-277', Anonymous Referee #3, 16 Aug 2025
reply
This is an excellent article, given the previous comments I have very few concerns.
The authors have collected previously published data, and measured 278 new samples to develop a more robust and spatially representative isosocape of Australia. They have provided a detailed paper that permits reproducibility, and a high resolution isoscape that can be used by a variety of disciplines.
Lines 30 - 39 lacks citations.
There should be a brief discussion in the paper that highlights the comparability of each sample type (plant, soil, water) to animal/human tissues.
Furthermore, a detailed breakdown of the total sample used to generate the isoscape, particularly the percent plant, soil, and water samples used. This information could easily be included in the supplemental information.
It is nice to see that the tiffs are included in the supplement.
Citation: https://doi.org/10.5194/essd-2025-277-RC4 -
AC3: 'Reply on RC4', Anthony Dosseto, 18 Aug 2025
reply
We thank Reviewer 3 for their constructive and thoughtful comments, which have helped us strengthen the manuscript.
Reviewer comment: Lines 30–39 lacks citations.
Response: We thank the reviewer for noting this omission. We have now added the appropriate citations in this section to support the statements made.Reviewer comment: There should be a brief discussion in the paper that highlights the comparability of each sample type (plant, soil, water) to animal/human tissues.
Response: We agree that this is an important consideration. Accordingly, we have added a short discussion outlining the comparability of plant, soil, and water samples to animal/human tissues, and the strengths and limitations of each sample type when applied in provenance studies.Reviewer comment: Furthermore, a detailed breakdown of the total sample used to generate the isoscape, particularly the percent plant, soil, and water samples used. This information could easily be included in the supplemental information.
Response: We appreciate this suggestion. We have now included a detailed breakdown of the dataset in the Supplementary Information (Table S1), specifying the proportions of sample types used. The database consists of 23.6% plant samples, 16.1% soil samples, and 25.7% water samples, with the remaining 34.6% comprising other categories (e.g., animal tissues, shells, rock).Reviewer comment: It is nice to see that the tiffs are included in the supplement.
Response: We thank the reviewer for this positive feedback and are pleased that the inclusion of the raster files is helpful.Citation: https://doi.org/10.5194/essd-2025-277-AC3
-
AC3: 'Reply on RC4', Anthony Dosseto, 18 Aug 2025
reply
Data sets
A bioavailable strontium isoscape of Australia. Initial contribution P. de Caritat et al. https://dx.doi.org/10.26186/150024
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 302 | 57 | 28 | 387 | 22 | 16 | 32 |
- HTML: 302
- PDF: 57
- XML: 28
- Total: 387
- Supplement: 22
- BibTeX: 16
- EndNote: 32
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1