the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 15 Mar 2021
The 4.2 ka event in East Asian monsoon region, precisely reconstructed by multi-proxies of stalagmite
Abstract. The 4.2 ka event is one of the most salient features of global climate change in the mid-late Holocene and influenced on the evolution of ancient civilizations. Although a lot of paleoclimate reconstructions have focused on it, the detailed structure and driving mechanism of the 4.2 ka event is still unclear. In this study, the variation of Asian summer monsoon (ASM) during 5000–3000 yr BP was reconstructed by using high-precision U-Th dating (average resolution of 7 yr) and multi-proxies (δ13C, δ18O, Ba / Ca, Sr / Ca, Mg / Ca) of stalagmite YK1306 from Yangkou Cave in southwestern China. The results showed that that the ASM weakened and precipitation decreased during 4600–4330 yr BP and 4070–3700 yr BP. During 4330–4070 yr BP, the ASM became strong, and precipitation increased. The multi-proxies variation of YK1306 showed a weak-strong-weak
structure of the ASM during the 4.2 ka event, which reappeared in different geologic records. However, westerlies and Australian-Indian summer monsoon (AISM) both showed the opposite change pattern (strong-weak-strong) with the ASM. This was resulted by the different phases of North Atlantic Oscillation (NAO) on a centennial scale, which regulated by the Atlantic Meridional Overturning Circulation (AMOC). In positive NAO-like, the strength of Azores high and westerly wind restrained the intensity of ASM. Thus, the ASM and the Middle East regions experienced bimodal drought and increased dust flux from the north in both regions during the 4.2 ka event. The strengthened meridional winds in the westerlies-dominated climatic regime (WDCR) lead more water vapor from the Indian Ocean and Arabian Sea transporting to in the WDCR, and subsequently increases precipitation in the WDCR. Meanwhile, the weakening of the AMOC results in the southward migration of the Intertropical Convergence Zone (ITCZ) and strengthens the AISM in the southern Hemisphere, finally results in the opposite change of the AISM contrast to the ASM. In addition, the strong ASM in the era of the Chinese Xia Dynasty maybe produce frequent ancient floods, which led to the decline of Longshan and Liangzhu cultures. The weakening of the ASM after 4070 yr BP contributed to the successful regulation of the ancient floodwaters by Dayu in Chinese history. Therefore, it is maybe credible that the official age for the establishment of the Xia Dynasty in 4070 yr BP. Benefit from the comprehensive comparison and analysis based on the unprecedented high-precise chronology, high-resolution and multi-proxy's stalagmite records, this study not only detailed described the evolution of the ASM during the 4.2 ka event, but also is conducive to verify the age of the first dynasty of China (the Xia Dynasty), and the legend of Dayu.
- Preprint
(1776 KB) - Metadata XML
-
Supplement
(587 KB) - BibTeX
- EndNote
Status: closed
-
CC1: 'Comment on cp-2021-20', Nick Scroxton, 21 Mar 2021
We would like to draw the authors attention to our recent submission in discussion at Climate of the Past (manuscript: cp-2020-138) that may help with the interpretation of the trace element signal from YK1306. Both the unidirectional change in trace elements, and the wet-dry-wet/dry-wet-dry pattern of the stable isotope records have both been identified in our submission, which comprises an Indian Ocean wide regional compilation of high-resolution hydroclimate proxies.
In particular these patterns are seen in the first and second principal components of the compilation (figure 3 of cp-2020-138). Our PC1, which we interpret as variability in Indian Ocean summer rainfall dynamics matches the trace element PC1 signal of YK1306. The decrease in rainfall is timed to 3.97 kyr BP (standard error of 94 years) in our study using a rampfit function and taking into account full age model uncertainty of five independent stalagmite records. A shift in tropical hydroclimate at 4.0 kyr BP is well recognised in the literature (examples: Marchant and Hooghiemstra, 2004, de Boer et al., 2014; Denniston et al., 2013; Gagan et al., 2004; Giosan et al., 2018; Li et al., 2018; MacDonald, 2011; Toth et al., 2012), And although it is sometimes interpreted as being caused by the 4.2 kyr event, both the timing and the shape of the anomaly does not match.
Our PC2 (particularly for the third PCA shown in orange which includes additional records from the Arabian Sea) matches the d18O and d13C signals in YK1306. We interpret those signals as likely deriving from winter monsoon variability as dominated by westerly derived moisture and winds through western disturbances. The triple wet-dry-wet (or dry-wet-dry depending on the loading (figure 4 of cp-2020-138)) pattern is evident too. The timings are approximately 4.6-4.25, 4.25-3.9, 3.9-3.6 kyr BP in our study, in good agreement with YK1306. We therefore believe there is significant congruence between the data of our studies and a reasonable agreement on underlying climatic processes.
With this in mind, one could hypothesise that the trace element signal in YK1306 might therefore reflect some kind of tropical forced variability in rainfall in south-west China via the Indian Summer Monsoon component. Meanwhile the stable isotopes respond, as suggested by the authors, to more westerly dominated climatic processes. We believe that a full consideration of our study in the interpretation of YK1306 will help develop the trace element interpretation in the manuscript, and begin to deconvolve the complex multiple drivers of climate variability in south-west China, and the Indian Ocean. Should you wish to use our data prior to final publication of our work, please contact us.
We will leave the full process of peer review to the appointed reviewers, but we have one other minor point: please make the data from this work publicly available in an archive such as NOAA, and also via a submission to the SISAL database.
Thank-you,
Nick Scroxton
(on behalf of coauthors)
Citation: https://doi.org/10.5194/cp-2021-20-CC1 -
AC1: 'Reply on CC1', chaojun chen, 24 Mar 2021
Dear Dr. Nick Scroxton,
I'm happy to discuss it with you. Happy that we have a common interest in research. I have read your manuscript (cp-2020-138). It is also a new way to analyze the 4.2 ka event recorded by regional paleoclimate by using this Monte Carlo principal component analysis method. On the whole, our viewpoints are similar, and the data can support each other. Thank you for affirming the scientific significance of our work. This will change the understanding of 4.2 ka event in the past. Accurate calibration of the 4.2ka event in low latitudes is not only helpful to the reconstruction of the paleoclimate, but also contribute to the evolution of ancient civilization.
We think that the variation of trace elements in YK1306 is well matched with isotope (Figure 3 and Figure S1), and PC1 is significantly correlated with isotope (Table S1). The slight difference in detail is due to the different resolution of trace elements and isotopes. Please see the section of Method. In addition, the Yangkou Cave is located on the top of the Jinfo Mountain (2100-2250 m a.s.l), about 1600 m higher that the Sichuan Basin. The Jinfo Mt. is a national reservation park, there is now industry, nor much anthropic activities, and the regional vegetation is subtropical evergreen broad-leaved forest. In addition, we monitored the trace elements in rain outside Yangkou cave during the period of 2012-2016. According to our monitoring results (unpublished, in preparing now), the impact of trace elements carried by the Asian summer monsoon is negligible. According to the monitoring work of cave drip water, we think that the variation of speleothems’ elements in this cave and nearby cave are all modulated by the regional hydrological conditions which closely correlated with the Asian summer monsoon (Chao-Jun Chen, Ting-Yong Li*. 2018. Geochemical characteristics of cave drip water respond to ENSO based on a 6-year monitoring work in Yangkou Cave, Southwest China. Journal of Hydrology, 561: 896-907. https://doi.org/10.1016/j.jhydrol.2018.04.061. Jian Zhang, Ting-Yong Li*. 2019. Seasonal and interannual variations of hydrochemical characteristics and stable isotopic compositions of drip waters in Furong Cave, Southwest China based on 12 years’ monitoring. Journal of Hydrology, 572: 40-50. https://doi.org/10.1016/j.jhydrol.2019.02.052.)
We are also very willing to share the original data in NOAA, as we did in the past. While, we hope to upload the data after the acceptance for publishing of this manuscript.
Great appreciation for your comments and constructive suggestions. Any further comment will be welcome.
Thank you very much.
Dr. Chao-Jun Chen and Prof. Ting-Yong Li
(on behalf of coauthors)
Citation: https://doi.org/10.5194/cp-2021-20-AC1
-
AC1: 'Reply on CC1', chaojun chen, 24 Mar 2021
-
RC1: 'Comment on cp-2021-20', Anonymous Referee #1, 14 Apr 2021
This manuscript aims for revealing the evolutionary structure of so-called 4.2 ka event mainly by the use of stalagmite records. The authors argued that the ASM during the 4.2 ka event has a weak-strong-weak structure and was possibly associated with the AMOC. The logical structure of this manuscript is built on a local stalagmite record covering the 5040–2920 yr BP in combination with extant stalagmite records and proxy evidence. In general, the results are not intriguing enough and the reasoning is not rigorous to some extent. In fact, Bradley et al. (2019) doubted the existence of the 4.2 ka event, and thought that it is a local event even in the so-called provenance region, the northern North Atlantic region. Rather, a series of multi-decadal- to century-scale fluctuations which superimposed on an overall decline in temperature occurred over the last 5000 years. My concerns are shown as below.
Bradley, R. S. and Bakke, J.: Is there evidence for a 4.2 ka BP event in the northern North Atlantic region?, Clim. Past, 15, 1665–1676, https://doi.org/10.5194/cp-15-1665-2019, 2019.What is the accurate criterion for the weak-strong-weak structure? How long does the 4.2 ka event last? Is it a millennial scale, multicentennial, century-scale or interdecadal event? I would suggest that the 4.2 ka event should be placed over a much longer perspective rather than such a short span of several hundred years. The association with Chinese Xia Dynasty and regulation of the ancient floodwaters by Dayu in Chinese history is highly speculative and lacks strong evidence.
A vegetation coverage above the cave is needed to facilitate the understanding of proxy mechanism.
Only a stalagmite record is selected to indicate the AISM change. More evidence is needed to verify their argument.
The inclusion of 8.2 ka event is confusing, when ice sheet does not melt entirely compared to 4.2 ka When no freshwater was delivered to the ocean.
Lines 224-227: Why is δ13C in dripping water controlled by surface precipitation considering modern monitoring is intra-annual? Here a mechanism description is needed.
Lines 299-300: the authors selected the lake proxies as validating evidence. However no necessary information is given, especially as for the dating uncertainties.
Line 308: the argument ice debris with the double peak pored into the NA during the 4.2 ka event is ambiguous. From Fig. 5B, it is hard to identify the so-called double peak structure. The same problem applies the argument in the line 309. Why it is the case when the Greenland temperature synchronously decreased during I and III stage.
Line 310: please indicate the period when a large amount of fresh water injected into the NA.
Line 312: would you show the AMOC intensity or NAO-like curves in Fig. 5 for comparison? To be frank, it is hard to believe the so-called weak-strong-weak AMOC structure as simulated by only one model.
Lines 326-328: Why does Yuexi peatland in southwest China record the dust flux from northern China?
Lines 389-398: I don't believe this section deliver any helpful information considering the floods reflected by SWD is not continuous or event-like. Actually, it is highly possible to for floods occurrence every several centuries.
Citation: https://doi.org/10.5194/cp-2021-20-RC1 -
RC2: 'Comment on cp-2021-20', Anonymous Referee #2, 19 Apr 2021
I find this article difficult to review. Throughout both the introduction and discussion, the authors make broad-brush arguments which are overly speculative, and in my opinion are not supported in this manuscript.
The manuscript presents an exceptional stalagmite archive with optimal U and Th characteristics for dating—2s age errors on individual U-Th samples of <20 years, with almost no 230-Th correction needed. The stable isotope record shows large, clear excursions within the ~2000 yr duration period, and additional Mg/Ca, Sr/Ca, and Ba/Ca ratios have also been measured, setting the stage for a strong manuscript. A robust, comprehensive analysis of the measured geochemical proxies is greatly lacking, however, making the manuscript weak, and difficult to review beyond the results section.
Some examples—
--The U-Th data reveal incredibly large variations in U content in the stalagmite over just these 2000 years… from 1,500 ppb to 25,580 ppb! Why is this? A U/Ca record alongside the Mg/Ca, Sr/Ca, and Ba/Ca records would have been a helpful addition.
--The d18O and d13C records on the mean age model show a very abrupt and large shift from heavier values to lighter values around 4330 yr BP, with d13C moving 5‰ lighter and d18O moving 2‰ lighter. [Note, a similar abrupt shift to lighter values appears to occur around 3500 y BP, though about half the amplitude.] The closest U-Th ages to the 4330 y BP shift (also the paper’s stage I to stage II transition boundary) are at 4440 y BP and 4230 y BP (aka 100 y away from the transition), and indeed the age model presented in Figure 2 shows the 95% confidence band to be +/-100 yrs at 4330 y BP. With such a significant transition in the stable isotope record, and an archive able to produce U-Th ages with < +/-20 yr precision, it is a shame to not have a more precise timing on this abrupt shift in stable isotopes.
--The d18O and d13C records show gradual transitions between stages I, II, and III (excluding the stage I to stage II transition)… so how were these stage boundaries’ ages defined?
--By eye it is easy to see the correlation between Sr/Ca and Ba/Ca in Figure S1 (r2 = 0.89, Table S1). A crossplot between the two ratios would be a nice addition, showing the slope. Mg/Ca has two large shifts that dominate its record, which are not replicated in Sr/Ca or Ba/Ca (Figure S1). Mg/Ca does not have a linear correlation with Sr/Ca nor Ba/Ca (r= 0.22 and 0.17, respectively), so I do not see any evidence to support the authors’ claim that all three ratios covary. As the authors chose to do a PC analysis, a P1 loading v P2 loading crossplot would have been useful to conclude which ratios are dominating the different components. The authors instead state (Line 189): “The variance of PC1 is 67.15% (Table 3). Therefore, we chose PC1 as the variable representing Ba/Ca, Sr/Ca, Mg/Ca“. The authors then make the sweeping claim “There was significant correlation between PC1 and δ13C, δ18O, Ba/Ca, Sr/Ca, Mg/Ca (Table S1).” Yet the actual r values between Mg/Ca, d13C, and d18O and PC1 are only 0.48, 0.24, and 0.37, or 23%, 6%, and 14% variance explained, respectively. The authors also state that PC1 (I believe largely Sr/Ca and Ba/Ca) increased during stage I and III, and decreased during stage II, however in Fig. 3 it appears PC1 was gradually decreasing through stage I and II, with a sudden increase at the stage II-III transition, and continued gradual increase through stage III, in contrast with the authors’ claims.
Citation: https://doi.org/10.5194/cp-2021-20-RC2
Status: closed
-
CC1: 'Comment on cp-2021-20', Nick Scroxton, 21 Mar 2021
We would like to draw the authors attention to our recent submission in discussion at Climate of the Past (manuscript: cp-2020-138) that may help with the interpretation of the trace element signal from YK1306. Both the unidirectional change in trace elements, and the wet-dry-wet/dry-wet-dry pattern of the stable isotope records have both been identified in our submission, which comprises an Indian Ocean wide regional compilation of high-resolution hydroclimate proxies.
In particular these patterns are seen in the first and second principal components of the compilation (figure 3 of cp-2020-138). Our PC1, which we interpret as variability in Indian Ocean summer rainfall dynamics matches the trace element PC1 signal of YK1306. The decrease in rainfall is timed to 3.97 kyr BP (standard error of 94 years) in our study using a rampfit function and taking into account full age model uncertainty of five independent stalagmite records. A shift in tropical hydroclimate at 4.0 kyr BP is well recognised in the literature (examples: Marchant and Hooghiemstra, 2004, de Boer et al., 2014; Denniston et al., 2013; Gagan et al., 2004; Giosan et al., 2018; Li et al., 2018; MacDonald, 2011; Toth et al., 2012), And although it is sometimes interpreted as being caused by the 4.2 kyr event, both the timing and the shape of the anomaly does not match.
Our PC2 (particularly for the third PCA shown in orange which includes additional records from the Arabian Sea) matches the d18O and d13C signals in YK1306. We interpret those signals as likely deriving from winter monsoon variability as dominated by westerly derived moisture and winds through western disturbances. The triple wet-dry-wet (or dry-wet-dry depending on the loading (figure 4 of cp-2020-138)) pattern is evident too. The timings are approximately 4.6-4.25, 4.25-3.9, 3.9-3.6 kyr BP in our study, in good agreement with YK1306. We therefore believe there is significant congruence between the data of our studies and a reasonable agreement on underlying climatic processes.
With this in mind, one could hypothesise that the trace element signal in YK1306 might therefore reflect some kind of tropical forced variability in rainfall in south-west China via the Indian Summer Monsoon component. Meanwhile the stable isotopes respond, as suggested by the authors, to more westerly dominated climatic processes. We believe that a full consideration of our study in the interpretation of YK1306 will help develop the trace element interpretation in the manuscript, and begin to deconvolve the complex multiple drivers of climate variability in south-west China, and the Indian Ocean. Should you wish to use our data prior to final publication of our work, please contact us.
We will leave the full process of peer review to the appointed reviewers, but we have one other minor point: please make the data from this work publicly available in an archive such as NOAA, and also via a submission to the SISAL database.
Thank-you,
Nick Scroxton
(on behalf of coauthors)
Citation: https://doi.org/10.5194/cp-2021-20-CC1 -
AC1: 'Reply on CC1', chaojun chen, 24 Mar 2021
Dear Dr. Nick Scroxton,
I'm happy to discuss it with you. Happy that we have a common interest in research. I have read your manuscript (cp-2020-138). It is also a new way to analyze the 4.2 ka event recorded by regional paleoclimate by using this Monte Carlo principal component analysis method. On the whole, our viewpoints are similar, and the data can support each other. Thank you for affirming the scientific significance of our work. This will change the understanding of 4.2 ka event in the past. Accurate calibration of the 4.2ka event in low latitudes is not only helpful to the reconstruction of the paleoclimate, but also contribute to the evolution of ancient civilization.
We think that the variation of trace elements in YK1306 is well matched with isotope (Figure 3 and Figure S1), and PC1 is significantly correlated with isotope (Table S1). The slight difference in detail is due to the different resolution of trace elements and isotopes. Please see the section of Method. In addition, the Yangkou Cave is located on the top of the Jinfo Mountain (2100-2250 m a.s.l), about 1600 m higher that the Sichuan Basin. The Jinfo Mt. is a national reservation park, there is now industry, nor much anthropic activities, and the regional vegetation is subtropical evergreen broad-leaved forest. In addition, we monitored the trace elements in rain outside Yangkou cave during the period of 2012-2016. According to our monitoring results (unpublished, in preparing now), the impact of trace elements carried by the Asian summer monsoon is negligible. According to the monitoring work of cave drip water, we think that the variation of speleothems’ elements in this cave and nearby cave are all modulated by the regional hydrological conditions which closely correlated with the Asian summer monsoon (Chao-Jun Chen, Ting-Yong Li*. 2018. Geochemical characteristics of cave drip water respond to ENSO based on a 6-year monitoring work in Yangkou Cave, Southwest China. Journal of Hydrology, 561: 896-907. https://doi.org/10.1016/j.jhydrol.2018.04.061. Jian Zhang, Ting-Yong Li*. 2019. Seasonal and interannual variations of hydrochemical characteristics and stable isotopic compositions of drip waters in Furong Cave, Southwest China based on 12 years’ monitoring. Journal of Hydrology, 572: 40-50. https://doi.org/10.1016/j.jhydrol.2019.02.052.)
We are also very willing to share the original data in NOAA, as we did in the past. While, we hope to upload the data after the acceptance for publishing of this manuscript.
Great appreciation for your comments and constructive suggestions. Any further comment will be welcome.
Thank you very much.
Dr. Chao-Jun Chen and Prof. Ting-Yong Li
(on behalf of coauthors)
Citation: https://doi.org/10.5194/cp-2021-20-AC1
-
AC1: 'Reply on CC1', chaojun chen, 24 Mar 2021
-
RC1: 'Comment on cp-2021-20', Anonymous Referee #1, 14 Apr 2021
This manuscript aims for revealing the evolutionary structure of so-called 4.2 ka event mainly by the use of stalagmite records. The authors argued that the ASM during the 4.2 ka event has a weak-strong-weak structure and was possibly associated with the AMOC. The logical structure of this manuscript is built on a local stalagmite record covering the 5040–2920 yr BP in combination with extant stalagmite records and proxy evidence. In general, the results are not intriguing enough and the reasoning is not rigorous to some extent. In fact, Bradley et al. (2019) doubted the existence of the 4.2 ka event, and thought that it is a local event even in the so-called provenance region, the northern North Atlantic region. Rather, a series of multi-decadal- to century-scale fluctuations which superimposed on an overall decline in temperature occurred over the last 5000 years. My concerns are shown as below.
Bradley, R. S. and Bakke, J.: Is there evidence for a 4.2 ka BP event in the northern North Atlantic region?, Clim. Past, 15, 1665–1676, https://doi.org/10.5194/cp-15-1665-2019, 2019.What is the accurate criterion for the weak-strong-weak structure? How long does the 4.2 ka event last? Is it a millennial scale, multicentennial, century-scale or interdecadal event? I would suggest that the 4.2 ka event should be placed over a much longer perspective rather than such a short span of several hundred years. The association with Chinese Xia Dynasty and regulation of the ancient floodwaters by Dayu in Chinese history is highly speculative and lacks strong evidence.
A vegetation coverage above the cave is needed to facilitate the understanding of proxy mechanism.
Only a stalagmite record is selected to indicate the AISM change. More evidence is needed to verify their argument.
The inclusion of 8.2 ka event is confusing, when ice sheet does not melt entirely compared to 4.2 ka When no freshwater was delivered to the ocean.
Lines 224-227: Why is δ13C in dripping water controlled by surface precipitation considering modern monitoring is intra-annual? Here a mechanism description is needed.
Lines 299-300: the authors selected the lake proxies as validating evidence. However no necessary information is given, especially as for the dating uncertainties.
Line 308: the argument ice debris with the double peak pored into the NA during the 4.2 ka event is ambiguous. From Fig. 5B, it is hard to identify the so-called double peak structure. The same problem applies the argument in the line 309. Why it is the case when the Greenland temperature synchronously decreased during I and III stage.
Line 310: please indicate the period when a large amount of fresh water injected into the NA.
Line 312: would you show the AMOC intensity or NAO-like curves in Fig. 5 for comparison? To be frank, it is hard to believe the so-called weak-strong-weak AMOC structure as simulated by only one model.
Lines 326-328: Why does Yuexi peatland in southwest China record the dust flux from northern China?
Lines 389-398: I don't believe this section deliver any helpful information considering the floods reflected by SWD is not continuous or event-like. Actually, it is highly possible to for floods occurrence every several centuries.
Citation: https://doi.org/10.5194/cp-2021-20-RC1 -
RC2: 'Comment on cp-2021-20', Anonymous Referee #2, 19 Apr 2021
I find this article difficult to review. Throughout both the introduction and discussion, the authors make broad-brush arguments which are overly speculative, and in my opinion are not supported in this manuscript.
The manuscript presents an exceptional stalagmite archive with optimal U and Th characteristics for dating—2s age errors on individual U-Th samples of <20 years, with almost no 230-Th correction needed. The stable isotope record shows large, clear excursions within the ~2000 yr duration period, and additional Mg/Ca, Sr/Ca, and Ba/Ca ratios have also been measured, setting the stage for a strong manuscript. A robust, comprehensive analysis of the measured geochemical proxies is greatly lacking, however, making the manuscript weak, and difficult to review beyond the results section.
Some examples—
--The U-Th data reveal incredibly large variations in U content in the stalagmite over just these 2000 years… from 1,500 ppb to 25,580 ppb! Why is this? A U/Ca record alongside the Mg/Ca, Sr/Ca, and Ba/Ca records would have been a helpful addition.
--The d18O and d13C records on the mean age model show a very abrupt and large shift from heavier values to lighter values around 4330 yr BP, with d13C moving 5‰ lighter and d18O moving 2‰ lighter. [Note, a similar abrupt shift to lighter values appears to occur around 3500 y BP, though about half the amplitude.] The closest U-Th ages to the 4330 y BP shift (also the paper’s stage I to stage II transition boundary) are at 4440 y BP and 4230 y BP (aka 100 y away from the transition), and indeed the age model presented in Figure 2 shows the 95% confidence band to be +/-100 yrs at 4330 y BP. With such a significant transition in the stable isotope record, and an archive able to produce U-Th ages with < +/-20 yr precision, it is a shame to not have a more precise timing on this abrupt shift in stable isotopes.
--The d18O and d13C records show gradual transitions between stages I, II, and III (excluding the stage I to stage II transition)… so how were these stage boundaries’ ages defined?
--By eye it is easy to see the correlation between Sr/Ca and Ba/Ca in Figure S1 (r2 = 0.89, Table S1). A crossplot between the two ratios would be a nice addition, showing the slope. Mg/Ca has two large shifts that dominate its record, which are not replicated in Sr/Ca or Ba/Ca (Figure S1). Mg/Ca does not have a linear correlation with Sr/Ca nor Ba/Ca (r= 0.22 and 0.17, respectively), so I do not see any evidence to support the authors’ claim that all three ratios covary. As the authors chose to do a PC analysis, a P1 loading v P2 loading crossplot would have been useful to conclude which ratios are dominating the different components. The authors instead state (Line 189): “The variance of PC1 is 67.15% (Table 3). Therefore, we chose PC1 as the variable representing Ba/Ca, Sr/Ca, Mg/Ca“. The authors then make the sweeping claim “There was significant correlation between PC1 and δ13C, δ18O, Ba/Ca, Sr/Ca, Mg/Ca (Table S1).” Yet the actual r values between Mg/Ca, d13C, and d18O and PC1 are only 0.48, 0.24, and 0.37, or 23%, 6%, and 14% variance explained, respectively. The authors also state that PC1 (I believe largely Sr/Ca and Ba/Ca) increased during stage I and III, and decreased during stage II, however in Fig. 3 it appears PC1 was gradually decreasing through stage I and II, with a sudden increase at the stage II-III transition, and continued gradual increase through stage III, in contrast with the authors’ claims.
Citation: https://doi.org/10.5194/cp-2021-20-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 1,829 | 710 | 94 | 2,633 | 175 | 42 | 83 |
- HTML: 1,829
- PDF: 710
- XML: 94
- Total: 2,633
- Supplement: 175
- BibTeX: 42
- EndNote: 83
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
