We thank Reviewer 1 (Brian Gulson) for his positive comments and suggestions.

In response, we are happy to modify our reporting of ²⁰⁴Pb-based isotope ratios to two decimal places as suggested, commensurate with our typical internal precision for SF-ICPMS measurements. Three decimal places were only employed in the first draft of the manuscript as this is the de facto norm for studies in the Earth Sciences.

We are also happy to accept the other minor changes to the text as suggested in his annotated Word document..

We thank Reviewer 2 for their comments on our manuscript and attempt to address the issues raised below.

Primary among these is the suggestion that there may be biases inherent to our dataset for a variety of reasons as outlined in the review.  At one level we agree with the reviewer in so much as biases of one type or another are commonplace in studies of this type employing Pb-isotope ratios of (relatively) small numbers of samples to approximate the isotopic composition of large terrains.  For example, studies aimed at establishing the isotopic composition of the continental crust using sediments have uncovered numerous potential biases associated with tectonic age (e.g., McDermot and Hawkesworth, 1990, Nature),  weathering overprints (e.g., Gallet et al., 1998, E.P.S.L), and the presence of zircons and other heavy minerals (Garçon et al., 2013, G-cubed) to name but a few. The question, however, is how significant are such effects and how can we best minimise their influence?

The reviewer’s first concern is that any lead precipitated in the form of iron oxides or phosphates would be inaccessible to our ammonium acetate leach solution. While chemically true we do not believe that this imparts any noticeable bias in our dataset. The sampling philosophy of the NGSA was always to collect naturally mixed and fine-grained fluvial/alluvial sediments from large catchments, thereby obtaining representative averages of the main soil and rock types contributing sediment through weathering. This allowed the ultra-low sampling density (~1 sample per catchment, which on average is 5200 km²) to remain representative of large-scale natural variations (Caritat and Cooper, 2011, Geoscience Australia Record 2011/020). Given that Pb is present in many extremely common rock-forming minerals contributing to fluvial sediments (e.g., feldspar, pyroxene, amphibole, clays etc) we do not believe that iron and Pb phosphates are the primary – or indeed a significant – sink of Pb when considering processes affecting catchment-level variations. Also we stress that the ammonium acetate leach is really only intended as a first step to remove any obvious anthropogenic influences (primarily aerosols and dust resulting from Industrial processes) and that any Pb components present in iron and Pb phosphates would ultimately be picked up by the subsequent aqua regia leach step, which is of course the one upon which we base our final maps.

The reviewer also notes that ‘Australia has a wide distribution of 3.5 to 2.5 billion-year-old granitic rocks, and that radiogenic Pb is easily extracted in a metamictic state around uranium- and thorium-rich microminerals such as zircon in the rocks, and is therefore extractable with aqua regia. Therefore, the Pb isotope ratios extracted by aqua regia may differ significantly from those of the basement rocks’. While metamict areas of zircon should be more susceptible to an aqua regia leach than pristine zircon, once more we do not believe that such very local effects will impart any measurable bias on our catchment scale dataset. It is well known that the primary sinks for Pb in granitic rocks are the major rock forming minerals, K-feldspar, plagioclase and biotite and while zircon obviously contains highly radiogenic Pb, this usually imparts very little leverage on the bulk rock isotopic signature of granites. The isotopic coherence of the aqua regia leach dataset with no high ²⁰⁶Pb/²⁰⁴ excursions which might reflect radiogenic input from metamict phases supports this conclusion. In addition, there have been many studies of the weathering of granitic rocks and these show that release of Pb from accessory phases generally occurs relatively quickly (in the first few hundred thousand years of exposure) and that, thereafter, the bulk of the Pb released is contributed from the major rock forming phases (e.g., Harlavan & Erel, 2002, Geochimica et Cosmichimica Acta). Once more, therefore, when considering the weathering of catchments covering  thousands of square kms, only some of which may comprise felsic rocks, we think it highly unlikely that the isotopic composition of metamict zircon will impart any observable bias to the resulting derivative sediment.

The reviewer’s third concern relates to analytical matters, in particular the possibility that natural Tl, contributed by the samples themselves, could affect our mass bias corrections using admixed thallium solution. While a valid question of course, it is important to note that natural thallium is often not efficiently removed from samples even when they are subjected to ion exchange chemistry and so our particular situation differs little from many other Pb-isotope studies involving Tl for mass bias correction.  The reason that this sample Tl contribution does not cause a significant problem is that variation in the Tl-isotopic composition of natural materials is in fact extremely low with only a very few environments (particularly ferromanganese sediments in the deep sea) exhibiting significant deviations from the bulk earth value (e.g., Nielsen, Thallium Isotopes, 2018, Encyclopedia of Geochemistry, Springer). As a result, even if we take the entire range of known natural Tl-isotope variation (around ~35 epsilon units), and insert these extreme values into our mass bias calculations, they will not change our derived mass bias corrected Pb-isotope ratios outside of our observed within run precision.

The reviewer makes some suggestions regarding our figures ‘I believe Figures 2 and 3, briefly describing the errors, are not very useful and should be deleted as they are both difficult to read’. This is slightly confusing because Fig 3 does not relate to our uncertainties but we agree that Fig 2 is rather inaccessible to non-specialists and is in fact not really necessary, especially as we provide a detailed analysis of analytical uncertainties in another paper concerned with the methodology itself (Desem et al., 2022, AppliedGeochemistry). Thus we are happy to remove Fig.2.

The reviewer also asks for ‘a brief description of the structural geology of Australia's major tectonic regional divisions and illustrations describing the lithology and basement rock types in each division. Also, the authors should illustrate the Pb isotopic ratio characteristics of each tectonic regional classification, including geologically old Pb, thorium (Th)-rich Pb, uranium (U)-rich Pb, and highly radioactive Pb (e.g., Hsu Y-K and Sabatini BJ, 2019; the case of ore without Pb-204)’. While this might seem to be a reasonable request, we believe that such information is readily available elsewhere and is actually beyond the scope of the ESSD journal format which clearly states ‘Any interpretation of data is outside the scope of regular articles’ and that ‘although examples of data outcomes may prove necessary to demonstrate data quality, extensive interpretations of data – i.e. detailed analysis as an author might report in a research article – remain outside the scope of this data journal. ESSD data descriptions should instead highlight and emphasize the quality, usability, and accessibility of the dataset, database, or other data product and should describe extensive carefully prepared metadata and file structures at the data repository’. We certainly plan additional publications which will attempt to interpret aspects of these data and these manuscripts will of course contain pertinent tectonic and lithological comparisons.

The reviewer suggests that ‘Figure 4 is also unnecessary because they just lays out the data of Table 4 with the error bars’ but we would argue that the visual representation of the data here does in fact add information in so much as it clearly illustrates the correlations which are not immediately obvious from the data table alone.

Finally, we are happy to accept all of the minor changes to the text suggested by Reviewer 2.