the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 07 Jan 2026
Concurrent hydrological closure and hominin presence in the Early Pleistocene Nihewan Basin (northern China): insights from stable isotopes
Abstract. The Nihewan Basin in northern China contains rich Early Pleistocene Palaeolithic sites, representing one of the earliest locations of hominins outside Africa. Here, we present the first long-term stable oxygen (δ18Oeq.cal) and carbon (δ¹³Ceq.cal) isotope record derived from ostracod shells, preserved in the composite 86.2-m NH-T sediment section at the northeastern part of the basin, with a timeframe between ca. 1.67 and 0.78 Ma. The study aimed at reconstructing the long-term climatic changes and hydrological dynamics of the Early Pleistocene Nihewan Basin and to assess their impact on hominin activities. Unexpectedly, we found a strong covariance of δ18Oeq.cal and δ¹³Ceq.cal values, clearly suggesting that the basin was mostly hydrologically closed. The dominance of evaporation implies that δ18Oeq.cal shifts track the hydrological state at the section location between closed settings with higher water levels (more standing waters) and open settings with low water levels (more flowing waters) instead of regional changes in precipitation/evaporation ratios alone. Moreover, we observed the concurrence of high δ18Oeq.cal and δ¹³Ceq.cal values and the increase in the marine-land temperature gradient (ΔT), indicating enhanced East Asian Summer Monsoon (EASM)-driven precipitation which led to wetter climate and increased biogenic productivity. Conversely, low δ18Oeq.cal and δ¹³Ceq.cal values reflect decreased EASM-driven precipitation and drier climate and reduced biogenic productivity. The new stable isotope data, combined with the synthetic archaeological record, suggest that hominin activities in the Nihewan Basin mostly coincided with periods of higher δ18Oeq.cal and δ¹³Ceq.cal values when more standing waters bodies and wetter climate prevailed in the region.
- Preprint
(1817 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2025-5999', Anonymous Referee #1, 03 Feb 2026
-
AC1: 'Reply on RC1', Ahmed Moghazi, 23 Jul 2026
- Lines 47-50. Rewrite simply as “Today, it represents one of the densest concentrations of Early Pleistocene Palaeolithic sites outside of Africa”. Here, I would also add Dennell (2009) and Bae (2024) who both spend some time talking about the importance of the Nihewan Basin in their books. I would also recommend citing in here someplace the recent Bae and Manthey (2025) paper as it points out that the Nihewan Basin is no longer considered to be a really “early” region for hominin occupation in eastern Asia. Sites like Xihoudu, Shangchen, and Longgupo are clearly older and the recent dating of the Yunxian Homo erectus site (Tu et al., 2026) is also older than the archaeology sites in the Nihewan Basin.
Response: Thank you for raising this concern regarding the chronological position of the Nihewan Basin in the context of earliest hominin locations outside Africa. Following a review of recent paleoanthropological discoveries and newly refined chronologies based on radiometric dating in East Asia, we have revised the text to mention that a growing number of archaeological sites across East Asia are now considered older than those in the Nihewan Basin and modified the wording accordingly. We have also incorporated the suggested references, including Dennell (2009), Bae (2024), and Bae and Manthey (2025).- Lines 50-54. This needs to be rewritten given the evidence from Xihoudu (not very far south of the Nihewan Basin) is dated to 2.43 Ma and Shangchen dates to 2.1 Ma. Mention should also be made someplace in here that most of the Nihewan Basin sites ages are based on relative dating techniques (paleomag, biostrat), while some sites like Xiashagou and Xujiayao (Tu et al., 2022 [cited in paper]; Shao et al., 2026 [not cited in paper]) have been radiometrically dated.
Response: Thank you for the suggestion. In addition to the revised text described above, we updated the text to include that the Xihoudu site represents the earliest evidence of hominin presence in China and possibly in the whole of Asia. We have also clarified that most Nihewan Basin archaeological sites are magnetostratigraphically dated (relative dating method), whereas sites such as Xiashagou and Xujiayao have been constrained by radiometric dating.
- Line 54. Start new paragraph with the first sentence (“Based on their…”).
Response: Thank you. We have started a new paragraph at this point, as recommended.
- Lines 68-71. Going out to the third digit seems a bit too overly precise, particularly given that these are likely paleomag dates. Recommend just going out to the second digit as presented in most paleomag studies in eastern Asia.
Response: We agree and adjusted the ages accordingly.
- Line 73. Rewrite “1.3 Ma” to “1.3 millions of years”. Technically, most readers read “Ma” as millions of years ago, but in this context you are only referring to a span of 1.3 millions of years.
Response: Thank you for pointing this out. We have revised the wording accordingly.
- Lines 128-130. You should add some justification and clarification as to what revisions you made to the age of the NH-T section in this study. As written, you leave the reader hanging, just saying you did it. But based on what?
Response: Thank you for the helpful comment. We agree that the original text did not explain the basis for the revised chronology. We have therefore revised the wording to clarify that, following a re-examination of the chronological anchoring points to the La2004 astronomical solution. Thus, the resulting astronomical age range originally reported as ca. 1.66-0.78 Ma in Moghazi et al. (2024b) has been slightly adjusted to ca. 1.67-0.78 Ma.
- Line 285 and Line 296. Provide the age range in brackets for these S and L layers unless it includes the entire time bracket for this section.
Response: We agree that providing the age ranges of the S and L layers improves clarity. We have, therefore, added the corresponding age ranges in the revised version of the manuscript.
- Line 308. Ibid.
Response: We added the age range here as well and in other lines in a consistent way.
- Section 5.1. It seems that your data crosscut both S and L layers. Is it not possible to determine if there are significant different in isotope ratios between S and L? Given the clear environmental differences between S and L layers, how can you explain then that your data is the same for both S and L?
Response: Thanks for this good observation. We agree that this point was not sufficiently clear in the original manuscript. A visual comparison shows that intervals characterized by relatively higher δ¹⁸O and δ¹³C values generally correspond to the correlated S intervals, whereas lower isotope values tend to coincide with the correlated L intervals, indicating an isotopic distinction between S and L. We have clarified this relationship in the revised manuscript. The correlation with the loess/paleosol sequence of the CLP was established due to the relatively comparable variations of the Early Pleistocene East Asian winter and summer monsoons reconstructed mainly from the grain-size end-member (EM) record of the NH-T section. This correlation was later used as base to track the localized hydrological signal changes inferred from the isotope record.
- Line 394. Replace “back from” with “between”. And given that Xujiayao is a part of the Nihewan Basin, this time spread should be pushed back to 160 ka (Shao et al., 2026).
Response: Thank you for bringing this newly published study to our attention. We have revised the manuscript accordingly and updated the age range to ca. 1.66-0.16 Ma (Dennell, 2013; Shao et al., 2026).
Citation: https://doi.org/10.5194/egusphere-2025-5999-AC1
-
AC1: 'Reply on RC1', Ahmed Moghazi, 23 Jul 2026
-
RC2: 'Comment on egusphere-2025-5999', Manuel R. Palacios-Fest, 26 May 2026
I find this investigation fascinating! It is an excellent compilation of information about climate change and its impact on previous hominids. I enjoyed the study from beginning to end.
I am intrigued by why the authors combined the stable isotope signatures of two different ostracode families. It is true that within the same genus, vital effects are not significantly different; merging the signatures of two families is somewhat risky. I would appreciate a better explanation. If the species represented different stratigraphic intervals, it would be better to use the stable isotope values, adjusting for the vital effects accordingly. For example, I don't have a problem with the authors merging L. inopinata with C. lacustris; both are limnocytherids and cold-water indicators, and their ecological requirements are similar or overlapping. However, mixing them with Ilyocypris sp. and Eucypris mareotica (two cypridids) may significantly alter the results. Can you trust the trends shown in Figures 2 and 3? I suggest they be careful when mixing species with different vital effects.
I recommend that the authors add a far-right column in Figure 2, defining the biozones identified in section 5.1.1 (Evolution between 1.67 to 1.30 Ma, the pre-MPT period). It would help the reader visually distinguish the biozones or biofacies (whichever term the authors prefer).
I found several references in the references list that weren't cited in the text: Deng et al. (2006a, b), Deng et al. (2007), Li et al. (2015), Mischke et al. (2021), Tian et al. (2002), Tian et al. (2004), Wang et al. (2005), Wang et al. (2010), Xie and Cheng (1990).
Citation: https://doi.org/10.5194/egusphere-2025-5999-RC2 -
AC2: 'Reply on RC2', Ahmed Moghazi, 23 Jul 2026
- I am intrigued by why the authors combined the stable isotope signatures of two different ostracode families. It is true that within the same genus, vital effects are not significantly different; merging the signatures of two families is somewhat risky. I would appreciate a better explanation. If the species represented different stratigraphic intervals, it would be better to use the stable isotope values, adjusting for the vital effects accordingly. For example, I don't have a problem with the authors merging L. inopinata with C. lacustris; both are limnocytherids and cold-water indicators, and their ecological requirements are similar or overlapping. However, mixing them with Ilyocypris sp. and Eucypris mareotica (two cypridids) may significantly alter the results. Can you trust the trends shown in Figures 2 and 3? I suggest they be careful when mixing species with different vital effects.
Response: We agree that stable isotope data based on ostracod valves from different families potentially introduce additional uncertainties due to different vital offsets. However, this approach was adopted because ostracod valves from closely related taxa were not sufficiently abundant and continuously present throughout the entire NH-T section. To minimize potential biases, corrected stable isotope values were calculated using a weighted average based on the relative abundance of ostracod valves of each species within individual samples. Although species-specific vital effects may differ, the applied corrections only produced relatively small shifts compared with the magnitude of the isotopic variability observed throughout the NH-T section. For example, the correction for samples containing C. lacustris + Ilyocypris sp. at 80.3 m changed the δ13C values from -7.07‰ to -8.52‰ and the δ18O values from -6.56‰ and -7.47‰, whereas the correction for samples containing L. flexa + Ilyocypris sp. at 77 m changed the δ13C values from -1.91‰ to -2.84‰ and the δ18O values from -3.20‰ to -3.75‰. However, these adjustments did not affect the major trends of the stable isotope variations shown in Figures 2 and 3.
- I recommend that the authors add a far-right column in Figure 2, defining the biozones identified in section 5.1.1 (Evolution between 1.67 to 1.30 Ma, the pre-MPT period). It would help the reader visually distinguish the biozones or biofacies (whichever term the authors prefer)
Response: Thank you for the suggestion. A new column was added to Figure 2 in the updated text.
- I found several references in the references list that weren't cited in the text: Deng et al. (2006a, b), Deng et al. (2007), Li et al. (2015), Mischke et al. (2021), Tian et al. (2002), Tian et al. (2004), Wang et al. (2005), Wang et al. (2010), Xie and Cheng (1990)
Response: Thank you for highlighting this issue. We have checked the reference list and corrected these inconsistencies in the revised manuscript.
Citation: https://doi.org/10.5194/egusphere-2025-5999-AC2
-
AC2: 'Reply on RC2', Ahmed Moghazi, 23 Jul 2026
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 1,174 | 473 | 112 | 1,759 | 128 | 252 |
- HTML: 1,174
- PDF: 473
- XML: 112
- Total: 1,759
- BibTeX: 128
- EndNote: 252
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Moghazi and colleagues present a new stable isotope analysis of deposits from the NH-T section in the Nihewan Basin, northern China. Given that my background is in paleoanthropology and not stable isotopes or ostracod shells I restrict my comments to the general paleoanthropological framework in which the paper is set. My comments are minor and are intended to better set the more up-to-date framework of the study in which the Nihewan Basin sits in Early Pleistocene paleoanthropological research. The paper is clearly acceptable pending minor revisions, depending of course on the stable isotope analysis.
Lines 47-50. Rewrite simply as “Today, it represents one of the densest concentrations of Early Pleistocene Palaeolithic sites outside of Africa”. Here, I would also add Dennell (2009) and Bae (2024) who both spend some time talking about the importance of the Nihewan Basin in their books. I would also recommend citing in here someplace the recent Bae and Manthey (2025) paper as it points out that the Nihewan Basin is no longer considered to be a really “early” region for hominin occupation in eastern Asia. Sites like Xihoudu, Shangchen, and Longgupo are clearly older and the recent dating of the Yunxian Homo erectus site (Tu et al., 2026) is also older than the archaeology sites in the Nihewan Basin.
Lines 50-54. This needs to be rewritten given the evidence from Xihoudu (not very far south of the Nihewan Basin) is dated to 2.43 Ma and Shangchen dates to 2.1 Ma. Mention should also be made someplace in here that most of the Nihewan Basin sites ages are based on relative dating techniques (paleomag, biostrat), while some sites like Xiashagou and Xujiayao (Tu et al., 2022 [cited in paper]; Shao et al., 2026 [not cited in paper]) have been radiometrically dated.
Line 54. Start new paragraph with the first sentence (“Based on their…”).
Lines 68-71. Going out to the third digit seems a bit too overly precise, particularly given that these are likely paleomag dates. Recommend just going out to the second digit as presented in most paleomag studies in eastern Asia.
Line 73. Rewrite “1.3 Ma” to “1.3 millions of years”. Technically, most readers read “Ma” as millions of years ago, but in this context you are only referring to a span of 1.3 millions of years.
Lines 128-130. You should add some justification and clarification as to what revisions you made to the age of the NH-T section in this study. As written, you leave the reader hanging, just saying you did it. But based on what?
Line 285. Provide the age range in brackets for these S and L layers unless it includes the entire time bracket for this section.
Line 296. Provide the age range in brackets for these S and L layers unless it includes the entire time bracket for this section.
Line 308. Ibid.
Section 5.1. It seems that your data crosscut both S and L layers. Is it not possible to determine if there are significant different in isotope ratios between S and L? Given the clear environmental differences between S and L layers, how can you explain then that your data is the same for both S and L?
Line 394. Replace “back from” with “between”. And given that Xujiayao is a part of the Nihewan Basin, this time spread should be pushed back to 160 ka (Shao et al., 2026).
References
Bae, C.J., 2024. The Paleoanthropology of Eastern Asia. University of Hawaii Press.
Bae, C.J. and Manthey, C., 2025. Out of Africa I revisited: Life history, energetics, and the evolutionary capacity for early hominin dispersals. Quaternary Science Reviews, 370, p.109690.
Dennell, R., 2008. The palaeolithic settlement of Asia. Cambridge University Press.
Shao, Q., Wang, F., Ge, J., Bahain, J.J., Voinchet, P., Hu, G., Jin, X., Grün, R., Bae, C.J. and Song, X., 2026. New U-series and coupled ESR/U-series dating of Xujiayao (northern China), the type site for Homo juluensis. Quaternary Science Reviews, 373, p.109742.
Depending on when this paper gets accepted, I can share the Tu et al., 2026 reference as well. It is currently embargoed.