the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 14 Nov 2024
A comparison between in situ monazite Lu–Hf and U–Pb geochronology
Abstract. In complex metamorphic terranes, monazite U–Th–Pb dates can span a wide concordant range, leading to ambiguous geological interpretations (e.g., slow protracted cooling versus multiphase growth). We present in situ monazite Lu–Hf analysis as an independent chronometer to verify U–Th–Pb age interpretations. Monazite Lu–Hf dates were attained via laser ablation inductively coupled plasma mass spectrometry equipped with collision/reaction cell technology (LA-ICP-MS/MS). In situ Lu–Hf dates for potential reference monazites with uncertainties < 1.6 % agree with published U–Th–Pb dates, validating the approach. We demonstrate the method on complex metamorphic samples from the Arkaroola region of the northern Flinders Ranges, South Australia, which exhibit protracted thermal and monazite growth histories due to high geothermal gradient metamorphism. In situ Lu–Hf dates reproduce the main U–Pb monazite age populations, demonstrating the ability to reliably resolve multiple age populations from polymetamorphic monazite samples.
- Preprint
(1966 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on gchron-2024-29', Nick Roberts, 08 Jan 2025
Nice paper, well presented, with a useful and clear case study.
There is already a paper on this method. This has been cited, and uses different instrumentation. The case study described here is a nice addition.
There is no comment on the fact that older monazites will be easier to date (in terms of measurable radiogenic Hf that is). How young will this method be useful for using this instrumentation? That is concentration specific of course, but clearly unpicking the Alpine-Himalayan orogen is not going to be easy.
It is a shame that the study does not include any of the most commonly used monazite RMs, e.g. Stern, 44069, Manangotry, Moacyr/Bananeira.
Error propagation: at this stage, this is all that can be done, and covers the basics of the calculations. A comment on the fact that long-term reproducibility is not accounted for, and may also be a large contributor to the total age uncertainty would be prudent.
The paper relies on its predecessors to describe common Hf, Yb corrections etc. I am not suggestion repetition, but perhaps comments on the key issues and important considerations would be useful.
Figures – I didn’t see it stated that bars/ellipses are 2sigma.
Data – The tables should comprise mass spectrometer signals for at least some of the measurements, as per widely shared recommendations for U-Pb. The decimal places are too many for the ratios.
Line 30 – ‘orogens’ would be more accurate that ‘terranes’
Line 44 – It is not clear how the two listed ‘problems’ with the approach of Wu et al, “hinder exploring the application of Lu-Hf monazite to its full potential”.
Line 56 – Were spot sizes mixed during each session? Were they mixed between samples and RMs, and does this matter? If not, then this needs to be demonstrated. Different spot sizes will change the downhole fractionation patterns, but it is understandable on a quadropole instrument that the data may be too imprecise to measure any difference accurately.
Line 57 – The reality of this method, is that this spot size is commonly larger or similar to the total length of metamorphic monazites found in typical pelites metamorphosed at mid-crustal conditions. U-Th-Pb spots are typically 5 to 15 microns, which is why multiple domains can be dated from single grains.
Line 74 – Indistinguishable – using this method and instrumentation. This does not mean that the ideology can be applied to all minerals, all instruments and all conditions. A matrix-matched approach should always be strived for, even if this is not possible at first. Non-matrix-matching allows for poor ‘traceability’ of the method.
Line 80 – Were trace elements checked against any published monazite data, or are they only considered to be non-quantitative? Were the Ce contents measured with EMPA, or just assumed? What Ce value was used if assumed?
Line 268 - I personally would not call these established – but they have published ID U-Pb data, and that is the point.
Line 269 – Perhaps also list the accuracy in terms of %, i.e. “accurate to <1%”.
Line 288 – True, but many labs only work with Phanerozoic monazite RMs – and it is unclear how the accuracy will degrade with younger and younger samples/RMs with this method.
Line 295 – Yes, indeed. I would also make the point that the companion trace element data are critical to unpick multiple populations.
Line 300 - missing a “to”
Citation: https://doi.org/10.5194/gchron-2024-29-RC1 -
RC2: 'Comment on gchron-2024-29', Stephanie Walker, 03 Feb 2025
This manuscript presents an important contribution by comparing in situ monazite Lu-Hf and U-Pb geochronology in a complex metamorphic terrane. Although the specific technique has already been outlined in a previous paper, this study is well-structured, methodologically rigorous, and effectively demonstrates the utility of Lu-Hf dating as an independent chronometer for validating U-Pb age interpretations. The authors provide a strong dataset with comprehensive analytical procedures. However, I have some concerns about the handling of the uncertainties which require further clarification.
General comments
Data processing: The matrix correction for Lu-Hf dating is based on apatite reference materials. While the justification is reasonable, the authors should explicitly state whether monazite-specific correction factors were tested and how any uncertainties from matrix mismatches were handled.
Uncertainty propagation: Error propagation was undertaken involving quadratic addition of uncertainties from various sources (eg analytical session, reference material age, decay constant etc). However, there is no discussion of how systematic errors (instrumental drift and long-term reproducibility) were assessed.
I’m guessing that the reference materials were processed in the same way as the unknowns? How do they compare over multiple sessions?
Statistical handling of isochrons:
- The study appropriately employs IsoplotR for isochron calculations, but the discussion lacks sufficient depth on the selection of anchored vs. unanchored regressions.
- The authors use a fixed initial 177Hf/176Hf of 3.55 ± 0.05 for isochron regressions. While this may be appropriate, it introduces a level of model dependence that should be discussed in more detail. Were alternative initial ratios tested?
- Some of the inverse isochrons have MSWD values greater than 2. The manuscript states that this suggests prolonged fluid-rock interaction, but alternative explanations (e.g. analytical scatter, common Hf incorporation) should also be considered.
Minor comments
Figures: The figures are clear and well-labelled, but the colour scheme in Fig 4 for the different microstructural domains could be more distinct to improve readability.
References: These are comprehensive, but there are some inconsistencies in the formatting such as missing DOIs for some references.
Supplementary data: Why are there such an absurd number of decimal places for the ratios? Obviously this depends on the precision of the values, but I imagine no more than two or three are actually significant.
Recommendation
While the manuscript presents valuable and well-executed research, improvements in uncertainty handling and statistical interpretation are required prior to publication. Once the above comments are addressed, I am confident that this will be a robust foundation for future monazite studies.
Citation: https://doi.org/10.5194/gchron-2024-29-RC2
Data sets
Monazite Lu–Hf and U–Pb data Alexander T. De Vries Van Leeuwen https://doi.org/10.25909/27441327.v1
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 231 | 48 | 9 | 288 | 13 | 9 |
- HTML: 231
- PDF: 48
- XML: 9
- Total: 288
- BibTeX: 13
- EndNote: 9
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1