Herriott et al. presents an analysis of detrital zircon (DZ) U-Pb geochronologic data as a means of determining maximum depositional ages (MDAs) for strata by comparing the youngest zircons measured from a bed by LA-ICPMS with the same grains measured by ID-TIMS. It is a thorough analysis that focuses on one dataset, and it also compiles data from a few other recent studies that have both ICPMS and TIMS paired analyses.  They show there is a semi-systematic offset between the ICPMS and TIMS data where the ICPMS data is younger – often beyond analytical uncertainty – than the TIMS date from the same zircon.  They discuss the possible causes and implications of this in detail.

I was a bit surprised to find out that all the data in this paper are already published in an Alaskan Survey publication that is easily obtainable and contains some of the same figures and tables (in addition to all the raw data tables).  I speculate that perhaps the great state of Alaska requires this form of publication first, as that first publication reads like more of a data dump. The current manuscript is perhaps seen as an opportunity to analyze the data in more detail and in a way that would appeal to a broader readership, but I can't help wonder if they both need to exist.

I must admit that I’m not entrenched in the DZ literature and therefore am somewhat ignorant to published discussions of the precision and accuracy of DZ MDAs. To be honest, I have just always assumed MDAs in general should be taken with a grain of salt, and those determined by ICPMS are likely inaccurate by some unknown amount, and that playing statistical games doesn’t get you much closer to assessing their accuracy.  I have also noticed numerous examples in my own lab where ID-TIMS dates on the same zircons dated by LA-ICPMS are older, sometimes by tens of Myr or more in the Paleozoic or Ediacaran, so I bring this observation to every MDA DZ dataset I see (i.e., big grain of salt).  Given that predisposition, I find the results and conclusions of Herriott et al. not too surprising and found parts of the paper to be overly verbose in coming to their conclusions. So, I guess assessing the value of this contribution would be best done within the context of someone who uses DZ MDAs, or wants to, but is less aware of the challenges of doing so. Herriott et al. cites numerous papers that appear to provide reviews of or new approaches to analyzing MDAs, so I would hope one of those folks could also chime in on this paper to assess its novelty.  Regardless, I see an obvious contribution from this paper is the reiteration that without paired LA-ICPMS and CA-ID-TIMS dates, you can’t have much confidence that n=1 zircon MDAs determined without chemical abrasion to remove Pb-loss are likely accurate.  Whether or not that matters of course depends on the question being asked in a given study.

For revisions, perhaps the authors could add some statements in the abstract and intro that delineate the novelty of this analysis. I also suggest shortening parts of the manuscript. For example, I bet sections 2.4.3., 3.1.1., 3.1.2. could be considerably condensed (by 50%?).  I would also include some text near the beginning that clearly states what data from this paper are already published and why those data need to be revisited here, in order to alleviate potential concerns about, or impressions that you are, double-dipping.

Here's some line-by-line comments:

27: I find the sentence beginning on this line fairly confusing. Try rewriting or breaking into two sentences.

34: I’ve never heard the word “stratal” used in this way before.  But I guess you mean “same age as the bed the zircons are found in”?

54: the previous two paragraphs assume people already know this, so move it up or perhaps it’s ok to delete.

176: “…to the work by LePain…” could be interpreted to mean that LePain did the correlation, though I think you mean you did the correlation between your samples and those in LePain.

199: I’m not sure what you mean by “in quadrature” here.  Maybe just cite the papers with the algorithms used.

212: If your intended audience includes geochronologists, you might report the MSWD statistic when discussing equivalence of MSWDs.

229-232: I have personally used the term accuracy to also include data interpretation, i.e., how well does a date measure the process you’re interested in.  But I like the suggestion here to use validity in that it separates the two, perhaps in useful ways.

240: out of curiosity, why did the ID-TIMS dates fail for these?

260: could you actually state what the kernel bandwidth is?

Fig 4: I’d put the dates for individual zircons on the figure beneath the sample name, and include you’re preferred depositional interpretation of the tephra samples as well.

Fig. 5 and associated:  I must say that after reading the abstract I was expecting the LA data to be resolvably and consistently younger than the TIMS data.  However, when I look at Fig. 5 it seems 6/10 single analysis pairs agree within 2sigma.  Some LA people may call that a win, whereas it’s being pitched here (at least how I read it) as a loss. I’d recommend revising some of the language with a focus on accurately describing the data comparison given analytical uncertainties. One could raise a similar question for Fig. 9, which may or may or may not look so bleak if analytical uncertainties were included.  But don’t get me wrong, I agree the dates are shifted young and I appreciate the cause is likely to be Pb-loss but for most readers, esp. of DZ data, they want to know if the mean overlaps with the true or not, not if the mean is shifted slightly younger but overlapping with the true given the uncertainty.

Fig. 6: these aerial photos are amazing.

Fig. 7:  I don’t think you should be plotting a KDE with the intent of comparing precision and accuracy of two datasets that measure the same thing with different instruments.  The KDE ignores actual measured uncertainties and therefore is not going to serve your purpose.  I’d plot a PDF in (b) instead.

337: I’d delete “suggest and” and replace with “and are”

374: “very good goodness” sounds funny

514-529: I don’t feel like I have enough information about what Copeland 2020 is proposing for me to get much out of this paragraph.

593-595:  I think this sentence is important, and what I was thinking while reading the previous few pages…

670: it would be useful to describe what the MLA algorithm does.  For me, I’ve never used it, so the following discussion about it is rather black box.

720, Fig. 9: I could have used a better description of what the nth Youngest Tandem LA-ICPMS date means. I got lost here.

724: typo

Review of Herriott et al.: Accuracy and validity of maximum depositional ages in light of tandem (laser ablation + isotope dilution) U–Pb detrital zircon geochronology, including n = 1 results from northern Alaska

The authors present new tandem U-Pb dates from Alaska and discuss their results both in the context of regional (AK) geological implications as well as much broader discussions about the applications of LA-ICPMS and CA-ID-TIMS geochronology for interpreting maximum depositional ages (MDAs). This is a carefully and thoughtfully crafted manuscript which serves as a case study for competing methods and procedures to constrain depositional ages. It is thoroughly referenced and well-written. Moreover, this is an important paper integrating the detrital zircon LA-ICPMS community with the ID-TIMS geochronology community toward a better approach for interpreting max depo ages that is not reliant on selecting from a menu of calculation metrics (or formulas). Rather, this case study demonstrates the utility of CA-TIMS for confidence in youngest single grain dates for MDA interpretations in samples with very few young zircon. 

I think the text is longer than it needs to be. Due in part to the authors trying to account for both the local geological considerations from their new data and the review-style discussion that is intended for a broader audience. Consequently, it feels a bit disjointed at times going back and forth between broader/philosophical discussions and the results of the study from AK. Moreover, many of the sections sort of have their own introductions, which adds up. 

I only have a few general comments/suggestions that might improve the manuscript for the broader community to understand the implications of the results and any implicit recommendations by the authors. 

General comments/suggestions. 

Small point (take or leave): Is there a better word than “stratal”? I presume the authors mean a near-depositional age when referring to “near stratal age”. Not sure I’ve seen this usage before. I think of strata as ‘layers of rock’ and thus strata are not deposited. Sediment is deposited and eventually becomes strata. Nit-picky, I know. I can adjust.

Matching the data presented with the main message(s) of the paper: The introduction nicely frames the background and review of the paper’s topic(s), but falls a bit short in setting up the motivation for the new data pursued/presented in Northern AK. In other words: the problem is presented well, but why is the Alaska dataset the go-to for solving this problem? For instance, if the introduction was more focused on the issues/challenges around interpreting sediment accumulation rates of clinothems then I could see how/why the new analyses from Torok and Nanushuk detrital (and ash) zircons would be pursued. Or if the intro focused more on use of the YSG and how many grains/dates are needed – thereby setting up the potential benefits of one good/reliable CA-TIMS date – that would set up the value of a suite of samples with few young grains that could be leveraged for the benefits of tandem dating. More of a thought/comment than a tangible suggestion. Obviously, this is up to the authors’ discretion, but improvements could be as simple as condense/focus the first four paragraphs of the introduction to set up the benefits of the AK dataset. 

Correlating the clinothem (timelines vs. facies): I wonder about the validity of using samples that have been correlated ~200km apart to calculate sediment accumulation rates on a time-transgressive clinothem. Does this correlation honor Walther’s Law or is it correlating facies rather than timelines (Figure 2)? In the subsurface some of these clinoforms have 100’s of meters of relief. For the sake of demonstrating how CA-TIMS dates can be used for estimation of sediment accumulation rates, I just want to confirm that the expected lateral facies variations are being accounted for when correlating timelines ~200km (dip view-ish) along a delta front clinoform. Or is this assumed to be negligible? Also, are there any other implications for revising the interpreted minimum sedimentation rates beyond achieving more realistic values?

Clarification of benchmarking with CA-ID-TIMS: What is the goal of this section? Are the authors recommending the community to benchmark DZ MDA results using CA-TIMS? If so, can this be clarified/specified for a broad audience (including DZ aficionados and such)? Some studies may seek to collect detrital zircon U-Pb dates exclusively for the purpose of constraining depositional ages and others may primarily be interested in provenance considerations. If someone collects ~300 detrital zircon dates from a sample for provenance information, are the authors suggesting new limits of interpretability of the upper bounds on depositional age if the sample has not been benchmarked by CA-TIMS? If one yields 3 to 4 young dates (or more?) from that sample, they will have to decide if it is worth an additional $500-$1000 (lab depending) to get accurate and precise results for a “half-age”. I certainly agree with the authors that if one wishes to rely on a limited number of young dates/grains to push a young MDA, then they would be wise to explore CA-TIMS. This dataset, and the compilation by Howard et al. (2025) make that point clear. I think the benchmarking section can be condensed into another section. If it remains, I suggest clarification for a broad readership. I was encouraged to see (in the conclusion section) acknowledgement of other considerations one must make when deciding to pursue CA-ID-TIMS with their detrital zircon results (e.g., feasibility, budget, accessibility, etc.).

A refined analysis of tandem date offsets? The authors present a comparison of tandem date pairs from their study, Herriott et al. (2019), and Rasmussen et al (2021). It’s not clear to me how this comparison differs from Howard et al (2025), which I think includes the Herriott et al (2019), Rasmussen et al (2021), and many more. Unless the authors have strong feelings about this comparison, this seems like another part of the paper that can be condensed. Perhaps focus on if/how their new comparison differs from the comprehensive dataset present by Howard et al (2025). In Figure 10 it is not obvious that all the data points fall significantly off the 1-1 line – suggest showing the 3 datasets separately so the offset can be expanded (zoomed in). 

Line 565: Should it be”… older than THE stratal age”?

Line 571: Should chronostratigraphy be pluralized as “chronostratigraphies”? Chronostratigraphic records? 

Line 814: Should that be “rendering”?

Line 853: Maybe clarify if both standards and unknowns were thermally annealed in the Herriott et al (2019) study so that it follows previous discussion.

Thank you for the opportunity to review the manuscript and thanks to the authors for clean, high-quality submission. Best wishes in getting it through to publication.