the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 21 Jul 2026
Carbon Cycle Perturbance During the Carnian Pluvial Episode (Larne Basin, Northern Ireland)
Abstract. The Late Triassic Carnian Pluvial Episode (CPE, ~232 Ma) marks a major global climatic transition characterized by intensification of humid conditions following a prolonged period of aridity through much of the Triassic. This episode triggered profound environmental and biotic change that ushered in modern environments and ecosystems. Here, we present a new high-resolution integrated geochemical and sedimentological record from the Knocksoghey Formation, Mercia Mudstone Group, in the Larne Basin, Northern Ireland. The high-resolution organic carbon isotopes document a ~4 ‰ negative excursion, whose onset coincides with a small, reproducible rise in Hg concentrations from background levels, thought to be associated with the emplacement of the Wrangellia Large Igneous Province (LIP). The geochemical anomalies stratigraphically closely precede a pulse of coarse siliciclastics, indicative of intensified weathering driven by enhanced rainfall, and are followed by a return to arid, restricted conditions, marked by deposition of stacked anhydrite nodules and beds. The CPE was previously suggested to be characterized by a succession of negative carbon isotope excursions (NCIEs) in sedimentary archives. However, our new data demonstrates a more protracted stepped carbon cycle perturbance, as reflected by a single, stratigraphically prolonged carbon isotope excursion, comparable in nature to NCIEs observed during other LIP-driven climatic perturbations, such as the Toarcian Oceanic Anoxic Event.
- Preprint
(1721 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 18 Sep 2026)
- RC1: 'Comment on egusphere-2026-2964', Anonymous Referee #1, 24 Jul 2026 reply
-
RC2: 'Comment on egusphere-2026-2964', Clemens Vinzenz Ullmann, 18 Aug 2026
reply
Marwa Mohamed Shahid and co-authors present a largely geochemical multi-proxy dataset of an interval of Upper Triassic strata from the Carnduff-1 core, identified by the authors as a record of the Carnian Pluvial Episode (CPE). The authors argue that these data show that - contrary to previously published accounts - the CPE was characterised by a stepped singular carbon isotope excursion, that Hg enrichment in the core is consistent with the perturbation being LIP-induced, and that weathering indices in the core and shape of the C isotope excursion as well as known biological turnover establish strong genetic similarity with other carbon isotope excursions that are linked with LIP volcanism, notably the Toarcian Oceanic Anoxic Event.
The authors - in my opinion - present a valuable dataset from an interesting location, supplemented further by some additional Hg data from a section in China, and therefore I think there is definitely merit in publication of these data. However, I equally think there are numerous aspects for this study that would benefit from further work to make the contribution as impactful as possible.
Identification and character of the CPE
I am not an expert of the CPE and I hope another reviewer with stronger research background on Triassic stratigraphy will be able to comment on some of the findings with more authority than I can. However, I feel that the authors are a bit too optimistic about the quality of their core record by making statements that would have far-reaching consequences for other CPE sections.
- Given that the authors present only relatively loose stratigraphic constraints for a succession in which I think stratigraphic completeness and consistent sedimentation rates are unlikely, how robust is the assertion that Carnduff-1 contains a superior CPE record that allows for a re-evaluation of the nature of the CPE carbon isotope excursion? Despite the large-scale changes in lithology, can changes in sedimentation rate that affect the shape of any isotope excursion or chemostratigraphic trend be ruled out, or at least constrained to some degree?
- If the authors are right in the proposal of a single, drawn-out, stepped carbon isotope excursion, then how are the other published accounts to be re-interpreted that note multiple isotopic perturbations separated by returns to pre-excursion values? Are they all affected by stratigraphic repetition, peculiar organic matter mixing, etc.? This seems rather an unlikely hypothesis to me given the alternative that Carnduff-1 is not as continuous and simple as the authors present it to be. In any case the authors do not elaborate on any of the issues that would need to affect other known sections to allow for the interpretation of a single isotope excursion to be correct, and I think this side of the story needs to be looked at as well to arrive at a plausible interpretation.
- The authors consistently assert that the CPE in Carnduff-1 is characterised by a stepped isotope excursion, but I really cannot see this in the presented data. The raw data is quite noisy, the CPE interval is rarely highlighted in the plots in detail, and the authors themselves use smoothing lines of the data (Fig. 4) that obscure any possible existence of steps. Can some information be added how these steps were identified and how robust they are?
- The authors note (based on the putative stepped nature of the CPE C isotope excursion) a strong similarity to the Toarcian OAE, and by extension other LIP-induced carbon isotope excursions. How similar to the T-OAE is this isotope excursion really, and how representative of LIP-induced carbon isotope excursions is the T-OAE itself? The authors make it sound as if multiple other such carbon isotope excursions could be taken for comparison, but even the discussion of the similarity to the T-OAE receives very little room. What other such stepped, symmetric negative CIEs could one compare to?
Geochemical analysis
The authors obtained valuable data, but these datasets have analytical issues and the lithologies of the core impose strong controls on what evaluations are possible based on these data.
- The authors note that analytical data will be made available upon publication of the study, which is great. However, I have been unable to access the dataset on Zenodo so cannot make a detailed assessment of the proxies myself for this review. In particular I would have liked to look at the Cl, S and trace element data in some more detail. I think that in a second round of reviews these data should be made available (to reviewers at least) so that potential issues not currently visible can be recognized and it can be judged if these data are formatted and presented in a way that is clear and useful to the user.
- Very little information about core condition and mineralogy is provided, so it is not clear if any alteration of the core has taken place after it was drilled. The authors refer to evaporite minerals such as halite and anhydrite, both of which would have been exposed to potentially high humidity if the core was not stored in a climate-controlled environment. Might the mineralogy have been affected since coring? Did the authors assess the presence of pore fluids in the core? Other cores of Mercia Mudstone contain large amounts of saline pore fluid that bleed out and cause formation of salt crusts. This can obscure the true mineralogy of the strata. Was this assessed by the authors?
- The authors present some methodological detail for their work, but some key information is missing or incomplete.
For XRF assessment, the authors give standard deviations for major element oxides obtained for NIST 2711a. While this is useful, it is not mentioned how many measurements of NIST 2711a this is based on (potentially about ten based on the number of samples analysed?), and it is unclear how these uncertainties relate to analytical uncertainty of the samples which largely are geochemically distinctly different to NIST 2711a. Are these uncertainties thought to be absolute uncertainties independent of concentration or are they thought to scale with concentration (i.e., would it be better to report relative standard deviations)? What are the relevant detection limits for the elements, how were detection limits identified, and are all the presented data above the limit of detection? How can the authors demonstrate that the data are accurate despite widely variable geochemical matrices? The authors use Rb, Sr, and Zr data for further interpretation but do not give any quality information for these data. Were these elements also measured via XRF using the described technique or derived differently? What are accuracy, precision and detection limits?
For C-org and TOC analysis, the authors present some information about reference materials used for correction of their measurements, but there is no reference to certified TOC content of EMA-P2 and therefore any information how to interpret the uncertainty of the TOC data. I gather that a single point calibration was carried out to shift isotope data of unknown samples based on L-Alanine, but was stretching of the isotopic scale due to instrumental fractionation taken into consideration? The analysed samples are geochemically very different to the standards. The standards are pure chemicals with very high TOC, while the samples have very low TOC and required the analysis of very large samples. This might mean that additional issues such as incomplete combustion or retention of some CO2 inside the sample may have affected the samples which would not be visible by testing only pure organic substances as standards. Can these issues be ruled out? The presented uncertainty of the EMA-P2 standard is already very large (0.7 permil 2 s.d.), and this seems to be an optimistic value regarding the core samples. Have any repeat analyses of core samples been carried out to see if they give a comparable repeatability? How should one translate the uncertainty of TOC in EMA-P2 to a similar uncertainty for the samples? Is this uncertainty largely driven by weighing uncertainty and therefore much lower for core samples than for the smaller standard samples, or is this related to instrumental instability? Are the reported TOC data for acid-treated material or for original rock samples? In the latter case, how was loss of material during acid treatment dealt with?
For Hg analysis the authors again present some useful information, but it is not entirely clear how it relates in detail to precision, accuracy, and detection limit of the analysis. Some of it may relate to the lack of identification of statistical measures employed and unclear terminology. Please add information what the presented uncertainties for Apple Leaves and Ryegrass refer to. Also, please state what the internal standard is and why there seem to be differences in reproducibility of samples, internal standard (what is it) and standard plant materials. From the figures and text it becomes clear that barring very few exceptions the Hg concentrations in the samples are very low. What is the detection limit of the technique and therefore what would approximate error bars for the samples be and are they actually above the limit of detection/quantification?
Presentation of the data and use of proxies
I noted that none of the figures includes any error bars, not even a representative uncertainty. It is therefore not possible to judge if patterns in the data relate to analytical noise, or if they are interpretable as geological features. Please add visual representation of uncertainty and (where relevant such as likely Hg) detection limit into the geochemical plots.
A wide suite of calculated weathering indices is employed in the study, none of which is introduced in any detail. It is not clear from the text how these are calculated and references are not included. Besides this lack of introduction it is also later noted by the authors themselves that the majority of these weathering proxies is compromised, particularly by carriers of Na and Ca. Little is mentioned about mineralogy of the core beyond halite and anhydrite (is calcite present above trace quantities?), but these seem to be quantitatively important and therefore the chemical weathering indices cannot be meaningfully used. While the authors acknowledge the issues with the proxies, this happens very late in the text (L264 and later), after an already extended description that reads as if the authors treat these data a genuinely reflective of weathering patterns. Even for the least compromised part of the succession it remains unclear how meaningful the weathering proxies really are.
Similarly the authors use Hg normalisation via TOC, while (very late in the text once more) acknowledging that this is against recommendations and likely not reliable. The issue of potential post-depositional degradation of TOC is not considered here either. Are TOC values considered trustworthy and primary, or how much TOC may have been lost? The authors note the presence of reduction spots which indicates chemical interaction of iron with a reducing substance such as organic carbon?
I would encourage a re-write of the text entirely removing the use of weathering proxies. In my mind even if they were applicable for use in the more sandy intervals there would not be any other part of the core to compare these values against. It may be preferable to instead reflect on mineralogy and maturity of the sands and how this relates to other known expressions of the CPE?
Overall assessment
To me, the presented sample look to provide a valuable yet somewhat compromised multiproxy dataset that allows to draw comparison to other CPE records, but I would not go as far as saying that it is of the exceptional quality required to revise entirely the nature of the CPE carbon isotope excursion(s). The conclusions the authors come to really are a little bit too bold for me to comfortably agree with. In my opinion it would be more appropriate to critically reflect on the limitations of the data more widely (nature of the lithology, breaks and variability in sedimentation) and put some additional work into a nuanced discussion with more comprehensive citation of relevant literature. In such a revised form I think the study would be quite suitable for Climate of the Past.
Detailed comments
L22: How are coarse siliciclastics indicative of intensified weathering? I would say that they may indicate enhanced erosion, unless they simply relate to a shift in the sediment transport pathways and reflect a river channel. It may be true that added rainfall would also have the capacity to lead to more weathering, but would this cause an increase in weathering intensity, or rather an increase in weathering flux via exporting a lot more detritus that is, however, less strongly weathered?
L23: I do not think that “deposition” is the best word for this.
L25: I do not think that this has been demonstrated convincingly. Please revise with more evidence, or rephrase.
L26: Please include more examples than the T-OAE, even though it may be enough to discuss one in detail.
L30: Is “wet” much different to “humid”?
L32: I would rephrase this to tie better with the following sentences and explain more what exactly these biotic changes were.
L34: The sequence of references appears to be incorrect here.
L37: There is an extra space before “The” here that can be deleted.
L45: What is the source of the palaeogeographic reconstruction? Please provide reference. Annotations on blue background are quite hard to read and seem a bit eclectic. Should it be “Pindos Volcanics” and “Orogenic Belt”?
L73: “strata” rather than “stratigraphy”?
L80: Is the Westbury Formation really consistently marine?
L84-86: This may be useful information but I am not fully sure how it integrates into the flow of the text here.
L100: In such an environment, how can one assume continuous sedimentation with mostly uniform sedimentation rates?
L106: What about syn-sedimentary faulting and subsidence?
L111: Why are three points in time noted here? Is the top of the Mercia Mudstone that diachronous?
L113: For what reasons are they considered equivalent, and why is this relevant here?
L128: “10-12” tonnes are a weight range, not a pressure. To derive the pressure one needs to normalise by the area this weight is applied to. Please update accordingly.
L135: Is carbonate known from this interval in Carnduff-1? How was the acid replaced? How much acid was used for how much rock powder?
L136: How was the rinsing carried out? How was sample loss avoided?
L143: The notation is odd here - please use super- and subscript for the organic carbon isotope description.
L148: In general it would be good to state precision of the weighing step (also for isotope work).
L150: What does the stated uncertainty relate to? Please consider using “ng/g” instead of ppb. Parts per billion is an ambiguous unit that could represent e.g., a molar fraction. “ng/g” cannot be misinterpreted.
L152: What concentration was found, and how many analyses were carried out for this certified reference material?
L155: It is great that duplicates were analysed, but what samples were duplicated? Most samples seem to have Hg concentrations much below 1 ng/g, so could not give a relative standard deviation of 2 % if the absolute standard deviation is 0.2 ng/g. Also, please consider the number of significant digits presented according to the relevant uncertainty. It appears odd that the typical difference between two duplicates is “roughly 2.16 %”. 2.16 % is a very precise value.
L158: I am wondering why materials give a standard deviation of 0.2 (L156), 0.4 (158), 0.6 (L152), and 0.8 (L158) ng/g. Are some of the materials heterogeneous? Does the standard deviation scale with concentration? Which uncertainty is most useful for the consideration of the samples? How was accuracy demonstrated?
L181: Some of the used font is tiny - so small that even when zooming in to 200 % I still struggle reading it. Please re-size ideally to at least 6pt in print size so that the figure can be easily read. Also, the codes for lithology are not easily visible on the log.
L186: Please highlight the “steps” in some way that makes them visible. I cannot see any clear steps.
L192: Why is this information presented again? It is already noted in L126.
L197: The high Na needs some explaining. Is it related to sedimentary halite or at least partly to pore fluids? Is the Na/Cl ratio approximately 1 mol/mol? Why is Na expressed as Na2O?
L205: None of these indices and abbreviations has been introduced, and there is no justification why they can be used here. A value of 13 (L206) is quite clear evidence that these lithologies are unsuitable for use of the CIA.
L219: Can a C isotope value be reliably be measured for samples with 0.02 wt%?
L220: This information is redundant. It is already noted in L138.
L222: How is the average of these values important or meaningful for this study?
L226: Is this actually measurable? Based on the presented analytical quality, the detection limit of the technique is likely 0.4 ng/g or even higher.
L227: What is a “background Hg concentration” - Is that the average of all samples that are not part of the two brief spikes? Why is it presented so precisely?
L230: Consider the uncertainty of the measurements. Conservatively, the measurement uncertainty of Hg is +/- 0.4 ng/g and the uncertainty of the TOC data has to be integrated into Hg/TOC as well. At a relative uncertainty of 6-7 % a value of 69 +/- 4 would be derived.
L233: See above; 11 ng/g is probably precise enough. Please check through all of the text, also L234, 236 and other places for precision of reporting.
L239: “Lithological evolution”?
L240: Does this refer to Northern Ireland? It was initially identified earlier.
L244: The sentence construction could be improved here (second which after L 243). It is not clear what “which” is now.
L250: Why “indeed” and what follows from that?
L252: I agree that there is a plausible climate interpretation, but are there no alternative viable hypotheses?
L254: See comments above - I do not think that the weathering index information is helpful here.
L260: It seems a bit late here to introduce this information, especially as the depositional environment does not allow for the use of such indices.
L271: Up to, or down to?
L272: How reliable is this assumption? What are the Cl readings for this sample?
L286: It seems that some work on thin sections would be more powerful here than interpretations of complex chemical signals?
L287: As above coarse-grained sands do not really point at increased weathering in my mind. At least not directly.
L293: I am not sure how this conclusion was reached. I would expect that the CIA of the clay fraction in the lower and upper interval would be much higher than the sandy interval, instead suggesting that this material is more weathered. Whether it indicates stronger weathering, however, would also then be unclear because of the strong lithological contrast and the lack of equivalent clay material from soils from which the sands derive.
L296: There seems to be an element of circular reasoning here. The C isotope curve is at least one of the tools to identify the CPE, so one should not then treat it as independent.
L297: Is “pluvial” not another word for “wetter”?
L301: I do not follow the logic here. What exactly are Si/Al, Zr/Al and (Rb+Zr/Sr) taken to be proxies for here? If they simply relate to terrigenous material, then they cannot be used, because as the authors state the entire succession is terrestrial and thus only contains terrigenous material.
L303: This may be true, but why and how is this relevant in the greater scheme of things? It seems a bit odd to just refer to one other core record.
L320: Please provide evidence for the “stepped” nature and that there is only one excursion.
L326: I also cannot see a 3 permil preceding positive excursion. The smoothed line in Fig. 4 shows at best 2 permil.
L330: “comprise multiple”
L331: It may be an expanded record, but is it more expanded than others, and does “expanded” necessarily equate to “complete”?
L333: Consider adding additional examples.
L344-7: Please provide references.
L349: A global temperature increase of 7-10 degrees is unrealistic in my opinion, and I do not think that the data presented in the cited papers really support such a magnitude.
L351: I think this paragraph is too short and unspecific to act as a detailed comparison to the T-OAE. Please elaborate further.
L362: CO2 does not directly lead to rainfall, though, I would think?
L364: Are there “non-fluvial” lakes?
L382: It may be worth also mentioning some of the complexity, however. There are also instances where the T-OAE has been linked with reduced TOC, and multiple marine epicontinental basins do not show clear evidence for anoxia.
L390: I think this figure would require some work to show that there are isotope steps, and how these compare to the T-OAE (and other CIEs). The logs are presented in the depth domain, but the arguments are in the time domain, so it appears the authors propose that no sedimentation rate changes occur in these successions? This appears quite unlikely.
L396: Which “other events” could be pointed out here?
L400: As noted above, I really cannot see a 3-4 positive excursion in the Carnduff-1 record.
L401: Also, as noted above, how reliable is such an assessment of symmetry? Besides, the Carnduff-1 record does not look to be entirely symmetrical even in the depth domain. The increasing limb is shorter and values do not return to background entirely.
L402: Please provide references for gravity flows in Mochras and ideally further afield.
L408: There seems to be an issue with the “13” in the delta value.
L414: It should read “Ruhl et al., 2020, 2022” I think?
L420: What is the meaning of “assessed Hg/TOC”
L423: I would recommend stating this limitation of low TOC much earlier and treat the data accordingly from the beginning.
L429: Can this really be firmly established? There is (very) little Hg even in these samples. Could any redistribution have occurred that juxtaposed a Hg spike at a lithological boundary?
L432: These R values are indeed low, but are only valid for interpretation if there is actual measurable variability that relates to primary concentrations. If there is TOC degradation, a primary TOC correlation may be obscured. Also, the error bars of the Hg measurements are likely on average larger than the values themselves; in such a scenario it would be very tricky to detect any significant correlation?
L434: Would this environment be conducive of pyrite formation? These are largely continental red beds and evaporites formed in an oxidising environment? Is there any covariation with the sediment colour?
L436: In red sediments, this seems to me to be a risky assumption. See also L440.
L445: It would be useful to annotate this figure with some information where the CPE is located in each chemostratigraphic profile.
L451: Why was Luojiagou selected? Please explain.
L453: I also cannot see any clear steps in this record.
L457: The onset of which CIE? There is a negative CIE around 20 m which does not show Hg enrichment.
L462: Enhanced with respect to what?
L472: I am not convinced that Carnduff-1 is the best record to show this. A more balanced discussion of other existing records would be needed as well, I think.
L473-475: This sentence really rather points at major changes in sedimentation rate - a topic not taken up and considered in the text.
L483: This last paragraph seems to be shoehorned into the text a little. Is this added because of the title of the funding research grant?
L489: Is precipitation not one of the variables measured by “climate”?
Figure A1: why relative to “carbonate”? Carbonate seems to be subordinate at best in these strata?
Citation: https://doi.org/10.5194/egusphere-2026-2964-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 143 | 52 | 10 | 205 | 13 | 16 |
- HTML: 143
- PDF: 52
- XML: 10
- Total: 205
- BibTeX: 13
- EndNote: 16
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Publisher’s note: a supplement was added to this comment on 27 July 2026.
This study examines the record of the Carnian Pluvial Event (CPE) in a drillcore through continental facies in northern Ireland. The main goal is to evaluate the relationship of the CPE to volcanic activity (proxied by Hg concentrations), which is assumed to be associated with the Wrangellia LIP.
Summary: There are many problems, both large and small, with this study that preclude publication in its present form. It can either be rejected or assigned “major revision followed by re-review”, depending on the editor’s preference.
Weathering proxy calculations: In looking at Figure 3, I realized that the calculated CIA values are completely in error. CIA values begin around 40-50 for fresh (unweathered) igneous rocks and go to 100 for highly weathered sediments. All of the values here are <25, which probably means that carbonate CaO has not been subtracted out. CIA is a proxy for silicate weathering intensity, so CaO associated with carbonates and phosphates must first be removed, yielding a silicate CaO value (designated CaO*). The authors seem completely unaware of this (while the first two authors may be students, why didn’t any of the more senior authors further down the authorship list catch such a serious problem??). The same thing applies to the CIW values.
Significance of volcanic proxy evidence: The authors document modest increases in Hg concentrations in the study core, but the significance of these data is uncertain. Rather than volcanic inputs, higher Hg concentrations might have resulted from redox changes (e.g., a shift to more reducing conditions, e.g., Shen et al. 2019) or decreased sedimentation rates (e.g., Liu and Algeo, 2020). Demonstrating that the increased Hg concentrations were due to enhanced volcanic Hg fluxes generally requires paired Hg isotope data, with a shift to near-zero values of D199Hg reflecting volcanic inputs.
Shen, J., Algeo, T.J., Chen, J., Planavsky, N.J., Feng, Q., Yu, J. and Liu, J., 2019. Mercury in marine Ordovician/Silurian boundary sections of South China is sulfide-hosted and non-volcanic in origin. Earth and Planetary Science Letters, 511, pp.130-140.
Liu, J. and Algeo, T.J., 2020. Beyond redox: Control of trace-metal enrichment in anoxic marine facies by watermass chemistry and sedimentation rate. Geochimica et Cosmochimica Acta, 287, pp.296-317.
Also, even if the increased Hg concentrations represent volcanic inputs, there is no evidence showing that the volcanic Hg came from the Wrangellia LIP—this is merely a speculation. Speculation is okay but it must be acknowledged as such in the text.
The Introduction is full of factual misstatements—I identified four errors on lines 34-42.
Project goals--What are the study goals? They need to be articulated in the final paragraph of the Introduction. The text is just stating what the authors did, not what the project goals were.
Figure 1—the Wrangellia LIP is shown as a huge feature, far larger than any modern continent. The actual size might have been ~20% of what is shown here.
Other figures—nearly all of the figures need various improvements per the comments that I have added to them.
Language—there are numerous errors in verb tense, subject-verb agreement, word choice, grammar and other issues. I have pointed some of them out with comments, but one of the native-English-speaking coauthors needs to go over the manuscript carefully and correct all of the language errors.