the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 30 Jun 2026
A Phanerozoic-style icehouse climate in the middle Ediacaran
Abstract. Geological evidence points to icehouse conditions during the Ediacaran Period (635 to 538.8 Ma) both before and during the emergence of animals in the fossil record. However, the temporal and spatial distributions of Ediacaran ice sheets are not well constrained due to uncertainties in palaeogeography, chronostratigraphy, and the depositional settings of candidate glaciogenic deposits. Here, we systematically evaluate the evidence for the depositional ages of candidate Ediacaran glaciogenic deposits and establish the likelihood that each deposit was formed by ice-driven processes using observation-driven weighting criteria. Our analysis supports the existence of discrete mid- and late Ediacaran icehouse intervals (MEIH and LEIH respectively), each followed by greenhouse conditions. Focussing on the older MEIH, we integrate our quality-controlled geological dataset with climate and ice sheet model simulations to characterise glacial conditions during this interval (~593 to 579 Ma), which encompasses the ‘Gaskiers glaciation’. Our results indicate that the MEIH was a Phanerozoic-style icehouse, with latitudinally-constrained and fluctuating ice sheets, marking a break from the preceding Cryogenian snowball Earth motif, and occurring before the first known appearance of animals in the fossil record. By mapping robustly identified glaciogenic deposits onto contrasting middle Ediacaran palaeogeographic reconstructions, we show how these deposits can be used to constrain hypotheses of Ediacaran continental configuration.
- Preprint
(1954 KB) - Metadata XML
-
Supplement
(1764 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2026-3617', Anonymous Referee #1, 27 Aug 2026
-
AC1: 'Reply on RC1', Thomas Wong Hearing, 21 Sep 2026
We are grateful for this thoughtful review of our manuscript. We response point-by-point, with our responses in bold and reviewer text in italics.
Specific comments:
94: For clarity, it would be good to specify here whether this is just reproduced or updated (updated according to the main text). It would also be helpful to describe the changes relative to the version published earlier somewhere.
This comment relates to Figure 1. This is updated from Wong Hearing et al. 2026 (now published in GSA Bulletin), with revised star ratings. All revised star ratings and revised age considerations are described in detail in the Supplementary Text and Supplementary Data files. We can make this more explicit in the main text and/or caption.
194-204: When first reading this paragraph, I had wondered whether it would be useful to show the different reconstructions somewhere, in particular the topography of the PALEOMAP reconstruction because of its importance for ice-sheet formation, and realised only later that they are shown in the Supplementary Information. It would be good to refer to the respective figure in the SI in this subsection.
Thank you for the suggestion, we will do this.
225: Please specify the present-day value used to scale the Ediacaran solar luminosity (it appears to be higher than the generally accepted value of about 1361 W/m2).
We used 1368 Wm-2 as the present-day value of solar luminosity. We can state this in the Methods. This value is slightly higher than the generally accepted present-day value but it is the standard to which FOAM was calibrated when it was developed and is therefore the appropriate value to use in FOAM-based studies.
308, subsection 3.2: I am missing a bit more discussion here, in particular concerning Fig. 3 and the plausible glaciogenic deposits deep in the tropics.
We can expand on this point in the text. There are several possible explanations. The position of the North China Craton is highly uncertain and should not be considered a reliable marker of tropical glacial deposits on its own. The South Australia deposit is a three star deposit most parsimoniously interpreted as a debris flow (and it is interpreted as such by people who have directly studied the rocks).
315-317: This is a bit confusing because the reader might interpret the "lower latitudes" in comparison to the ones in the PALEOMAP reconstruction rather than lower latitudes absolutely. I would suggest to rephrase the sentence to avoid the confusion.
We will rephrase this in the manuscript.
336-337: This is a bit confusing. This selection does obviously not include the MEIH deposits, right? This should be explained better!
We will rephrase this in the manuscript. The comment relates to Figure 3 and its caption. Figure 3 shows MEIH deposits (~593 to 579 Ma) in the top row and deposits more loosely constrained to only an ‘Ediacaran’ (635 to 538.8 Ma) age on the bottom row. We see the confusion and will clarify in the revised caption and figure.
346-348: This is too vague. You should be able to explore the mechanism based on model diagnostics.
We understand Reviewer 1’s point of view but we deliberately did not expand on this point here. Our intention is instead to only describe the results of direct interest to the reader in the context of the present study, and to refer readers interested in learning more about the mechanisms at play to previously-published, dedicated work (Pohl et al., 2014, 2016b). We remain open to editorial decision on this point, we would strongly prefer to keep the focus as it is. This is motivated by the fact that some authors have recently gone further into the analysis of the climatic mechanisms involved in this instability elsewhere, and these analyses and results are complex. They constitute a whole new study.
Improving our understanding of the climatic mechanisms at play notably required comparing climatic results obtained under similar boundary conditions using different general circulation models. Specifically, we conducted a model intercomparison based on climatic simulations run using FOAM, IPSL-CM5A2, and HadCM3. That study is currently under minor revisions at Science Advances. We hope it will be published soon, and if so, will make sure to properly refer to this new article during the proof stage. For now, a preprint was posted on HAL (https://hal.science/hal-05731743); as of September 2nd, it is still awaiting moderation and we therefore provide a temporary link to the Reviewers: https://sdrive.cnrs.fr/s/a48TCHrL4MZSkW2 (valid until January 2nd, 2027). Our inter-model comparison notably shows that this climatic instability only appears in models characterized by a weak latitudinal SST gradient, independent of model complexity. As a rapid glance at the preprint above should show, the underlying analyses are complex and lengthy, and this is why we believe that including such analyses would distract the reader from the main message of the present contribution. We nevertheless revised the text to hopefully make it clearer and more informative:
“Our simulations indicate that a sudden global climate cooling accompanied by an expansion of sea ice into the middle latitudes occurs if atmospheric pCO2 decreases from 6 PAL to 4 PAL (figs. S3-4). A similar climatic instability has previously beenreported for Late Ordovician FOAM simulations (Pohl et al., 2014, 2016b). It was attributed to the ocean heat transport simulated in the virtual absence of landmasses in the northern hemisphere, allowing sea ice to suddenly extend to the middle latitudes and climate to rapidly cool. This mechanism similarly applies to the MEIH interval, characterized by a similar continental configuration and sea-ice expansion."
351-353: Could be moved to the Methods section. Also, could you comment on whether the ice sheet is approaching equilibrium at this point?
These lines could be moved, but we think they add clarity by orienting the reader in the subsequent results section. Given that this is only 6 lines we prefer to keep them here.
Secondly, we do not know if the ice sheet was approaching equilibrium because we do not account for ice sheet-climate feedbacks. This would require a much more involved modelling approach, iterating multiple rounds of FOAM-IPSL-GRISLI simulations (following directly the approach of Pohl et al., 2016).
386: "[S]ome uncertainty" seems rather euphemistic, please delete "some".
We will do this.
400-402: Well, yes... and no... You should keep in mind that topography is usually rather uncertain as well and that high topography at low latitudes could lead to glaciation in terms of glaciers on tropical mountains.
The distribution of topographic highs likely did impact the locations of glacier development. We can revise the statement to include that “palaeogeography” includes “topography”, as well as plate position.
404-406: This cannot be the only reason since the radiative forcing from the change in CO2 is more than twice the radiative forcing from the change in the solar constant, if my back-of-the-envelope calculation is correct.
We thank Reviewer #1 for raising this point. Our own back-of-the-envelope calculations, however, suggest that changes in solar luminosity between the Late Ordovician and our Ediacaran simulations ((1320-1303.12)/4 = 4.22 W m–2) are more than enough to induce a net cooling despite a concomitant increase in pCO2 from 12 PAL to 16 PAL (=3360 ppm to 4480 ppm), which in itself represents an additional 5.35*log(4480/3360) = 1.54 W m–2 (using the formula of Myhre et al., 1998, to represent the radiative forcing associated with changing atmospheric pCO2). It follows that an atmospheric pCO2 of 26.4 PAL (~7400 ppm) would be required to obtain an Ediacaran radiative forcing equivalent to an Ordovician pCO2 of 12 PAL (using, respectively, Ediacaran and Ordovician solar luminosity values).
408: One thing your equilibrium simulations also do not take into account is that orbital forcing is continuously varying in reality. This should be discussed somewhere.
We will add a brief discussion on this point. However, as we have modelled the hardest-to-initiate glacial conditions scenario, we show that glaciation was likely to develop in these palaeogeography pCO2 conditions regardless of orbital parameters.
441: I agree. Ice-sheet models are particularly sensitive to the choice of certain model parameters, something which is not discussed in the manuscript!
While we deliberately used the same set of standard model parameters as in Pohl et al. (2016b), permitting one-to-one comparison, we agree with Reviewer 1 that sensitivity tests to key model parameters should be conducted in future work trying to constrain the size and geometry of the ice sheet (which is beyond the scope of the present contribution). This point was briefly made in this version of the manuscript on lines 439-441 and we will expand on it in response to the Reviewer’s comment.
We will add the following at the sentence break in line 441:
“Of particular interest in future work aiming to constrain the extent and volume of the ice sheet would be the use of an energy balance model in place of the degree-day scheme to represent ablation (van de Wal 1996) and sensitivity tests to ice-sheet basal dragging strength (Pohl et al., 2016b).”
Citation: https://doi.org/10.5194/egusphere-2026-3617-AC1
-
AC1: 'Reply on RC1', Thomas Wong Hearing, 21 Sep 2026
-
RC2: 'Comment on egusphere-2026-3617', Anonymous Referee #2, 31 Aug 2026
Hearing et al. review Ediacaran glacial deposits and compile their latitudes in two paleogeographic models. They then construct climate and ice sheet models for the mid-Ediacaran glaciations. To my knowledge these are the first attempts to model the Ediacaran ice extent using real climate models and data. I think the review and modeling is useful for the community, to recognize that there is no strong evidence for low-latitude Ediacaran glaciation, nor any need for exotic geophysical phenomena, like inertial interchange true polar wander, to explain the data. I really appreciate how the authors discuss and depict the uncertainties in this model. This is the second time I have reviewed a version of this manuscript and it is much improved.
The paper is well written and I have few additional comments for improvement. My only suggestion is that Figure 5 could be improved a bit to have a bigger impact and be used more by others. Get rid of the dashed lines and the different shades—these are confusing and distracting. I don’t know why the authors use the dashes for the LEIH—just use the same paleogeographic reconstructions and finish this plot and maybe put a question mark for extent. Putting dashes down to the equator makes readers think that there is evidence that it could be global in extent. It is also unclear why they put the dashes from 635-580 Ma. There is no evidence for ice during this period, so they should delete this as it will perpetuate in the literature otherwise. It is also unclear why they make the Cryogenian solid, high latitude ice. There is no low-latitude ice from 661-640 Ma (see Tasistro-Hart et al., 2025, PNAS; Tasistro-Hart et al., 2026, Geology). Cleaning up this diagram will make the paper have a much bigger impact—people will actually use this and cite this. If you keep it confusing with weird dashed intervals, no one will use this and cite this because they would have to explain the weird dashed intervals.
Citation: https://doi.org/10.5194/egusphere-2026-3617-RC2 -
AC2: 'Reply on RC2', Thomas Wong Hearing, 21 Sep 2026
We are grateful for the constructive reviews from the reviewer. We respond to their main points below, with Reviewer comments in italics and our responses in bold.
The paper is well written and I have few additional comments for improvement. My only suggestion is that Figure 5 could be improved a bit to have a bigger impact and be used more by others. Get rid of the dashed lines and the different shades—these are confusing and distracting. I don’t know why the authors use the dashes for the LEIH—just use the same paleogeographic reconstructions and finish this plot and maybe put a question mark for extent. Putting dashes down to the equator makes readers think that there is evidence that it could be global in extent. It is also unclear why they put the dashes from 635-580 Ma. There is no evidence for ice during this period, so they should delete this as it will perpetuate in the literature otherwise. It is also unclear why they make the Cryogenian solid, high latitude ice. There is no low-latitude ice from 661-640 Ma (see Tasistro-Hart et al., 2025, PNAS; Tasistro-Hart et al., 2026, Geology). Cleaning up this diagram will make the paper have a much bigger impact—people will actually use this and cite this. If you keep it confusing with weird dashed intervals, no one will use this and cite this because they would have to explain the weird dashed intervals.
Yes, this figure ought to be cleaned up and we will do this. We address each point separately below.
We think there is some misunderstanding here with the extent of ice in the Cryogenian. We depict the Cryogenian glacial intervals as solid low to high latitude ice from 717 to 661 Ma and 640 to 635 Ma. We use a light blue fill to illustrate uncertainty of onset of Marinoan glaciation following Hoffman et al., 2017). We will put additional time labels on the x-axis (at 50 or 100 Myr intervals) to help clarify the timescales. We will expand the figure legend to include all elements used (including the light blue shading which is only described in the caption at present). We will also make clear on the y-axis that this is referring to land ice.
The uncertainty around the LEIH is there because we have not applied the same degree of review to the LEIH deposits as we have for the MEIH deposits which are the focus of this submission. We will replace this interval with a light grey bar to mid latitudes, indicating that there is no evidence for high to low latitude glaciation (as with the Cryogenian) but that there is still uncertainty about the true extent of LEIH deposits.
The pre-600 Ma uncertainty reflects recent work showing a dearth of early Ediacaran deposits (Cantine et al., 2024) and hints of pre-600 Ma (~620 Ma) glacial deposits (Lan et al., 2024), suggesting there may be more earlier Ediacaran glaciations than currently realised, i.e. this represents real uncertainty in the literature that should be reflected in summary figures. We will improve the clarity of this uncertainty by representing the region in a solid light grey fill and include explicit explanation of this on the figure as well as in the caption.
Citation: https://doi.org/10.5194/egusphere-2026-3617-AC2
-
AC2: 'Reply on RC2', Thomas Wong Hearing, 21 Sep 2026
-
EC1: 'Comment on egusphere-2026-3617', Gerilyn (Lynn) Soreghan, 03 Sep 2026
Hearing et al. evaluate ages and depositional origins of Ediacaran glaciogenic (and candidate glaciogenic) deposits and combine these data with climate- and ice-sheet modeling to explore Ediacaran glacial history, ultimately inferring the existence of two icehouse intervals that behaved in a manner akin to the Phanerozoic icehouses (in contrast to snowball states).
I found the manuscript to be well and clearly written, and appreciated the clear and logical explanations of the age- and depositional “grading” systems, as well as the modeling, and assessment of paleogeographic constraints. Two reviewers also evaluated the manuscript.
Reviewers have provided detailed and constructive comments. Overall, I expect to be encouraging the authors to submit a revised version that addresses the comments and suggestions in the two very constructive reviews.
Citation: https://doi.org/10.5194/egusphere-2026-3617-EC1 -
AC3: 'Reply on EC1', Thomas Wong Hearing, 21 Sep 2026
We are grateful to the Editor, Lynn Soreghan, for her handling of our submission. We really appreciate the feedback from the Editor and both reviewers. We are sure that we can address the comments raised across the very constructive reviews and we think addressing these will help improve the clarity of the manuscript.
Kind regards,
Thomas Wong Hearing, on behalf of all authors
List of references cited across our responses
Cantine, M.D., Rooney, A.D., Knoll, A.H., Gómez-Pérez, I., al Baloushi, B. and Bergmann, K.D. 2024. Chronology of Ediacaran sedimentary and biogeochemical shifts along eastern Gondwanan margins. Communications Earth & Environment, 5, 1–9, https://doi.org/10.1038/s43247-024-01630-1.
Hoffman, P.F., Abbot, D.S., et al. 2017. Snowball Earth climate dynamics and Cryogenian geology-geobiology. Science Advances, 3, e1600983, https://doi.org/10.1126/sciadv.1600983.
Lan, Z., Huyskens, M.H., Ren, R. and Yin, Q.-Z. 2024. A potentially New Early Ediacaran glaciation. Journal of Earth Science, 35, 1810–1819, https://doi.org/10.1007/s12583-024-1979-7.
Myhre, G., Highwood, E.J., Shine, K.P. and Stordal, F. 1998. New estimates of radiative forcing due to well mixed greenhouse gases. Geophysical Research Letters, 25, 2715–2718, https://doi.org/10.1029/98GL01908.
Pohl, A., Donnadieu, Y., Le Hir, G., Buoncristiani, J.-F. and Vennin, E. 2014. Effect of the Ordovician paleogeography on the (in)stability of the climate. Clim. Past, 10, 2053–2066, https://doi.org/10.5194/cp-10-2053-2014.
Pohl, A., Donnadieu, Y., Le Hir, G., Ladant, J.-B., Dumas, C., Alvarez-Solas, J. and Vandenbroucke, T.R.A. 2016. Glacial onset predated Late Ordovician climate cooling. Paleoceanography, 31, 800–821, https://doi.org/10.1002/2016PA002928.
van de Wal, R.S.W. 1996. Mass-balance modelling of the Greenland ice sheet: a comparison of an energy-balance and a degree-day model. Annals of Glaciology, 23, 36–45, https://doi.org/10.3189/S0260305500013239.
Citation: https://doi.org/10.5194/egusphere-2026-3617-AC3
-
AC3: 'Reply on EC1', Thomas Wong Hearing, 21 Sep 2026
Data sets
Supplementary Data T. W. Wong Hearing et al. https://zenodo.org/records/17158910
Model code and software
Supplementary Code T. W. Wong Hearing and A. Pohl https://github.com/twwh01/MEIH/
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 198 | 102 | 30 | 330 | 75 | 24 | 27 |
- HTML: 198
- PDF: 102
- XML: 30
- Total: 330
- Supplement: 75
- BibTeX: 24
- EndNote: 27
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
This manuscript combines a systematic evaluation of glaciogenic deposits for the Ediacaran with climate and ice-sheet model simulations to explore the glaciation history during this time period. The authors find evidence for two discrete icehouse intervals in the middle and late Ediacaran (MEIH and LEIH, respectively). Based on the combined evidence, they suggest that the MEIH was a "Phanerozoic-style" icehouse and provide some constraints on uncertain paleogeographic reconstructions at the time, which I find a particularly interesting thought and contribution, albeit challenging because of other uncertainties.
Overall, this is a very good and important paper which I would like to see published after revision. My major concerns relate to the discussion of uncertainties in terms of paleogeographic reconstructions and model parameters, see my detailed comments below. I hope these remarks will prove useful to the authors during revision of the manuscript!
Specific comments:
l. 94: For clarity, it would be good to specify here whether this is just reproduced or updated (updated according to the main text). It would also be helpful to describe the changes relative to the version published earlier somewhere.
l. 194-204: When first reading this paragraph, I had wondered whether it would be useful to show the different reconstructions somewhere, in particular the topography of the PALEOMAP reconstruction because of its importance for ice-sheet formation, and realised only later that they are shown in the Supplementary Information. It would be good to refer to the respective figure in the SI in this subsection.
l. 225: Please specify the present-day value used to scale the Ediacaran solar luminosity (it appears to be higher than the generally accepted value of about 1361 W/m2).
l. 308, subsection 3.2: I am missing a bit more discussion here, in particular concerning Fig. 3 and the plausible glaciogenic deposits deep in the tropics.
l. 315-317: This is a bit confusing because the reader might interpret the "lower latitudes" in comparison to the ones in the PALEOMAP reconstruction rather than lower latitudes absolutely. I would suggest to rephrase the sentence to avoid the confusion.
l. 336-337: This is a bit confusing. This selection does obviously not include the MEIH deposits, right? This should be explained better!
l. 346-348: This is too vague. You should be able to explore the mechanism based on model diagnostics.
l. 351-353: Could be moved to the Methods section. Also, could you comment on whether the ice sheet is approaching equilibrium at this point?
l. 386: "[S]ome uncertainty" seems rather euphemistic, please delete "some".
l. 400-402: Well, yes... and no... You should keep in mind that topography is usually rather uncertain as well and that high topography at low latitudes could lead to glaciation in terms of glaciers on tropical mountains.
l. 404-406: This cannot be the only reason since the radiative forcing from the change in CO2 is more than twice the radiative forcing from the change in the solar constant, if my back-of-the-envelope calculation is correct.
l. 408: One thing your equilibrium simulations also do not take into account is that orbital forcing is continuously varying in reality. This should be discussed somewhere.
l. 441: I agree. Ice-sheet models are particularly sensitive to the choice of certain model parameters, something which is not discussed in the manuscript!