the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 11 May 2026
Technical Note: A retroactive method for identifying subpopulations of zoned zircon in (U-Th)/He data
Abstract. Zircon (U-Th)/He thermochronometery (ZHe) is a widely used tool for investigating and dating thermal events such as uplift, exhumation, incision, and plutonic emplacement, among others. The utility of this thermochronometer relies on its ability to measure and predict the production, distribution, and ultimately diffusion of radioactive decay products, a thermoregulated process. Natural zircon crystals frequently display internal chemical zonation as a result of precipitation environment, and metamorphism. This internal chemical heterogeneity unevenly partitions radioactive actinides, alpha particle production, and resultant accumulated radiation damage within the grain. The heterogeneous distribution of radiogenic He and radiation damage accumulation change the diffusion kinetics of these zircon, hindering our ability to calculate dates and ultimately develop thermal history interpretations. Models reveal that the effect of zoned actinides and associated radiation damage is magnified for samples characterized by negative ZHe date-eU trends produced by protracted thermal histories including extended residence in the partial retention zone. Measuring the degree and distribution of zoned radiation damage accumulation requires destructive characterization processes which precludes characterized zircon from whole grain ZHe dating. It is otherwise difficult to isolate zoned zircon during standard grain picking procedures, implying that many zircon grains that could be used for (U-Th)/He dating have spatially heterogeneous accumulated radiation damage. I developed a retroactive test to identify populations of potentially zoned zircon within a dataset utilizing the HeFTy v2.3.1 (Ketcham, 2025) forward modeler and “zoned” grain function. This test allows us to reproduce the behavior of endmember zonation styles in date-eU space, with the ultimate goal of identifying and reclaiming these data to develop more robust thermal history interpretations. Considering zircon damage zonation may be helpful when interrogating samples with complex ZHe date-eU patterns that are otherwise difficult to interpret.
- Preprint
(1225 KB) - Metadata XML
-
Supplement
(4 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2026-1672', Anonymous Referee #1, 02 Jul 2026
-
AC1: 'Reply on RC1', Jenna Kaempfer, 11 Sep 2026
Responses to reviewer comments are itallicised.
This work attempts to explain the dispersion present within a selection of highly dispersed samples. By generating a series of synthetic samples with different zoning characteristics and predicting ages using forward modelling. These new data are shown to overlap with measured dates as a means of showing than zoning may have played a significant role in the dispersion of grains. The workflow presented is understandable and can be seen as a simple way of screening grains to better understand dispersion and extract some thermal history information. However, in its current form, this article feels underdeveloped and highly localized to just this small collection of samples. Several inclusions could be made to make this more impactful and to ensure this approach is more broadly applicable.
The author is thankful for the comments and suggestions made by Reviewer 1 about the quality and presentation of the proposed methodology. Individual comments are addressed below.
Scientific Significance
I feel this work does not offer a ‘substantial’ contribution to geochronology as its limited scope within a small collection of samples from the Northern Madison Range in Montana. Testing this method on a number of other cases studies would be more convincing, which would possibly allow for more zoning combinations to be attempted. From my understanding the data need to test this workflow would be readily available in any publication of zircon (U-Th-Sm)/He dates and could warrant a wider selection of samples to be tested.
You are correct that this contribution would benefit from the inclusion of additional sample collections. Samples from the Northern Madison Range are used as an example of complex thermal histories due to its multiple exhumation and reheating events, resulting in a negative date-eU trend. Datasets within this technical paper were chosen because they exhibit diverse outliers that can be identified and explained using the new protocol. The author and colleagues are preparing a separate manuscript which includes multiple published sample sets with a variable trends and sample dispersion in date-eU, alongside independent characterization analysis of zircon zonation types found within each sample. The author believes the inclusion of additional sample sets in this technical note would significantly expand the discussion to topics beyond the base functionality of the presented method.
Additionally, could the newly acquired information about zoning be then reincorporated into an inverse model that shows how this work can help explore the model space better? This thermal history approach feels like it only is capable of forward modelling based on a given thermal history, and it use in inverse modelling could be very beneficial.
The author agrees that integrating this method with inverse modelling techniques would be valuable. That is the ultimate goal of this work. The author will include a “Future Work” section detailing possible applications and integrations of this work.
Scientific quality:
The work is well written, though a deeper analysis of the results is merited. In its current form the comparison between observed and modelled data appears to be visually based (overlapping in plots). A more robust statistical test that actually compares this dataset would help show how well these modelled data are. Some comparisons could help better fine tune forward modelling and find a better fit for zone/rim ratios.
Thank you for your suggestion to include statistical testing of the modeled dataset. A section on statistical comparison will be included in the revised manuscript.
Presentation quality:
Though not necessary for publication, the quality of figures and inclusion of perhaps an additional one would benefit this work. I do feel the plotting of data could be improved (perhaps coloring samples point by eU would benefit). Additionally, a conceptional figure showing a zircon with enriched cores or rims would perhaps help non-specialists.
Figure quality will be addressed in the manuscript revision. A conceptual figure demonstrating zonation types will be added, thank you for this suggestion.
Citation: https://doi.org/10.5194/egusphere-2026-1672-AC1
-
AC1: 'Reply on RC1', Jenna Kaempfer, 11 Sep 2026
-
RC2: 'Comment on egusphere-2026-1672', Anonymous Referee #2, 18 Jul 2026
Kaempfer presents a reverse-engineered model of zoned zircon (U-Th)/He thermochronology. The manuscript addresses a genuine problem that is well laid out and explained. Current diffusive models do not account for internal chemical zonation, including radiogenic daughter products. Other complexities, such as radiation damage, modify the He diffusion kinetics, which can enable a more detailed time-temperature path interpretation. The significant problem occurs for FT corrections only when zonation causes most eU to be concentrated either within/outside of 15 μm of the rim. So, this issue is most prevalent in zircon grains with metamorphic overgrowths that can cause large overdispersion.
The manuscript currently remains largely conceptual and qualitative, which may be the right breadth/scope for a technical note in GChron. Greater discussion of assumptions, limitations, parameter sensitivity, and the non-unique nature of interpreting date-eU scatter would make this technical note stronger. With a few key revisions, this manuscript is suitable for publication in GChron. Outlined below are several points that would bolster the readership, support interpretations more strongly, and streamline the flow.
- The central claim is that overlap between forward-modeled hypothetical zoned grains and observed outlier populations can be used to identify likely zoned zircon. However, the manuscript currently lacks validation using zircon grains whose zoning has been characterized independently (e.g., CL imaging, LA-ICP-MS mapping or depth profiling, Raman damage mapping). However, I do understand the utility of forward modeling in this case, as independent characterization is not part of traditional workflows, and this model could be used to interrogate previously published data. Also, traditional workflows are destructive, so there is no easy or cheap way to include independent characterization without mounting the grain, then plucking and dissolving the half-grain, which would introduce some error. It is worth noting in the manuscript that in situ thermochronology techniques can map U and He concentrations (e.g., Danišík et al., 2017; Pickering et al., 2020; Vermeesch et al., 2012). Depth profiling unpolished grains on double-sided sticky tape, which can then be plucked and dissolved as whole grains for (U-Th)/He, is another way to assess core-rim U relationships (e.g., Orme et al., 2015; Pujols and Stockli, 2021). Perhaps the manuscript needs more explicit guidance that this modeling technique is not conclusive and should be used alongside other evidence that the grain is zoned. At present, Figure 3 demonstrates only that a hypothetical zonation scenario can reproduce the behavior of some outliers in a published dataset. This is suggestive but does not independently demonstrate that the outliers are truly zoned grains.
- The workflow requires a "best-fit" time-temperature path that already reproduces the inlier data. This is an important assumption that warrants further discussion, especially because many datasets showing significant date dispersion are those where thermal histories remain poorly constrained. Discuss how uncertainty in thermal history propagates into predictions of zoned-grain behavior. Would different acceptable thermal histories produce similar hypothetical date-eU arrays? A brief sensitivity analysis would substantially strengthen the manuscript.
- A clearer explanation of the hypothetical grain parameters that are iteratively adjusted to match real data should be moved to the abstract and placed earlier in the discussion. This is the key process that drives the forward model. I suggest providing evidence that these values are representative of common ZHe datasets, examples from published zircon populations, and discussion of expected effects when actual core-rim domain sizes or concentrations differ substantially. Why was 50 ppm chosen as the minimum eU concentration for non-enriched zones?
- I am confused between outcomes B and C. Is it simply that B’s hypothetical grains are closer to the outlier data in comparison to C? Do you recommend changing just the eU for B? Is there a simple rule to get the hypothetical grains to move vertically in age-eU space to better match the outlier (i.e., increase the rim enrichment if the grains need to move down or left)? Do you recommend changing the eU for B and/or C first, or also changing Rs? Does Rs have to be changed too?
- Additionally, this “zoning” occurs exclusively on the exterior of the grains, so more discussion of stopping distances, He implantation, and FT corrections is needed. In addition to Hourigan et al. (2005), it would be prudent to cite and discuss Orme et al. (2015): https://doi.org/10.1002/2015GC005818. In fact, the ZHe dataset in that paper might be a nice comparative dataset to the current Madison Range, Montana, as the Orme et al. data does not have protracted cooling. Providing another dataset that this technique successfully reproduces with core-rim relationships would substantially strengthen the approach's broader utility and robustness.
- Equations (1) and (2) would benefit from a brief derivation. Readers would benefit from a short explanation of how these core-rim U relationships were created.
- NCZ acronym is not defined.
Figure comments:
Fig 2. Hard to read HeFTy screenshot.
If allowed, Figure A1 should be a main figure in the technical note, not relegated to the supplement.
Figure A1 outcomes: Could benefit from a sketch of what each outcome would look like in age-eU space.
Thank you for the opportunity to review this manuscript. It presents a valuable contribution to understanding complex core-rim relationships in ZHe datasets.
Citation: https://doi.org/10.5194/egusphere-2026-1672-RC2 -
AC2: 'Reply on RC2', Jenna Kaempfer, 11 Sep 2026
Responses to reviewer comments are italicized
Kaempfer presents a reverse-engineered model of zoned zircon (U-Th)/He thermochronology. The manuscript addresses a genuine problem that is well laid out and explained. Current diffusive models do not account for internal chemical zonation, including radiogenic daughter products. Other complexities, such as radiation damage, modify the He diffusion kinetics, which can enable a more detailed time-temperature path interpretation. The significant problem occurs for FT corrections only when zonation causes most eU to be concentrated either within/outside of 15 μm of the rim. So, this issue is most prevalent in zircon grains with metamorphic overgrowths that can cause large overdispersion.
The manuscript currently remains largely conceptual and qualitative, which may be the right breadth/scope for a technical note in GChron. Greater discussion of assumptions, limitations, parameter sensitivity, and the non-unique nature of interpreting date-eU scatter would make this technical note stronger. With a few key revisions, this manuscript is suitable for publication in GChron. Outlined below are several points that would bolster the readership, support interpretations more strongly, and streamline the flow.
The author is grateful for this reviewer’s thoughtful comments and suggestions. Comments are addressed individually below.
- The central claim is that overlap between forward-modeled hypothetical zoned grains and observed outlier populations can be used to identify likely zoned zircon. However, the manuscript currently lacks validation using zircon grains whose zoning has been characterized independently (e.g., CL imaging, LA-ICP-MS mapping or depth profiling, Raman damage mapping). However, I do understand the utility of forward modeling in this case, as independent characterization is not part of traditional workflows, and this model could be used to interrogate previously published data. Also, traditional workflows are destructive, so there is no easy or cheap way to include independent characterization without mounting the grain, then plucking and dissolving the half-grain, which would introduce some error. It is worth noting in the manuscript that in situ thermochronology techniques can map U and He concentrations (e.g., Danišík et al., 2017; Pickering et al., 2020; Vermeesch et al., 2012). Depth profiling unpolished grains on double-sided sticky tape, which can then be plucked and dissolved as whole grains for (U-Th)/He, is another way to assess core-rim U relationships (e.g., Orme et al., 2015; Pujols and Stockli, 2021). Perhaps the manuscript needs more explicit guidance that this modeling technique is not conclusive and should be used alongside other evidence that the grain is zoned. At present, Figure 3 demonstrates only that a hypothetical zonation scenario can reproduce the behavior of some outliers in a published dataset. This is suggestive but does not independently demonstrate that the outliers are truly zoned grains.
Thank you for these robust comments and suggestions. You are correct that the intended utility of this method is to retroactively screen outliers in published data. The author agrees, individually characterized zoned grains would validate the method’s ability to reproduce outlier distribution due to zonation within a dataset. The author and colleagues are currently building a dataset of characterized zoned grains to support and validate the protocol proposed here, however the author plans to include characterized data from Orme et al. (2015) for validation following a suggestion made below.
- The workflow requires a "best-fit" time-temperature path that already reproduces the inlier data. This is an important assumption that warrants further discussion, especially because many datasets showing significant date dispersion are those where thermal histories remain poorly constrained. Discuss how uncertainty in thermal history propagates into predictions of zoned-grain behavior. Would different acceptable thermal histories produce similar hypothetical date-eU arrays? A brief sensitivity analysis would substantially strengthen the manuscript.
This is a very good point, thank you for this comment. The need for a “best-fit” time-temperature path is necessary to relate the hypothetical zoned grains to the dataset’s unique thermal history. A figure detailing how hypothetical data in date-eU space respond to time-temperature path adjustments in date-eU space will be included in the revised manuscript. Additionally, the revised technical note will include statistical testing to determine how well hypothetical data compare to sample outliers, following a suggestion from Reviewer 1.
- A clearer explanation of the hypothetical grain parameters that are iteratively adjusted to match real data should be moved to the abstract and placed earlier in the discussion. This is the key process that drives the forward model. I suggest providing evidence that these values are representative of common ZHe datasets, examples from published zircon populations, and discussion of expected effects when actual core-rim domain sizes or concentrations differ substantially. Why was 50 ppm chosen as the minimum eU concentration for non-enriched zones?
Thank you for this suggestion to present hypothetical grain parameters alongside measured grain parameters for direct comparison. The parameters of the hypothetical grains results from a brief survey of samples with variable formation and thermal histories (Moser et al., 2021- Mecca Hills; Kaempfer et al., 2021- Archean Wyoming Craton; Armstrong et al., 2024- Punchbowl Formation), though this is not explicitly stated. This includes setting the minimum eU of non-enriched zones at 50 ppm, which was just below the lowest eU value identified in the above datasets. The revised technical note will include a broader discussion of grain parameter selection. Additionally, the inclusion of figure A1 alongside a step-by-step explanation of iterative testing will be included.
- I am confused between outcomes B and C. Is it simply that B’s hypothetical grains are closer to the outlier data in comparison to C? Do you recommend changing just the eU for B? Is there a simple rule to get the hypothetical grains to move vertically in age-eU space to better match the outlier (i.e., increase the rim enrichment if the grains need to move down or left)? Do you recommend changing the eU for B and/or C first, or also changing Rs? Does Rs have to be changed too?
Yes, proximity of sample data to hypothetical data is different in outcomes B and C. In light of your questions, these outcomes will be grouped together as there is no practical difference in the treatment of these two outcomes. There are too many variables to decisively say that changing a given grain parameter will change the position of hypothetical data in date-eU space, given the complexity of the system and the impact of unique thermal histories. For example, changing the Rs of a hypothetical grain with a monotonic cooling history will have a different effect than changing the Rs of a hypothetical grain with a history of multiple re-heating and annealing events. The exception is adjusting the grain bulk eU value, which changes data positions along the x-axis. Generally, the Rs of hypothetical data should represent the Rs of sample data for the first iteration of testing, and eU should be set to mimic the eU of outlier populations within the sample data. An expanded discussion of how and when to adjust parameters will be included in the revised manuscript.
- Additionally, this “zoning” occurs exclusively on the exterior of the grains, so more discussion of stopping distances, He implantation, and FT corrections is needed. In addition to Hourigan et al. (2005), it would be prudent to cite and discuss Orme et al. (2015): https://doi.org/10.1002/2015GC005818. In fact, the ZHe dataset in that paper might be a nice comparative dataset to the current Madison Range, Montana, as the Orme et al. data does not have protracted cooling. Providing another dataset that this technique successfully reproduces with core-rim relationships would substantially strengthen the approach's broader utility and robustness.
The author agrees that further descriptions of stopping distances and zone width would strengthen the proposed methodology. Thank you for your excellent suggestion to include Orme et al. (2015) as a comparative dataset, as this data includes characterized zoned grains allowing for independent verification of hypothetical grains in date-eU space.
- Equations (1) and (2) would benefit from a brief derivation. Readers would benefit from a short explanation of how these core-rim U relationships were created.
Noted. An expanded discussion of these equations will be included in the Appendix in the revised technical note.
- NCZ acronym is not defined.
Thank you for catching this oversight. An explanation will be included in the revised technical note.
Figure comments:
Fig 2. Hard to read HeFTy screenshot.
Noted. The quality of this figure will be addressed during revision.
If allowed, Figure A1 should be a main figure in the technical note, not relegated to the supplement.
Figure A1 outcomes: Could benefit from a sketch of what each outcome would look like in age-eU space.
Thank you for these comments on Figure A1. The author will include examples of the distribution of zoned outliers for each of the listed outcomes. This figure will also be moved to the Methodology section.
Thank you for the opportunity to review this manuscript. It presents a valuable contribution to understanding complex core-rim relationships in ZHe datasets.
Citation: https://doi.org/10.5194/egusphere-2026-1672-AC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 246 | 72 | 30 | 348 | 26 | 29 | 25 |
- HTML: 246
- PDF: 72
- XML: 30
- Total: 348
- Supplement: 26
- BibTeX: 29
- EndNote: 25
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
This work attempts to explain the dispersion present within a selection of highly dispersed samples. By generating a series of synthetic samples with different zoning characteristics and predicting ages using forward modelling. These new data are shown to overlap with measured dates as a means of showing than zoning may have played a significant role in the dispersion of grains. The workflow presented is understandable and can be seen as a simple way of screening grains to better understand dispersion and extract some thermal history information. However, in its current form, this article feels underdeveloped and highly localized to just this small collection of samples. Several inclusions could be made to make this more impactful and to ensure this approach is more broadly applicable.
Scientific significance:
I feel this work does not offer a ‘substantial’ contribution to geochronology as its limited scope within a small collection of samples from the Northern Madison Range in Montana. Testing this method on a number of other cases studies would be more convincing, which would possibly allow for more zoning combinations to be attempted. From my understanding the data need to test this workflow would be readily available in any publication of zircon (U-Th-Sm)/He dates and could warrant a wider selection of samples to be tested.
Additionally, could the newly acquired information about zoning be then reincorporated into an inverse model that shows how this work can help explore the model space better? This thermal history approach feels like it only is capable of forward modelling based on a given thermal history, and it use in inverse modelling could be very beneficial.
Scientific quality:
The work is well written, though a deeper analysis of the results is merited. In its current form the comparison between observed and modelled data appears to be visually based (overlapping in plots). A more robust statistical test that actually compares this dataset would help show how well these modelled data are. Some comparisons could help better fine tune forward modelling and find a better fit for zone/rim ratios.
Presentation quality:
Though not necessary for publication, the quality of figures and inclusion of perhaps an additional one would benefit this work. I do feel the plotting of data could be improved (perhaps coloring samples point by eU would benefit). Additionally, a conceptional figure showing a zircon with enriched cores or rims would perhaps help non-specialists.