Review of Schmidt et al. “Zircon luminescence dating revisited”

In this paper the authors characterised two zircon samples in terms of radioelement concentration in different grains, thermoluminescence spectra, signal bleaching, and auto regeneration for zircon luminescence dating. The paper presents interesting new information about the luminescence of zircon, and was elegantly written. I see no major issue - However, I feel that there is a gap between what are presented (experimental data) and what are discussed at the end. For instance, one of the main outcomes of this study is the too low OSL intensity after 1.5 years of regeneration, when up to a few grains in holes of a single grain disc were used. But could the authors use the information of regenerated signal intensity to estimate how many grains they need for a successful auto regenerated OSL dating? I think that the paper becomes clearer and easier to read, if the authors tidy up the discussion, reducing the parts of “what could be done”, which are not directly based on the results of this paper. Other minor points are listed below.

Sumiko Tsukamoto

1. p. 2, Line 43: Guerin should be Guérin
2. p. 2. Line 48: ug^-1 (-1 should be superscript)
3. p. 2-3, second and third paragraphs of the Introduction: It is not really clear to me why the anti-correlation of U concentration and TL sensitivity leads to age underestimation, nor why averaging the ages of individual grains is expected to yield higher ages.
4. p. 3, auto-regeneration dating: I think that information is missing how long the storage time should be to allow “dating precision of a few years” (line 89).
5. p. 4, 2.1: I do not have an access to Bussy et al. (2000). Please add the type of heavy liquid and its density, which were used to separate of zircon. Please also explain the purpose of the HF etching. Except rutile, other minerals which are mentioned at line 127 have much lower density than zircon, and it should be able to separate them from zircon by density.
6. p. 9 line 235-239 and Fig. S1: in the text the two methods of measuring the U and Th abundance are phrased “bulk and structural”, whereas in Fig. S1, they are “bulk and crystal”. It is easier for readers if one term is used throughout the paper.
7. p. 12, line 258 “this trend is notably more pronounced…”: I am not sure if the trend of weak CL intensity in higher U and Th contents is more notable with elevated radioelement concentration.
8. p. 14, 3.4, line 309: Is there an explanation of why one large zircon grains tend to emit brighter OSL signal than multiple grains in a hole? (if a single grain sits in a hole, the volume of the grain tend to be larger than the sum of multiple grains in a hole??) How does this information help draw the conclusion of the paper (in other words, is Fig. 6 necessary?)
9. p. 16, about OSL components: The way the authors wrote this part gives an impression that they think that the OSL components of quartz and zircon are somehow relatable. But the fact that the signal can be separated into components does not mean that such components physically exist nor can be correlated to quartz OSL components. I think it is ok to say that the photoionization cross section of zircon OSL signal is similar to quartz slow OSL components (and this is consistent with the result of the bleaching test, the signal bleaches slightly slower than the quartz OSL), but no more.
10. p. 21, fig. 9: what are the small black dots?
11. p. 25, 4.2.3, no. 1: Please elaborate what stimulation and detection wavelengths could be effective for measuring zircon OSL, if the stimulation source is pulsed?
12. p. 27, 4.2.4: this part of discussion seems almost nothing to do with the results presented in this paper.

Review of gchron-2024-10 Zircon luminescence dating revisited by Christoph Schmidt and colleagues

In this paper, Schmidt and colleagues assess the viability of the use of zircons for luminescence dating, exploring signal resetting, thermoluminescence spectra, internal radionuclide concentrations for individual grains, and final assessing the use of the auto-regeneration technique. The paper is very well written and structured, and is a welcome contribution to the literature on luminescence dating beyond quartz and feldspar minerals. The other reviewer has identified a structural issue relating to the discussion section, whereby there is a gap between the presented experimental data and the discussion, and I agree with the reviewer that closing this gap, and relating the discussion more closer to the presented data will be of benefit to the paper. Overall, this is a very interesting paper that is well articulated, clearly structured and easy to follow.

Without overly duplicating the suggestions in the other review, I have a few minor comments:

L61: can you add a sentence to explain with there is a negative relationship between U concentration and TL sensitivity? Is relationship only seen with U, or Th as well?

L125: 2 days of etching in HF seems like a long time. How did you design this part of your preparation procedure and/or can you reference this? Were you able to assess whether you have a purified zircon separate after these extended etches, and do you have an estimate of how much of the zircon grains themselves were etched in the process?

L330 onwards: I’m not really sure why you are comparing your zircon OSL components with the quartz OSL ones in this paragraph because they’re unrelated. It would be better to compare your suite of cross-section data to those published by Smith (1989). The dataset is small and limited to only two samples, but are you at a point where you can speculate about the presence of fast, medium, slow etc components for zircons yet, or is your cross-section data too variable?

P27: I agree with the first reviewer that this discussion seems unrelated to the data presented in the paper, and detracts from the findings of your work. You could consider removing this and adding key points into conclusions as directions for future work.

Figure 7: I don’t see the need for the S1 and S3 quartz cross-section reference lines in this figure. I’d also be really interested to see beyond the distribution of cross-sections, and see the size of the cross-sections fitted in conjunction with the number of components fitted. We did this in Figure 2c in Durcan and Duller (2024) https://doi.org/10.1016/j.radmeas.2024.107260 – I’m not saying you should replicate this figure, but it would be interesting to see in what type of fits you see the fastest component appearing for example. You’d also be able to see if there was a component that was ubiquitously present across all grains (or not perhaps).