the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 21 Oct 2025
How unstable was the environment during the Penultimate Glacial in the South-Western Mediterranean? Vegetation, climate and human dynamics during MIS 6
Abstract. The impact of rapid climate variability on Neanderthal population in Europe during the Last Glacial (Marine Isotope Stages 4-2), including Dansgaard-Oeschger cycles and Heinrich events, has been the subject of a long-standing debate. However, few studies have focused on the nature and impact of such rapid variations on human population during earlier periods. A growing number of high-resolution paleoclimatic archives supports the persistence of rapid oscillations during the penultimate glaciation (MIS 6), and the close response of Mediterranean ecosystems to these. Still, few palynological sequences in the Mediterranean region offer sufficient resolution to document vegetation dynamics during this time. Pollen records are especially lacking in the western Mediterranean, a key region to understand the connection between North Atlantic and Mediterranean climatic influences. This region is also traditionally considered a climatic refugium for human population during unfavourable periods. We provide new palynological data covering MIS 6 from the long and continuous marine record of ODP 976 in the Alboran Sea. A total of 200 samples, spanning the interval from 196 to 127 ka Before Present (BP), reveal both long-term trends and rapid fluctuations in regional vegetation composition. A multi-method approach, including modern analogues, regression, and machine learning approaches, was applied to ODP 976 pollen assemblages to reconstruct the annual/seasonal temperatures and precipitation. Results show that three phases can be identified. The first phase (187–166 ka BP) is characterized by significant oscillations of temperate trees and rather cool and humid conditions during early MIS 6, coincident with a sapropel layer deposition in both the western and eastern Mediterranean. In the second phase (165–144 ka BP), arid herbaceous vegetation is dominant, marking the main imprint of glacial maxima conditions and reduced climate variability. The third phase (144–129 ka BP) is marked by the development of Ericaceae and increased annual precipitations. At the end of MIS 6 glaciation, an strong cooling and intense episode of steppe and semi-desert expansion is identified as Heinrich Stadial 11 (135–129 ka BP), marking a distinct pattern for Termination II in the Western Mediterranean. Rapid oscillations appear like a pervasive feature of the Penultimate glacial in the SW Mediterranean, though they present reduced amplitude and frequency compared to the Last Glacial. A synthesis of human occupation shows that a mosaic of traditional (Mode 2) and innovative (Mode 3) technological features is observed. Although the data are scarce, Neanderthal seems to have continuously inhabited Western Mediterranean regions across the penultimate glacial. The severe climate conditions during Heinrich Stadial 11 (~133–129 ka BP) might have played a role in the apparent population contraction at the end of MIS 6, and perhaps also in the definitive abandonment of Lower Palaeolithic industries.
Competing interests: At least one of the (co-)authors is a member of the editorial board of Climate of the Past.
Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.- Preprint
(2988 KB) - Metadata XML
-
Supplement
(2644 KB) - BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2025-5024', Anonymous Referee #1, 10 Nov 2025
-
RC2: 'Comment on egusphere-2025-5024', Anonymous Referee #2, 11 Nov 2025
Revision Charton et al. (2025)
The study by Charton et al. (2025) presents new high-resolution pollen data and quantitative reconstructions of temperature and precipitation, which have been used to relate it to previous archaeological interpretations concerning human populations during MIS 6–5. Thanks to the high-resolution pollen analysis carried out (200 samples), the authors have also been able to explore short and intense paleoclimatic periods, such as the HS11. Therefore, despite certain issues described below in the section “Major comments,” the study provides valuable data and interesting multidisciplinary interpretations, making it a noteworthy contribution to the scientific community, and deserves publication after major revisions.
Major comments:
One of the main issues concerns the chronology of the studied core, which relies on correlations with other records through tie-points. Independent chronological control in marine archives for ages above the radiocarbon dating limit has been a common challenge in paleoclimate studies. However, in this case, the correlation of the studied core with the marine record MD01-2444, which also lacks independent age control and chronology, and which in turn is aligned with the Greenland δD record (Dome C ice-core), is problematic. This situation may lead to paleoclimatic reconstructions being subject to circular reasoning. Since the chronological framework represents one of the weaknesses of this study, it would be necessary to present the age alignments more explicitly to the reader, for example, by showing the depth correlations from the onset of MIS 6 to the present between the studied record and MD01-2444, as well as the tie-points linking MD01-2444 with the Greenland core. By showing the depth correlations for the last 200 ka together with the absolute ages (displayed alongside the depth axis) of records that are independently dated (in this case, the Antarctic Dome C ice-core), the reader can understand that the tie-points used here are potentially robust and that these alignments correspond to the same climatic variations of the penultimate glacial period. Similarly, no correlation graph is presented between the ODP 976 core and the Chinese speleothem records from Dongge Cave.
Regarding the quantitative reconstructions, the authors should provide a clearer specification of the modern pollen training set used for the reconstructions. It is indicated that the dataset of Peyron et al. (2013, 2017) and updated by Dugerdil et al. (2021) and Robles et al. (2023) was employed. However, it is essential to include the total number of harmonized taxa used for reconstructions and should also mention any taxa that have been excluded, for example, aquatics. It would also be helpful to show a map displaying the location of all the modern samples used. They should also specify the area, including the latitude and longitude of this training set. It would also be useful to see the maximum bias for each method in Table 3. To better understand which taxa are responsible for the changes observed in the reconstructions, I also suggest including a figure showing the tolerance spectra of the main pollen taxa for each of the climatic parameters used (MAAT, MTCO, TCON, MAP, WINTERRR, PCON).
At the end of section 5.1, a delay of around 1000 years is mentioned with respect to the marine SST of ODP 976 (Jiménez-Amat and Zahn, 2015; Martrat et al., 2014). Considering that this is the same record, why does this delay occur? Is this due to a delay in the continental response compared to that recorded in the sea? Is the age model of Jiménez-Amat and Zahn (2015) and Martrat et al. (2014) based on the same tie-points as this study, and therefore the age chronology is the same? If not, then referring to a delay of only 1,000 years when dealing with interpretations for MIS 6, and considering that the chronological control is not independent and relies on correlations between warm events, appears at the very least, risky and potentially overinterpreted.
In section 5.4, the authors analyze the HS11 signal in OPD 976 and other records. Although they mention the difference in intensities recorded between the eastern and western Mediterranean, they do not delve into the climatic mechanisms that could have led to this difference between the two areas.
Minor comments:
In Methods, I suggest including the average resolution of pollen samples in age. Taking into account the beginning (195.72 ka BP) and end of the record (127.47 ka BP), and the 200 pollen samples analyzed, the average resolution should be approx. 341 years/sample. Please verify this.
Mention the meaning of “mcd” when it first appears in Table 1 or in Line 215, not in Figure 3.
Explain why a radius of 400 km is used to calculate the average value of climate parameters. Pollen influence is reduced beyond a radius of 175 km (Rojo et al., 2016). However, other studies suggest that pollen can be transported up to distances of 200-300 km (Fernández-Rodríguez et al., 2014), and even over distances greater than 500 km (Bayr et al., 2023; Bogawski et al., 2019). The authors could use some of these studies to explain the radius used to obtain the recent average values.
Lines 263-266: Authors should mention the total number of harmonized taxa that compose the training set.
Line 300: “Jeanne Rampal, personal communication”. Provide the age of the communication.
Lines 872 and 917: Use HS11 instead of “Heinrich Stadial 11”
Figure 6: Explain why Pinus was included as pioneer vegetation in Ioannina and why it was not included in other records.
To facilitate reading between the text and the figures, I suggest including the letters of the curves when referring to a record in a figure, especially when referring to records in Figure 7.
New references cited:
Rojo, J. et al. Modeling olive pollen intensity in the Mediterranean region through analysis of emission sources. Science of The Total Environment 551-552, 73-82 (2016).
Fernández-Rodríguez, S. et al. Identification of potential sources of airborne Olea pollen in the Southwest Iberian Peninsula. International Journal of Biometeorology 58, 337-348 (2014).
Bayr, D. et al. Pollen long-distance transport associated with symptoms in pollen allergics on the German Alps: An old story with a new ending? Science of The Total Environment 881, 163310 (2023).
Bogawski, P., Borycka, K., Grewling, Ł. Kasprzyk, I. Detecting distant sources of airborne pollen for Poland: Integrating back-trajectory and dispersion modelling with a satellite-based phenology. Science of The Total Environment 689, 109-125 (2019).
Citation: https://doi.org/10.5194/egusphere-2025-5024-RC2 -
RC3: 'Comment on egusphere-2025-5024', Anonymous Referee #3, 21 Jan 2026
As an archaeologist specializing in Pleistocene hunter-gatherer adaptations rather than a paleoclimatologist, I want to be transparent that my evaluation focuses primarily on the archaeological interpretations and the human-environment dynamics discussed in this paper. With that acknowledged, this study provides a valuable paleoenvironmental framework that is sorely needed for understanding human occupation patterns during MIS 6 in the Western Mediterranean, a period that remains critically understudied compared to the Last Glacial.
The paper deserves publication for several key reasons from an archaeological perspective:
Major strengths:
- Filling a critical gap in human-environment research: The archaeological record for MIS 6 is notoriously sparse and poorly dated. This high-resolution paleoenvironmental sequence from the Alboran Sea provides the chronological and environmental context that archaeologists desperately need to understand the Lower to Middle Palaeolithic transition and Neanderthal population dynamics during this period. The 200-sample resolution (~341 years/sample) allows correlation with human occupation phases at much more detailed level.
- The climate refugia hypothesis: The paper's demonstration that the SW Mediterranean maintained relatively stable (though harsh) conditions during MIS 6, combined with the archaeological synthesis showing continuous human presence in Southern France and Iberia, strengthens the long-standing refugia model for this region. This has important implications for understanding Neanderthal metapopulation dynamics and the persistence of "source" populations.
- HS11 as a potential demographic bottleneck: The identification of Heinrich Stadial 11 (~133-129 ka BP) as a particularly severe event in the Western Mediterranean is archaeologically significant. The observation that many MIS 6 sites show chronological boundaries compatible with TII onset (~136 ka BP) is compelling and suggests this event may have been a catalyst for population contraction and technological reorganization. This interpretation aligns well with genetic evidence for Neanderthal demographic stability only emerging after MIS 5e.
- The archaeological synthesis (Section 5.6): This section provides a useful compilation of dated MIS 6 sites across SW Europe, distinguishing between cave and open-air contexts, and discussing the technological mosaic of Acheulean and Early Middle Palaeolithic assemblages. The regional comparison (Italy vs. Iberia vs. Southern France) offers insights into differential human responses to glacial conditions.
Suggestions for improvement:
- Expand discussion of behavioral adaptations: While the paper mentions changes in land use, mobility patterns, and subsistence strategies during MIS 6 (citing Diez-Martín, Sánchez-Yustos, and Rios-Garaizar), these could be developed further. How do the three environmental phases identified in ODP 976 correlate with observable changes in lithic technology, site function, or faunal exploitation patterns? For instance, does the shift from Early MIS 6 (humid, variable) to Middle MIS 6 (full glacial, stable) correspond to changes in raw material procurement ranges or prey selection documented at sites like Bolomor or Lazaret?
- Consider Neanderthal niche modeling: The brief mention of Yaworsky et al. (2024) showing reduced niche space at ~145 ka BP is intriguing but underexplored. How do the climate reconstructions presented here compare with modeled Neanderthal climatic tolerances? This could strengthen the argument about population contraction during full glacial conditions.
- The Lower to Middle Palaeolithic transition: The paper suggests that MIS 6 glaciation "could have played a role in the final abandonment of Acheulean industries." This is an important claim that deserves more nuanced treatment. What specific environmental pressures might have favored Levallois technology over bifacial production? Were these pressures most acute during the cold-dry middle phase or during the rapid oscillations of early MIS 6? The environmental data presented here could potentially contribute to this long-standing debate.
- Site-level correlations: While the archaeological synthesis is valuable, the temporal resolution of most archaeological sites remains too coarse for direct correlation with specific interstadials or stadials identified in ODP 976. The paper should acknowledge this limitation more explicitly and perhaps identify which sites have the chronological precision to test human-environment correlations at sub-orbital scales.
- Clarify the refugia concept: The paper uses "climate refugia" somewhat interchangeably for both vegetation and human populations. It would be worth discussing whether areas that served as refugia for temperate vegetation necessarily corresponded to optimal habitats for Neanderthals, who were evidently capable of exploiting a broader range of environments than many plant communities.
- HS11 and human response: The claim that HS11 "did not lead to complete extinction of hominin groups but might have induced deep demographical and technological reorganization" is compelling but would benefit from more specific archaeological evidence. Are there observable differences in assemblage composition, site density, or raw material use between pre-HS11 and post-HS11 contexts in the region?
- The comparison of MIS 6 with MIS 4-2 (Section 5.5) from the same core is methodologically valuable. However, the archaeological implications of the "subdued" millennial variability during MIS 6 compared to the Last Glacial could be discussed more explicitly. Did reduced climate oscillation frequency translate to different patterns of population mobility or technological change?
Conclusion:
This paper makes a significant contribution by providing the high-resolution paleoenvironmental framework necessary to understand human-environment dynamics during a critical but understudied period. The identification of three distinct environmental phases during MIS 6, the characterization of HS11 as a severe event in the Western Mediterranean, and the archaeological synthesis of human occupation patterns all advance our understanding of Neanderthal adaptations during the penultimate glaciation. I recommend publication after revisions that strengthen the archaeological interpretations and address the chronological limitations of site-level correlations.Citation: https://doi.org/10.5194/egusphere-2025-5024-RC3
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 361 | 200 | 35 | 596 | 43 | 25 | 25 |
- HTML: 361
- PDF: 200
- XML: 35
- Total: 596
- Supplement: 43
- BibTeX: 25
- EndNote: 25
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
In this study Charton et al. carried out a high-resolution pollen analysis of the penultimate glaciation part of a marine core from the Alboran Sea. The authors discussed the pollen data and deduced vegetation changes and climate inferences and compared them to other paleoclimatic records from the study area making some interesting inferences. Detailed records from the MIS6 glaciation in the western Mediterranean are quite rare so that adds interest to this study.
I suggest publication in Egusphere but some changes should be done before publication. Here below are my comments:
-Line 43: This sentence is not clear - explain this further.
-Line 44: in plural: neanderthals seem to have..
-Line 128: be more specific?
-Line 136: chose a better word than connect. relate? correlate?
-Line 157: Not really a climate reference for the study area
-Line 165: Picea does not grow naturally in the study area
-Line 169: The Atlantic currents? which ones? can you be more specific for your study site?
-Line 194: add: correlation to orbital configuration
-Line 199: The chronological control of the sequence is based on direct correlation to very distant and various records, such as a speleothem in China and a record from the coast off Portugal, which in turn was aligned to the Antarctic temperature record (even considering the bipolar seesaw...) ...the authors then assumed that vegetation changes in the Mediterranean were synchronous to Chinese and Antarctic temperatures. There is a possibility that this can be true but this approach limits the output in terms of lags or leads of the response of the vegetation in your record to climate events worldwide... For example, in lines 402-403: How is it possible to observe a delay for TII in your record if the chronology was tuned to that particular speleothem record? If there is not an objective direct chronological control of your core you cannot really discuss about this matter
-Line 215: change "here" for "In this study"
-Line 224: Change "along" for "in"
-Line 234: remove "As already pointed out,"
-Figure 3. I imagine this figure was meant to be represented horizontally. However, vertically the numbers in the scales are upside down and should be rotated 180º.
-Lines 294-314. In this section, the results are already interpreted in terms of climate - I would leave those interpretations for the discussion and just describe the results here
-Line 303: Ericaceae rather indicates humid conditions, not cold climate
-Line 341: Present-day climatic conditions are shown in Figure 5 by stars - are those from the location of the marine site? from any other specific location? please specify
-Line 341: Remove "all along" and change it for "during"
-Figure 5 and text: The substages MIS6e-a were defined in an isotopic record but do not seem to agree with the results here - why do you pay so much attention to that?
-Fig. 5 legend: it is a black star (not red)
-Lines 380-381. This sentence is too vague...the pollen results do not really agree with the MIS6 glacial substages (a-e) and/or D-O dynamics that are only defined for the last glaciation...maybe it would be better to say "D-O-like dynamics" and I would remove the glacial substages.
-Lines 382-391: As these three phases don't coincide with the previously defined substages, I would remove those from the text and add ages when those phases occur in your record
-Line 402-403. Please give the ages for the beginning and end of the termination II in your record and describe how you identified that termination in the ODP 976 pollen record
-Line 407: W?
-Line 413: Do you mean the abrupt drop of temperatures at the beginning of HS11?
-Line 416: I would rather call this section "Hydroclimate connection..."
-In Figure 3. Pioneer vegetation is represented by brown color, as Ericaceae here. Please be consistent and use the same color for the different categories in all figures
-Line 455: Could you please explain why deciduous forest is important in better noticing the MIS7-6 transition?
-Lines 477-486. Maybe representing site MD01-2444 in Figure 6 would be usefull, even though it is not a Mediterranean site "per se", it looks like ODP 976 share a lot of similarities with it
-Lines 481-486: Could it be due to lower summer insolation generating lower evaporation in summer and favoring the growth of Ericaceae, which requires humidity throughout the year?
-Line 498: I would just say Moroccan Mountains - Cedrus could also originate from the Atlas
-Line 505: Change across for throughout
-Line 507: Not just rapid, but enhanced development of forest in that area
-Line 508: Please explain why altitude is important here
-Line 523: Please be more specific of why do you think it is compatible with DO-like variabilty. Is it the duration of the events? the intensity?
-Line 527: What does "disentangle the character" mean? investigate?
-Line 529. What do you mean with consistent? similar? in what? time? intensity? be more specific
-Line 532: Remove "the" and change it for "its more semiarid"
-Line 434. Confirm this, as earlier you said Ericaceae was a "cold" indicator (see line 303) - be consistent
-Line 539. This affirmation is very vague. There are significant differences between the Cueva Gitana and ODP 976, for example maximum humidity seems to occur at 148 kyr, which is not replicated in your record
-Fig. 7. The text of the legends of the different plots should be written in a bigger size.
Also, the scale in the synthetic pollen figure from ODP 976 is missing
In addition, the pollen data do not really agree with the substage subdivision done by Railsback fro MIS 6 - I would remove that subdivision.
-Lines 566-572: This paragraph is repetitive, see the beginning of the discussion in lines 382-415. I would remove it or synthesize those two paragraphs in one. Also, check more text in this section for further repetition
-Lines 573-583. Check for repetitions in the sapropel discussion with the previous section about the ORLs - this could be synthesized...maybe adding the ORL text in this section?
-Line 589. How do the isotopic and sea surface temperature records reflect precipitation? Please explain
-The size of figure 9 as it is shown in the pdf here is too small - one can barely read the text-numbers so it should be bigger.
-Line 903: replace "consistent with" with similar to
-Line 917: Change "double u for "W"