the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 21 Jan 2025
ESD Ideas: Cenozoic Ice Volume as a Driver of Geomagnetic Events
Abstract. This study investigates the relationship between Cenozoic ice volume changes and geomagnetic events, including reversals and incomplete reversals, revealing that high frequencies of these events responded to ice volume increases over the past 49 million years. Geomagnetic events forming chrons or wiggles shorter than 0.1 Myr are particularly sensitive to ice volume changes. The findings suggest that future global warming could suppress geomagnetic activity, highlighting the impact of climate-driven ice volume changes on Earth's magnetic field dynamics.
- Preprint
(689 KB) - Metadata XML
- BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on esd-2024-43', Nicolas Thouveny, 29 Jan 2025
REVIEW OF Cenozoic Ice Volume as a Driver of Geomagnetic Events
By JAISCHENG CHEN
In this very short paper the author attempts to establish a correlation between the Ice volume variations and the frequency of the Earth magnetic field reversals during the Cenozoic Era. A speculative discussion follows linking the increase of ice volume to high frequency reversals particularly short subchrons (. The conclusion forecasts a possible decrease of geomagnetic activity due to the future global warming.
This paper suffers from several and severe defaults:
1) The introduction is a very short summary of distinct fundamental concepts of geophysics and paleoclimatology, each illustrated by one single reference : Gubbins, 2008; Doake 1977; Zachos et al. 2001.
2) The paleomagnetic event serie used for this study is identified only by the refs: Cande and Kent, 1992, Gee and Kent 2007 ( which one is used ?). No graphical presentation, nor description is provided to help the reader understand the data sets and the differences between the studied clusters ( duration < 0.03 Ma or >0.03 Ma , <0.1 Ma or > 0.1 Ma).
3) The Doake (1977) reference provides the basic physical concept of the article: polar ice caps provoke an increase of Earth’s rotation speed (conservation of the angular momentum) and thus an increase of the outer core fluid motion, providing more energy to the dynamo, resulting in stronger geomagnetic field .
However this old and single ref should also be questioned and completed by other references.
- The Doake (1977) hypothesis (in fact firstly introduced in 1975 by Olausson and Svenonius) was weakened by Kent (Nature 1982) who denounced the paleoclimatic biases on the natural remanent magnetization intensity records in sediment cores supposed (at this time) to represent the geomagnetic field intensity). Few experimental studies however supported the idea that the Earth’s rotation speed acts on the geomagnetic field intensity ( e.g. Miyagoshi and Hamano, in Phys. Rev. Lett., 2013).
However, several other studies suggested a causal link on the EMF intensity by orbital forcing ( precession, obliquity and eccentricity): see Fuller 2006, Thouveny et al. 2008, or Zhou et al. 2023 (and references therein).
4) Some studies listed together in line 28 and 29 are presented as if they all agreed. This is wrong: they provided different (sometimes contradictory) observations and proposed different hypotheses: Fuller 2006 draw a relationship between the occurrence of reversals and the obliquity period; Thouveny et al. 2008 observed that the majority of dipole moment lows and excursions of the last 800 000 years occurred within interglacial episodes, in coincidence with obliquity minima. The most recent high resolution studies paleomag and cosmonuclide isotopes along sediment sequences (e.g. Simon et al. 2016, 2018, 2020), as well as high precision Ar/Ar dating of excursional lava flow series allowed to demonstrate that the last reversal occurred during an interglacial and that most excursions of the last 800 ka occurred in narrow relation with interglacials. This totally cancels the hypothesis of Worm (1997) claiming that excursions occurred during glaciations.
5) Macroscopically for the past 50 Ma, the correlation seems to be supported: frequency of reversals (chron durations shorter than 0.1 Ma) and Ice volume seem to present a significant correlation coefficient. However several important discrepancies appear at ca 20 Ma, 10 Ma and mostly for the last Ma.
Moreover, during the 50 - 60 Ma interval, the massive occurrence of reversals pointing at 55 Ma corresponds to an Ice free Greenhouse world. This observation totally kills the initial hypothesis and forces the author (line 69-75) to find another mechanism (catastrophic water distribution and mass transfers…) that are absolutely not explained, neither constrained.
6) The most fundamental criticism is based on an obvious and revealing contradiction in
lines 9-10 and lines 82 – 85.
Indeed, the sentence (line 82): “such accelerated melting would decrease the occurrence of geomagnetic events”, is in complete contradiction with the sentence (line 83): “…relationship between ice sheet melting … and the increase in geomagnetic events”, the later itself contradicting the last sentence of the abstract (line 9 and 10).
This contradiction points the melting confusion of two different concepts: “occurrence of geomagnetic events” and “geomagnetic activity”.
7) Over the last decades, paleomagnetic studies of lava flows and sediments (e.g. Valet, Meynadier, Guyodo, 2005 in Nature), completed by cosmogenic nuclide studies (e.g. Simon et al. 2016, JGR; 2018, 2020 EPSL; Valet et al. 2024, QSR) demonstrate that reversals and excursions are associated (or even triggered by) dipole moment collapses. Therefore, at multi-million years scales, frequent reversals imply more frequent time intervals of weak dipole field.
This article is based on the hypothesis that the increase of the Earth’s speed rotation due to the mass accumulation in the polar regions (formation of ice caps) increases the energy of fluid motion in the outer core. But this mode should increase the geomagnetic dipole moment and thus provide a stronger stability of the dipole field polarity, i.e. a lower frequency of reversals. On the contrary the author claims that the glaciations (resp. deglaciations) of polar regions are responsible for increases (resp. decreases) of reversal frequency.
Conclusion : The author’s observation that higher frequency of reversals correlates with heavier polar ice masses is in contradiction with the principle of conservation of angular momentum that imply a faster Earth rotation and a stronger dipole field.
Note finally that several other studies suggested that on longer time scales (Ga) the control of the geomagnetic reversal frequency (and occurrence of superchrons), is influenced (or driven) by the heat transfert in the mantle) (e.g. Olson and Amit, 2015, 2019 Frontiers in Earth Science; Franco et al. 2019, Nature Sci. Reports).
Citation: https://doi.org/10.5194/esd-2024-43-RC1 -
AC1: 'Reply on RC1', Jiasheng Chen, 27 Mar 2025
In response to Reviewer 1’s Comment 5, which states that “Macroscopically for the past 50 Ma, the correlation seems to be supported: frequency of reversals and ice volume seem to present a significant correlation coefficient,” we have revised our manuscript to focus exclusively on the long-term, million-year-scale relationship between ice volume and geomagnetic reversals.
Additionally, Comments 3 and 4 raised concerns regarding the uncertainty in the relationship between glacial cycles and Earth's magnetic field at orbital timescales. Given this, we have removed discussions related to geomagnetic chrons shorter than 0.1 Ma, which are tied to the ~100 kyr eccentricity cycle of Milankovitch forcing, as well as any claims regarding a causal link between orbital forcing and variations in geomagnetic field intensity.
The revised manuscript now focuses solely on chrons and cryptochrons, which are distinguished based on a 0.03 Myr duration threshold used in the construction of the Geomagnetic Polarity Time Scale (GPTS). Chrons serve as the foundation of the GPTS, while cryptochrons are excluded from it.
By considering both chrons and cryptochrons, our analysis reveals a significant positive correlation between the frequency of geomagnetic events and Cenozoic ice volume changes. This suggests that, on million-year timescales, variations in ice volume may have influenced Earth's magnetic field dynamics.
Below are the point-by-point responses to Reviewer 1's comments, where each comment is followed by the corresponding response:
Comment 1: REVIEW OF Cenozoic Ice Volume as a Driver of Geomagnetic Events
By JAISCHENG CHEN
In this very short paper the author attempts to establish a correlation between the Ice volume variations and the frequency of the Earth magnetic field reversals during the Cenozoic Era. A speculative discussion follows linking the increase of ice volume to high frequency reversals particularly short subchrons (. The conclusion forecasts a possible decrease of geomagnetic activity due to the future global warming.
This paper suffers from several and severe defaults:
The introduction is a very short summary of distinct fundamental concepts of geophysics and paleoclimatology, each illustrated by one single reference : Gubbins, 2008; Doake 1977; Zachos et al. 2001.
Answer 1: The introduction has been rewritten to include more studies on the relationship between ice volume and the geomagnetic field, research focused on the Quaternary period, and our investigation of the Cenozoic. The new introduction is as follows:
Changes in Earth's geomagnetic field, including reversals, are linked to alterations in the outer liquid core (Gubbins, 2008). Global ice volume variations, primarily driven by Antarctic and Greenland ice sheets, influence Earth's rotation through the conservation of angular momentum. When the global ice volume increases, a significant amount of water is stored on land as glaciers, which alters the Earth's mass distribution and leads to an increase in its moment of inertia. According to the conservation of angular momentum, in the absence of external torques, this increase in the moment of inertia causes a reduction in the Earth's rotation rate. Following the establishment of ice sheets, vertical movements and internal adjustments occur due to changes in surface loading. As a result, the Earth's flattening increases, leading to a reduction in its rotational speed (Peltier, 2004). However, there remains controversy regarding the relationship between Earth's rotation rate and its magnetic dipole moment after the formation of glaciers, as derived from simple formulas and magnetohydrodynamic geodynamo simulations (Olausson and Svenonius, 1975; Doake, 1977; Takehiro and Yozo, 2013). Despite this, these theoretical studies suggest that the establishment or demise of ice sheets may be sufficient to trigger geomagnetic reversals.
At orbital timescales, ice sheet variations are primarily governed by Milankovitch cycles. Over the past 870 kyr, geomagnetic field variations and Asian monsoon precipitation records have exhibited a strong correlation with orbital eccentricity, with a particularly high coherence at the 100 kyr periodicity. Changes in ice volume driven by orbital eccentricity influence Earth's rotational velocity, which in turn affects geomagnetic variability, leading to a complex phase response (Zhou et al., 2023). Some studies have also suggested a causal link between orbital forcing and geomagnetic field intensity (Fuller, 2006; Thouveny et al., 2008). Marine sediment reconstructions of geomagnetic paleointensity over the past 1.5 Myr, when compared with oxygen isotope records, reveal significant coherence at certain time intervals and frequency bands. However, the varying phase relationship between these records suggests a potential nonlinear connection or merely a coincidental coherence between two independent signals within overlapping frequency bands. As a result, no consensus has been reached regarding the relationship between ice sheet dynamics and geomagnetic field variability throughout the Quaternary.
Since the Cenozoic, Earth's climate has transitioned from a greenhouse to an icehouse state. Over million-year timescales, global ice volume has undergone significant fluctuations, such as the initiation of ice sheets during the Eocene-Oligocene transition and the expansion of the East Antarctic Ice Sheet in the Miocene (Zachos et al., 2001). However, the potential link between these long-term ice volume changes and geomagnetic variability remains largely unexplored.
Comment 2: The paleomagnetic event serie used for this study is identified only by the refs: Cande and Kent, 1992, Gee and Kent 2007 ( which one is used ?). No graphical presentation, nor description is provided to help the reader understand the data sets and the differences between the studied clusters ( duration < 0.03 Ma or >0.03 Ma , <0.1 Ma or > 0.1 Ma).
Answer 2: The chron data used in this study are derived from Gee and Kent (2007), which represents an update of the Geomagnetic Polarity Time Scale originally established by Cande and Kent (1992), with the primary differences occurring before the Cenozoic. Cryptochron data, on the other hand, are sourced from Cande and Kent (1992). During the construction of the Geomagnetic Polarity Time Scale, cryptochrons—defined as geomagnetic events with durations of less than 0.03 Ma—were classified as geomagnetic "tiny wiggles" and were excluded from the formal GPTS, as they may be associated with incomplete reversals. Instead, they were listed separately in the original dataset. However, since cryptochrons also represent geomagnetic anomaly signals, they are included in our analysis to provide a more comprehensive assessment of geomagnetic variability.
The revised manuscript is as follows:
As new seafloor forms and cools at mid-ocean ridges, it captures records of Earth's magnetic field. Polarity chrons, which represent intervals between magnetic reversals, serve as the basis for the Geomagnetic Polarity Time Scale (Cande and Kent, 1992). Gee and Kent (2007) updated the geomagnetic polarity reversal record based on marine magnetic anomalies, with significant revisions to the M-sequence of the Mesozoic. However, these adjustments fall outside the temporal scope of our study. To distinguish these chrons from shorter-duration anomalies, a 0.03 Myr threshold is arbitrarily applied. Anomalies shorter than this threshold, referred to as cryptochrons, correspond to brief intervals between reversals or incomplete reversals. Incomplete reversals are primarily associated with fluctuations in magnetic intensity and direction. Datasets on cryptochrons over the Cenozoic Era are also publicly available (Cande and Kent, 1992). All geomagnetic records are derived from magnetic anomalies preserved in mid-ocean ridge basalts. Using these records as a basis for reconstructing geomagnetic field variations minimizes potential climatic interference. Both reversals and incomplete reversals are collectively defined as geomagnetic events in this paper.
Previous studies have primarily used the frequency of geomagnetic reversals (Biggin et al., 2012; Pétrélis et al., 2011) to reflect changes in Earth's magnetic field. In contrast, we also consider incomplete reversals and employ a similar methodology to calculate the frequency of geomagnetic events (FGE). It was determined using a moving window approach with a 2 Myr window width and 1 Myr increments. Three groups of geomagnetic events (FGE) were categorized using a cutoff duration of 0.03 Myr (Fig. 1a–c). This duration serves as the threshold that distinguishes these chrons from cryptochrons. For example, the calculation of FGE>0.03 involves listing the chrons from the GPTS, each of which has a duration greater than 0.03 Myr. Each chron begins and ends with a polarity reversal or geomagnetic event. These geomagnetic events are then arranged chronologically, and the moving window method is applied to compute the FGE within each window. This approach is similarly applied to all other FGE calculations.
Comment 3: The Doake (1977) reference provides the basic physical concept of the article: polar ice caps provoke an increase of Earth’s rotation speed (conservation of the angular momentum) and thus an increase of the outer core fluid motion, providing more energy to the dynamo, resulting in stronger geomagnetic field .
However this old and single ref should also be questioned and completed by other references.
- The Doake (1977) hypothesis (in fact firstly introduced in 1975 by Olausson and Svenonius) was weakened by Kent (Nature 1982) who denounced the paleoclimatic biases on the natural remanent magnetization intensity records in sediment cores supposed (at this time) to represent the geomagnetic field intensity). Few experimental studies however supported the idea that the Earth’s rotation speed acts on the geomagnetic field intensity ( e.g. Miyagoshi and Hamano, in Phys. Rev. Lett., 2013).
However, several other studies suggested a causal link on the EMF intensity by orbital forcing ( precession, obliquity and eccentricity): see Fuller 2006, Thouveny et al. 2008, or Zhou et al. 2023 (and references therein).
Answer 3: Doake (1977), Olausson and Svenonius (1975), and Takehiro and Yozo (2013) all suggested that the establishment or melting of ice sheets could have a significant impact on Earth's magnetic field. However, the phase relationship between changes in Earth's rotation rate and geomagnetic field intensity remains unresolved.
Several other studies have proposed a causal link between orbital forcing (precession, obliquity, and eccentricity) and variations in geomagnetic field intensity (Fuller, 2006; Thouveny et al., 2008; Zhou et al., 2023, and references therein). These studies are primarily focused on the Quaternary period and examine geomagnetic variations on orbital timescales.
In contrast, our revised study focuses on million-year-scale variations during the Cenozoic. Nevertheless, we have summarized relevant Quaternary research in the introduction to provide context, as noted in Comment 1.
Comment 4: Some studies listed together in line 28 and 29 are presented as if they all agreed. This is wrong: they provided different (sometimes contradictory) observations and proposed different hypotheses: Fuller 2006 draw a relationship between the occurrence of reversals and the obliquity period; Thouveny et al. 2008 observed that the majority of dipole moment lows and excursions of the last 800 000 years occurred within interglacial episodes, in coincidence with obliquity minima.
The most recent high resolution studies paleomag and cosmonuclide isotopes along sediment sequences (e.g. Simon et al. 2016, 2018, 2020), as well as high precision Ar/Ar dating of excursional lava flow series allowed to demonstrate that the last reversal occurred during an interglacial and that most excursions of the last 800 ka occurred in narrow relation with interglacials. This totally cancels the hypothesis of Worm (1997) claiming that excursions occurred during glaciations.
Answer 4: On orbital timescales, the relationship between ice volume fluctuations and geomagnetic field variations during the Quaternary remains inconclusive. The revised manuscript shifts focus to longer, million-year timescales, examining the connection between large-scale ice sheet formation and retreat and Earth's magnetic field dynamics.
Comment 5: Macroscopically for the past 50 Ma, the correlation seems to be supported: frequency of reversals (chron durations shorter than 0.1 Ma) and Ice volume seem to present a significant correlation coefficient. However several important discrepancies appear at ca 20 Ma, 10 Ma and mostly for the last Ma.
Moreover, during the 50 - 60 Ma interval, the massive occurrence of reversals pointing at 55 Ma corresponds to an Ice free Greenhouse world. This observation totally kills the initial hypothesis and forces the author (line 69-75) to find another mechanism (catastrophic water distribution and mass transfers…) that are absolutely not explained, neither constrained.
Answer 5: In response to the reviewers’ comments, the revised manuscript focuses exclusively on the past 50 Ma, examining the macro-scale relationship between the frequency of geomagnetic events and ice volume. For all geomagnetic events—including polarity changes associated with both chrons and cryptochrons—the correlation coefficient between their occurrence frequency and ice volume reaches 0.76.
However, slight discrepancies are observed at ca. 20 Ma, 10 Ma, and the last 2 Ma. Given that geomagnetic reversals are influenced by multiple factors—including core-mantle boundary heat flux distribution, mantle convection, the formation of lower mantle superplumes, heterogeneity in lower mantle electrical conductivity, and plate tectonics (Amit and Olson, 2015; Olson and Amit, 2015; Franco et al., 2019; Li et al., 2016; Biggin et al., 2012)—ice sheet dynamics alone cannot fully account for every detail of geomagnetic field variations.
The Cenozoic marks a critical period in Earth's tectonic evolution, during which major continental masses migrated to their present-day configurations. Notably, the collision of the Indian subcontinent with Asia between 55 and 45 million years ago led to the formation of the Himalayan orogeny. Concurrently, around 50 million years ago, the Solar System underwent a chaotic orbital transition, characterized by orbital instability and chaotic diffusion. This astronomical perturbation has been implicated in major climatic events, including the PETM (Zeebe and Lourens, 2019). Given the fundamental differences in plate configurations, mantle dynamics, and astronomical parameters before and after 50 Ma, the variations in Earth's magnetic field prior to 50 Ma may need to be interpreted in the context of plate tectonics (Pétrélis et al., 2011), mantle convection processes (Biggin et al., 2012; Franco et al., 2019), and their influence on the geodynamo. This also highlights the complexity of the mechanisms underlying Earth's magnetic field variations.
However, our investigation into the relationship between ice volume and the geomagnetic field during the Cenozoic provides a new perspective for understanding the mechanisms driving geomagnetic variability. It is only during the Cenozoic icehouse climate that the large fluctuations in ice volume have had a more pronounced impact on Earth's magnetic field.
Comment 6: The most fundamental criticism is based on an obvious and revealing contradiction in
lines 9-10 and lines 82 – 85.
Indeed, the sentence (line 82): “such accelerated melting would decrease the occurrence of geomagnetic events”, is in complete contradiction with the sentence (line 83): “…relationship between ice sheet melting … and the increase in geomagnetic events”, the later itself contradicting the last sentence of the abstract (line 9 and 10).
This contradiction points the melting confusion of two different concepts: “occurrence of geomagnetic events” and “geomagnetic activity”.
Answer 6: As explained in Answer 3, the theoretical connection between ice sheet formation and retreat and geomagnetic field variations has been outlined. However, the phase relationship between these two factors, as reported in different studies, is not consistent. On the Cenozoic scale, our results show that periods of higher ice volume are associated with increased frequency of geomagnetic reversals. The erroneous statements related to this have been corrected in the revised manuscript.
Comment 7:Over the last decades, paleomagnetic studies of lava flows and sediments (e.g. Valet, Meynadier, Guyodo, 2005 in Nature), completed by cosmogenic nuclide studies (e.g. Simon et al. 2016, JGR; 2018, 2020 EPSL; Valet et al. 2024, QSR) demonstrate that reversals and excursions are associated (or even triggered by) dipole moment collapses. Therefore, at multi-million years scales, frequent reversals imply more frequent time intervals of weak dipole field.
This article is based on the hypothesis that the increase of the Earth’s speed rotation due to the mass accumulation in the polar regions (formation of ice caps) increases the energy of fluid motion in the outer core. But this mode should increase the geomagnetic dipole moment and thus provide a stronger stability of the dipole field polarity, i.e. a lower frequency of reversals. On the contrary the author claims that the glaciations (resp. deglaciations) of polar regions are responsible for increases (resp. decreases) of reversal frequency.
Conclusion : The author’s observation that higher frequency of reversals correlates with heavier polar ice masses is in contradiction with the principle of conservation of angular momentum that imply a faster Earth rotation and a stronger dipole field.
Note finally that several other studies suggested that on longer time scales (Ga) the control of the geomagnetic reversal frequency (and occurrence of superchrons), is influenced (or driven) by the heat transfert in the mantle) (e.g. Olson and Amit, 2015, 2019 Frontiers in Earth Science; Franco et al. 2019, Nature Sci. Reports).
Answer 7: Theoretical calculations suggest that the formation and retreat of ice sheets are sufficient to trigger geomagnetic reversals (Olausson and Svenonius, 1975; Doake, 1977; Takehiro and Yozo, 2013). However, the phase relationship between these two factors remains unresolved. Our study on Cenozoic ice volume and geomagnetic reversal frequency indicates that periods of high ice volume correspond to increased geomagnetic reversal events. We have identified this pattern, though the specific mechanisms still require further validation through additional model results.
Several other studies have suggested that, on longer timescales (Ga), the frequency of geomagnetic reversals—and the occurrence of superchrons—are influenced (or driven) by heat transfer in the mantle. As outlined in Answer 5, while ice volume has a significant impact on the geomagnetic field during the Cenozoic, prior to this period, factors such as plate distribution and Earth's orbit differed substantially, and in the absence of ice sheets, other factors become more pronounced in influencing the geomagnetic field.
Citation: https://doi.org/10.5194/esd-2024-43-AC1
-
AC1: 'Reply on RC1', Jiasheng Chen, 27 Mar 2025
-
RC2: 'Comment on esd-2024-43', M.J. Dekkers, 02 Feb 2025
Review Chen “Cenozoic Ice Volume as a Driver of Geomagnetic Events”
For Earth System Dynamics
The manuscript discusses a potential relation between ice volume and what is referred to as ‘geomagnetic events’, with a focus on short duration events lasting < 0.1 Myr (whether this would be short remains an unanswered question). The current lay-out suffers from several serious issues listed below (random order).
The idea seems to be inspired from Doake (1977) which is not a recent publication on this issue. Potential newer ideas (agreeing or disagreeing with Doake) should be reviewed. Referencing stops at the turn of the 21st century, no newer relevant work on the topic?
The data quality should be assessed; at present they are taken at face value from old sources. This applies at least to the geomagnetic events time series.
Are the ice volume and geomagnetic data placed on the same time scale? For frequency analysis this probably does not matter too much but here data seem to be plotted vs time.
The number of short geomagnetic events shorter 0.03 Myr (excursions or reversal excursions) may be grossly underestimated in the geomagnetic time series. Before 2 Ma or so, the existing data base (as off 2025) is very incomplete.
The 0.1 Myr ice age period applies only after the Mid-Pleistocene Transition (1.2-0.6 Ma). It may vary in older periods.
How the data were processed and with which software package(s) remains unclear.
There is no assessment of the contrast between icehouse and greenhouse state. Visually there is a similar amount of variability in Figure 1h. The claim that minimal geomagnetic variation would occur during the icehouse-greenhouse transition is not assessed. It reads as merely postulated.
What about the Jurassic (greenhouse) as an example? Reversal frequency is higher than 10x per Myr; this is full reversals, no information is available about excursions. If ice volume is a driver, what is the reason for geomagnetic variability during greenhouse times?
The implications/inferences should be assessed with the most meaningful data sets (why are the selected references the best?) and not against a few seemingly randomly picked references.
With the geomagnetic event data base being incomplete (short events are bound to be under-represented) inferences on links between future climate and geomagnetic field variability are on thin ice.
Future global warming is sometimes equated to heading to a uni-polar glacial state but whether completely ice-free conditions will be reached is an open question.
Citation: https://doi.org/10.5194/esd-2024-43-RC2 -
AC2: 'Reply on RC2', Jiasheng Chen, 27 Mar 2025
Comment 1: The manuscript discusses a potential relation between ice volume and what is referred to as ‘geomagnetic events’, with a focus on short duration events lasting < 0.1 Myr (whether this would be short remains an unanswered question). The current lay-out suffers from several serious issues listed below (random order).
The idea seems to be inspired from Doake (1977) which is not a recent publication on this issue. Potential newer ideas (agreeing or disagreeing with Doake) should be reviewed. Referencing stops at the turn of the 21st century, no newer relevant work on the topic?
Answer 1: In response to the reviewer's suggestion, we have updated the references. In addition to the theoretical calculations by Olausson and Svenonius (1975) and Doake (1977), the study by Takehiro and Yozo (2013) also conducted magnetohydrodynamic geodynamo simulations. All of these studies demonstrate that changes in ice volume and Earth's rotation have a significant impact on the geomagnetic field.
Comment 2: The data quality should be assessed; at present they are taken at face value from old sources. This applies at least to the geomagnetic events time series.
Are the ice volume and geomagnetic data placed on the same time scale? For frequency analysis this probably does not matter too much but here data seem to be plotted vs time.
The number of short geomagnetic events shorter 0.03 Myr (excursions or reversal excursions) may be grossly underestimated in the geomagnetic time series. Before 2 Ma or so, the existing data base (as off 2025) is very incomplete.
The 0.1 Myr ice age period applies only after the Mid-Pleistocene Transition (1.2-0.6 Ma). It may vary in older periods.
How the data were processed and with which software package(s) remains unclear.
Answer 2: All geomagnetic records are derived from magnetic anomalies preserved in mid-ocean ridge basalts. Using these records as a basis for reconstructing geomagnetic field variations minimizes potential climatic interference.
The Geomagnetic Polarity Time Scale (GPTS) was initially established by Cande and Kent (1992) based on oceanic magnetic anomalies. The record was later updated by Gee and Kent (2007), with significant revisions focusing on the M-series of the Mesozoic. While our study primarily concentrates on the Cenozoic, we have adopted the Gee and Kent (2007) model. Cryptochrons, defined as magnetic anomalies with a duration of less than 0.03 Myr, were excluded from the Geomagnetic Polarity Time Scale when it was first established by Cande and Kent (1992). However, these cryptochrons are listed separately in the literature. Currently, no updated studies on these cryptochrons exist. Therefore, the data on geomagnetic field changes is exclusively drawn from Gee and Kent (2007) and Cande and Kent (1992).
Ice volume data has a resolution of 1 Myr, and the frequency of geomagnetic events is also calculated with a 1 Myr time step. The comparison between the two datasets is conducted on the same temporal scale, which satisfies the resolution requirements for our analysis.
It is possible that the number of cryptochrons before 2 Ma is underestimated, but no new data can confirm this. The study uses only mid-ocean ridge basalt records for cryptochrons, and no updated data sources are available.
The eccentricity cycle of 0.1 Myr becomes significant only after the Mid-Pleistocene Transition in the ice volume records. Reviewer 1 also raised concerns about the connection between ice volume and geomagnetic field changes on the orbital scale. Consequently, we have removed related studies on orbital-scale changes and focused on the relationship between ice volume and geomagnetic field variations on the million-year scale.
The calculation of the frequency of geomagnetic events involves counting geomagnetic events within a sliding window. This process can be handled in Excel, with the main procedure outlined below:Previous studies have primarily used the frequency of geomagnetic reversals (Biggin et al., 2012; Pétrélis et al., 2011) to reflect changes in Earth's magnetic field. In contrast, we also consider incomplete reversals and employ a similar methodology to calculate the frequency of geomagnetic events (FGE). It was determined using a moving window approach with a 2 Myr window width and 1 Myr increments. Three groups of FGE were categorized using a cutoff duration of 0.03 Myr (Fig. 1a–c). This duration serves as the threshold that distinguishes these chrons from cryptochrons. For example, the calculation of FGE>0.03 involves listing the chrons from the GPTS, each of which has a duration greater than 0.03 Myr. Each chron begins and ends with a polarity reversal or geomagnetic event. These geomagnetic events are then arranged chronologically, and the moving window method is applied to compute the FGE within each window. This approach is similarly applied to all other FGE calculations.
Comment 3: There is no assessment of the contrast between icehouse and greenhouse state. Visually there is a similar amount of variability in Figure 1h. The claim that minimal geomagnetic variation would occur during the icehouse-greenhouse transition is not assessed. It reads as merely postulated.
What about the Jurassic (greenhouse) as an example? Reversal frequency is higher than 10x per Myr; this is full reversals, no information is available about excursions. If ice volume is a driver, what is the reason for geomagnetic variability during greenhouse times?
Answer 3: Reviewer 1 raised concerns regarding the relationship between geomagnetic field variations and ice volume prior to 49 Ma. Around 50 Ma, the collision between the Indian plate and the Eurasian continent, coupled with a chaotic orbital transition in the Solar System (Zeebe and Lourens, 2019), resulted in significant geological and orbital changes that distinguish the pre-50 Ma period from subsequent epochs.
Geomagnetic field variations are influenced by various factors, including plate tectonics (Pétrélis et al., 2011), mantle convection processes (Biggin et al., 2012; Franco et al., 2019), and others. Our findings primarily highlight the connection between Cenozoic ice volume and geomagnetic field changes, offering a novel perspective for understanding geomagnetic variations during this period. In the absence of ice volume changes, other factors, such as those mentioned, likely exert a more prominent influence on the Earth's magnetic field.
Comment 4:The implications/inferences should be assessed with the most meaningful data sets (why are the selected references the best?) and not against a few seemingly randomly picked references.
With the geomagnetic event data base being incomplete (short events are bound to be under-represented) inferences on links between future climate and geomagnetic field variability are on thin ice.
Future global warming is sometimes equated to heading to a uni-polar glacial state but whether completely ice-free conditions will be reached is an open question.
Answer 4: As explained in Answer 2, the data have been clarified. Other references have also been updated, including additional studies on the impact of ice volume on the geomagnetic field and other factors influencing geomagnetic variations, as cited in Answer 3. The revised manuscript now exclusively focuses on the relationship between ice volume and geomagnetic field variations on the Cenozoic million-year timescale. Future climate warming involves shorter timescales, and the geomagnetic response to such changes cannot be inferred from the relationship observed over million-year scales. Accordingly, relevant sections have been removed in the updated manuscript.
Citation: https://doi.org/10.5194/esd-2024-43-AC2
-
AC2: 'Reply on RC2', Jiasheng Chen, 27 Mar 2025
Status: closed
-
RC1: 'Comment on esd-2024-43', Nicolas Thouveny, 29 Jan 2025
REVIEW OF Cenozoic Ice Volume as a Driver of Geomagnetic Events
By JAISCHENG CHEN
In this very short paper the author attempts to establish a correlation between the Ice volume variations and the frequency of the Earth magnetic field reversals during the Cenozoic Era. A speculative discussion follows linking the increase of ice volume to high frequency reversals particularly short subchrons (. The conclusion forecasts a possible decrease of geomagnetic activity due to the future global warming.
This paper suffers from several and severe defaults:
1) The introduction is a very short summary of distinct fundamental concepts of geophysics and paleoclimatology, each illustrated by one single reference : Gubbins, 2008; Doake 1977; Zachos et al. 2001.
2) The paleomagnetic event serie used for this study is identified only by the refs: Cande and Kent, 1992, Gee and Kent 2007 ( which one is used ?). No graphical presentation, nor description is provided to help the reader understand the data sets and the differences between the studied clusters ( duration < 0.03 Ma or >0.03 Ma , <0.1 Ma or > 0.1 Ma).
3) The Doake (1977) reference provides the basic physical concept of the article: polar ice caps provoke an increase of Earth’s rotation speed (conservation of the angular momentum) and thus an increase of the outer core fluid motion, providing more energy to the dynamo, resulting in stronger geomagnetic field .
However this old and single ref should also be questioned and completed by other references.
- The Doake (1977) hypothesis (in fact firstly introduced in 1975 by Olausson and Svenonius) was weakened by Kent (Nature 1982) who denounced the paleoclimatic biases on the natural remanent magnetization intensity records in sediment cores supposed (at this time) to represent the geomagnetic field intensity). Few experimental studies however supported the idea that the Earth’s rotation speed acts on the geomagnetic field intensity ( e.g. Miyagoshi and Hamano, in Phys. Rev. Lett., 2013).
However, several other studies suggested a causal link on the EMF intensity by orbital forcing ( precession, obliquity and eccentricity): see Fuller 2006, Thouveny et al. 2008, or Zhou et al. 2023 (and references therein).
4) Some studies listed together in line 28 and 29 are presented as if they all agreed. This is wrong: they provided different (sometimes contradictory) observations and proposed different hypotheses: Fuller 2006 draw a relationship between the occurrence of reversals and the obliquity period; Thouveny et al. 2008 observed that the majority of dipole moment lows and excursions of the last 800 000 years occurred within interglacial episodes, in coincidence with obliquity minima. The most recent high resolution studies paleomag and cosmonuclide isotopes along sediment sequences (e.g. Simon et al. 2016, 2018, 2020), as well as high precision Ar/Ar dating of excursional lava flow series allowed to demonstrate that the last reversal occurred during an interglacial and that most excursions of the last 800 ka occurred in narrow relation with interglacials. This totally cancels the hypothesis of Worm (1997) claiming that excursions occurred during glaciations.
5) Macroscopically for the past 50 Ma, the correlation seems to be supported: frequency of reversals (chron durations shorter than 0.1 Ma) and Ice volume seem to present a significant correlation coefficient. However several important discrepancies appear at ca 20 Ma, 10 Ma and mostly for the last Ma.
Moreover, during the 50 - 60 Ma interval, the massive occurrence of reversals pointing at 55 Ma corresponds to an Ice free Greenhouse world. This observation totally kills the initial hypothesis and forces the author (line 69-75) to find another mechanism (catastrophic water distribution and mass transfers…) that are absolutely not explained, neither constrained.
6) The most fundamental criticism is based on an obvious and revealing contradiction in
lines 9-10 and lines 82 – 85.
Indeed, the sentence (line 82): “such accelerated melting would decrease the occurrence of geomagnetic events”, is in complete contradiction with the sentence (line 83): “…relationship between ice sheet melting … and the increase in geomagnetic events”, the later itself contradicting the last sentence of the abstract (line 9 and 10).
This contradiction points the melting confusion of two different concepts: “occurrence of geomagnetic events” and “geomagnetic activity”.
7) Over the last decades, paleomagnetic studies of lava flows and sediments (e.g. Valet, Meynadier, Guyodo, 2005 in Nature), completed by cosmogenic nuclide studies (e.g. Simon et al. 2016, JGR; 2018, 2020 EPSL; Valet et al. 2024, QSR) demonstrate that reversals and excursions are associated (or even triggered by) dipole moment collapses. Therefore, at multi-million years scales, frequent reversals imply more frequent time intervals of weak dipole field.
This article is based on the hypothesis that the increase of the Earth’s speed rotation due to the mass accumulation in the polar regions (formation of ice caps) increases the energy of fluid motion in the outer core. But this mode should increase the geomagnetic dipole moment and thus provide a stronger stability of the dipole field polarity, i.e. a lower frequency of reversals. On the contrary the author claims that the glaciations (resp. deglaciations) of polar regions are responsible for increases (resp. decreases) of reversal frequency.
Conclusion : The author’s observation that higher frequency of reversals correlates with heavier polar ice masses is in contradiction with the principle of conservation of angular momentum that imply a faster Earth rotation and a stronger dipole field.
Note finally that several other studies suggested that on longer time scales (Ga) the control of the geomagnetic reversal frequency (and occurrence of superchrons), is influenced (or driven) by the heat transfert in the mantle) (e.g. Olson and Amit, 2015, 2019 Frontiers in Earth Science; Franco et al. 2019, Nature Sci. Reports).
Citation: https://doi.org/10.5194/esd-2024-43-RC1 -
AC1: 'Reply on RC1', Jiasheng Chen, 27 Mar 2025
In response to Reviewer 1’s Comment 5, which states that “Macroscopically for the past 50 Ma, the correlation seems to be supported: frequency of reversals and ice volume seem to present a significant correlation coefficient,” we have revised our manuscript to focus exclusively on the long-term, million-year-scale relationship between ice volume and geomagnetic reversals.
Additionally, Comments 3 and 4 raised concerns regarding the uncertainty in the relationship between glacial cycles and Earth's magnetic field at orbital timescales. Given this, we have removed discussions related to geomagnetic chrons shorter than 0.1 Ma, which are tied to the ~100 kyr eccentricity cycle of Milankovitch forcing, as well as any claims regarding a causal link between orbital forcing and variations in geomagnetic field intensity.
The revised manuscript now focuses solely on chrons and cryptochrons, which are distinguished based on a 0.03 Myr duration threshold used in the construction of the Geomagnetic Polarity Time Scale (GPTS). Chrons serve as the foundation of the GPTS, while cryptochrons are excluded from it.
By considering both chrons and cryptochrons, our analysis reveals a significant positive correlation between the frequency of geomagnetic events and Cenozoic ice volume changes. This suggests that, on million-year timescales, variations in ice volume may have influenced Earth's magnetic field dynamics.
Below are the point-by-point responses to Reviewer 1's comments, where each comment is followed by the corresponding response:
Comment 1: REVIEW OF Cenozoic Ice Volume as a Driver of Geomagnetic Events
By JAISCHENG CHEN
In this very short paper the author attempts to establish a correlation between the Ice volume variations and the frequency of the Earth magnetic field reversals during the Cenozoic Era. A speculative discussion follows linking the increase of ice volume to high frequency reversals particularly short subchrons (. The conclusion forecasts a possible decrease of geomagnetic activity due to the future global warming.
This paper suffers from several and severe defaults:
The introduction is a very short summary of distinct fundamental concepts of geophysics and paleoclimatology, each illustrated by one single reference : Gubbins, 2008; Doake 1977; Zachos et al. 2001.
Answer 1: The introduction has been rewritten to include more studies on the relationship between ice volume and the geomagnetic field, research focused on the Quaternary period, and our investigation of the Cenozoic. The new introduction is as follows:
Changes in Earth's geomagnetic field, including reversals, are linked to alterations in the outer liquid core (Gubbins, 2008). Global ice volume variations, primarily driven by Antarctic and Greenland ice sheets, influence Earth's rotation through the conservation of angular momentum. When the global ice volume increases, a significant amount of water is stored on land as glaciers, which alters the Earth's mass distribution and leads to an increase in its moment of inertia. According to the conservation of angular momentum, in the absence of external torques, this increase in the moment of inertia causes a reduction in the Earth's rotation rate. Following the establishment of ice sheets, vertical movements and internal adjustments occur due to changes in surface loading. As a result, the Earth's flattening increases, leading to a reduction in its rotational speed (Peltier, 2004). However, there remains controversy regarding the relationship between Earth's rotation rate and its magnetic dipole moment after the formation of glaciers, as derived from simple formulas and magnetohydrodynamic geodynamo simulations (Olausson and Svenonius, 1975; Doake, 1977; Takehiro and Yozo, 2013). Despite this, these theoretical studies suggest that the establishment or demise of ice sheets may be sufficient to trigger geomagnetic reversals.
At orbital timescales, ice sheet variations are primarily governed by Milankovitch cycles. Over the past 870 kyr, geomagnetic field variations and Asian monsoon precipitation records have exhibited a strong correlation with orbital eccentricity, with a particularly high coherence at the 100 kyr periodicity. Changes in ice volume driven by orbital eccentricity influence Earth's rotational velocity, which in turn affects geomagnetic variability, leading to a complex phase response (Zhou et al., 2023). Some studies have also suggested a causal link between orbital forcing and geomagnetic field intensity (Fuller, 2006; Thouveny et al., 2008). Marine sediment reconstructions of geomagnetic paleointensity over the past 1.5 Myr, when compared with oxygen isotope records, reveal significant coherence at certain time intervals and frequency bands. However, the varying phase relationship between these records suggests a potential nonlinear connection or merely a coincidental coherence between two independent signals within overlapping frequency bands. As a result, no consensus has been reached regarding the relationship between ice sheet dynamics and geomagnetic field variability throughout the Quaternary.
Since the Cenozoic, Earth's climate has transitioned from a greenhouse to an icehouse state. Over million-year timescales, global ice volume has undergone significant fluctuations, such as the initiation of ice sheets during the Eocene-Oligocene transition and the expansion of the East Antarctic Ice Sheet in the Miocene (Zachos et al., 2001). However, the potential link between these long-term ice volume changes and geomagnetic variability remains largely unexplored.
Comment 2: The paleomagnetic event serie used for this study is identified only by the refs: Cande and Kent, 1992, Gee and Kent 2007 ( which one is used ?). No graphical presentation, nor description is provided to help the reader understand the data sets and the differences between the studied clusters ( duration < 0.03 Ma or >0.03 Ma , <0.1 Ma or > 0.1 Ma).
Answer 2: The chron data used in this study are derived from Gee and Kent (2007), which represents an update of the Geomagnetic Polarity Time Scale originally established by Cande and Kent (1992), with the primary differences occurring before the Cenozoic. Cryptochron data, on the other hand, are sourced from Cande and Kent (1992). During the construction of the Geomagnetic Polarity Time Scale, cryptochrons—defined as geomagnetic events with durations of less than 0.03 Ma—were classified as geomagnetic "tiny wiggles" and were excluded from the formal GPTS, as they may be associated with incomplete reversals. Instead, they were listed separately in the original dataset. However, since cryptochrons also represent geomagnetic anomaly signals, they are included in our analysis to provide a more comprehensive assessment of geomagnetic variability.
The revised manuscript is as follows:
As new seafloor forms and cools at mid-ocean ridges, it captures records of Earth's magnetic field. Polarity chrons, which represent intervals between magnetic reversals, serve as the basis for the Geomagnetic Polarity Time Scale (Cande and Kent, 1992). Gee and Kent (2007) updated the geomagnetic polarity reversal record based on marine magnetic anomalies, with significant revisions to the M-sequence of the Mesozoic. However, these adjustments fall outside the temporal scope of our study. To distinguish these chrons from shorter-duration anomalies, a 0.03 Myr threshold is arbitrarily applied. Anomalies shorter than this threshold, referred to as cryptochrons, correspond to brief intervals between reversals or incomplete reversals. Incomplete reversals are primarily associated with fluctuations in magnetic intensity and direction. Datasets on cryptochrons over the Cenozoic Era are also publicly available (Cande and Kent, 1992). All geomagnetic records are derived from magnetic anomalies preserved in mid-ocean ridge basalts. Using these records as a basis for reconstructing geomagnetic field variations minimizes potential climatic interference. Both reversals and incomplete reversals are collectively defined as geomagnetic events in this paper.
Previous studies have primarily used the frequency of geomagnetic reversals (Biggin et al., 2012; Pétrélis et al., 2011) to reflect changes in Earth's magnetic field. In contrast, we also consider incomplete reversals and employ a similar methodology to calculate the frequency of geomagnetic events (FGE). It was determined using a moving window approach with a 2 Myr window width and 1 Myr increments. Three groups of geomagnetic events (FGE) were categorized using a cutoff duration of 0.03 Myr (Fig. 1a–c). This duration serves as the threshold that distinguishes these chrons from cryptochrons. For example, the calculation of FGE>0.03 involves listing the chrons from the GPTS, each of which has a duration greater than 0.03 Myr. Each chron begins and ends with a polarity reversal or geomagnetic event. These geomagnetic events are then arranged chronologically, and the moving window method is applied to compute the FGE within each window. This approach is similarly applied to all other FGE calculations.
Comment 3: The Doake (1977) reference provides the basic physical concept of the article: polar ice caps provoke an increase of Earth’s rotation speed (conservation of the angular momentum) and thus an increase of the outer core fluid motion, providing more energy to the dynamo, resulting in stronger geomagnetic field .
However this old and single ref should also be questioned and completed by other references.
- The Doake (1977) hypothesis (in fact firstly introduced in 1975 by Olausson and Svenonius) was weakened by Kent (Nature 1982) who denounced the paleoclimatic biases on the natural remanent magnetization intensity records in sediment cores supposed (at this time) to represent the geomagnetic field intensity). Few experimental studies however supported the idea that the Earth’s rotation speed acts on the geomagnetic field intensity ( e.g. Miyagoshi and Hamano, in Phys. Rev. Lett., 2013).
However, several other studies suggested a causal link on the EMF intensity by orbital forcing ( precession, obliquity and eccentricity): see Fuller 2006, Thouveny et al. 2008, or Zhou et al. 2023 (and references therein).
Answer 3: Doake (1977), Olausson and Svenonius (1975), and Takehiro and Yozo (2013) all suggested that the establishment or melting of ice sheets could have a significant impact on Earth's magnetic field. However, the phase relationship between changes in Earth's rotation rate and geomagnetic field intensity remains unresolved.
Several other studies have proposed a causal link between orbital forcing (precession, obliquity, and eccentricity) and variations in geomagnetic field intensity (Fuller, 2006; Thouveny et al., 2008; Zhou et al., 2023, and references therein). These studies are primarily focused on the Quaternary period and examine geomagnetic variations on orbital timescales.
In contrast, our revised study focuses on million-year-scale variations during the Cenozoic. Nevertheless, we have summarized relevant Quaternary research in the introduction to provide context, as noted in Comment 1.
Comment 4: Some studies listed together in line 28 and 29 are presented as if they all agreed. This is wrong: they provided different (sometimes contradictory) observations and proposed different hypotheses: Fuller 2006 draw a relationship between the occurrence of reversals and the obliquity period; Thouveny et al. 2008 observed that the majority of dipole moment lows and excursions of the last 800 000 years occurred within interglacial episodes, in coincidence with obliquity minima.
The most recent high resolution studies paleomag and cosmonuclide isotopes along sediment sequences (e.g. Simon et al. 2016, 2018, 2020), as well as high precision Ar/Ar dating of excursional lava flow series allowed to demonstrate that the last reversal occurred during an interglacial and that most excursions of the last 800 ka occurred in narrow relation with interglacials. This totally cancels the hypothesis of Worm (1997) claiming that excursions occurred during glaciations.
Answer 4: On orbital timescales, the relationship between ice volume fluctuations and geomagnetic field variations during the Quaternary remains inconclusive. The revised manuscript shifts focus to longer, million-year timescales, examining the connection between large-scale ice sheet formation and retreat and Earth's magnetic field dynamics.
Comment 5: Macroscopically for the past 50 Ma, the correlation seems to be supported: frequency of reversals (chron durations shorter than 0.1 Ma) and Ice volume seem to present a significant correlation coefficient. However several important discrepancies appear at ca 20 Ma, 10 Ma and mostly for the last Ma.
Moreover, during the 50 - 60 Ma interval, the massive occurrence of reversals pointing at 55 Ma corresponds to an Ice free Greenhouse world. This observation totally kills the initial hypothesis and forces the author (line 69-75) to find another mechanism (catastrophic water distribution and mass transfers…) that are absolutely not explained, neither constrained.
Answer 5: In response to the reviewers’ comments, the revised manuscript focuses exclusively on the past 50 Ma, examining the macro-scale relationship between the frequency of geomagnetic events and ice volume. For all geomagnetic events—including polarity changes associated with both chrons and cryptochrons—the correlation coefficient between their occurrence frequency and ice volume reaches 0.76.
However, slight discrepancies are observed at ca. 20 Ma, 10 Ma, and the last 2 Ma. Given that geomagnetic reversals are influenced by multiple factors—including core-mantle boundary heat flux distribution, mantle convection, the formation of lower mantle superplumes, heterogeneity in lower mantle electrical conductivity, and plate tectonics (Amit and Olson, 2015; Olson and Amit, 2015; Franco et al., 2019; Li et al., 2016; Biggin et al., 2012)—ice sheet dynamics alone cannot fully account for every detail of geomagnetic field variations.
The Cenozoic marks a critical period in Earth's tectonic evolution, during which major continental masses migrated to their present-day configurations. Notably, the collision of the Indian subcontinent with Asia between 55 and 45 million years ago led to the formation of the Himalayan orogeny. Concurrently, around 50 million years ago, the Solar System underwent a chaotic orbital transition, characterized by orbital instability and chaotic diffusion. This astronomical perturbation has been implicated in major climatic events, including the PETM (Zeebe and Lourens, 2019). Given the fundamental differences in plate configurations, mantle dynamics, and astronomical parameters before and after 50 Ma, the variations in Earth's magnetic field prior to 50 Ma may need to be interpreted in the context of plate tectonics (Pétrélis et al., 2011), mantle convection processes (Biggin et al., 2012; Franco et al., 2019), and their influence on the geodynamo. This also highlights the complexity of the mechanisms underlying Earth's magnetic field variations.
However, our investigation into the relationship between ice volume and the geomagnetic field during the Cenozoic provides a new perspective for understanding the mechanisms driving geomagnetic variability. It is only during the Cenozoic icehouse climate that the large fluctuations in ice volume have had a more pronounced impact on Earth's magnetic field.
Comment 6: The most fundamental criticism is based on an obvious and revealing contradiction in
lines 9-10 and lines 82 – 85.
Indeed, the sentence (line 82): “such accelerated melting would decrease the occurrence of geomagnetic events”, is in complete contradiction with the sentence (line 83): “…relationship between ice sheet melting … and the increase in geomagnetic events”, the later itself contradicting the last sentence of the abstract (line 9 and 10).
This contradiction points the melting confusion of two different concepts: “occurrence of geomagnetic events” and “geomagnetic activity”.
Answer 6: As explained in Answer 3, the theoretical connection between ice sheet formation and retreat and geomagnetic field variations has been outlined. However, the phase relationship between these two factors, as reported in different studies, is not consistent. On the Cenozoic scale, our results show that periods of higher ice volume are associated with increased frequency of geomagnetic reversals. The erroneous statements related to this have been corrected in the revised manuscript.
Comment 7:Over the last decades, paleomagnetic studies of lava flows and sediments (e.g. Valet, Meynadier, Guyodo, 2005 in Nature), completed by cosmogenic nuclide studies (e.g. Simon et al. 2016, JGR; 2018, 2020 EPSL; Valet et al. 2024, QSR) demonstrate that reversals and excursions are associated (or even triggered by) dipole moment collapses. Therefore, at multi-million years scales, frequent reversals imply more frequent time intervals of weak dipole field.
This article is based on the hypothesis that the increase of the Earth’s speed rotation due to the mass accumulation in the polar regions (formation of ice caps) increases the energy of fluid motion in the outer core. But this mode should increase the geomagnetic dipole moment and thus provide a stronger stability of the dipole field polarity, i.e. a lower frequency of reversals. On the contrary the author claims that the glaciations (resp. deglaciations) of polar regions are responsible for increases (resp. decreases) of reversal frequency.
Conclusion : The author’s observation that higher frequency of reversals correlates with heavier polar ice masses is in contradiction with the principle of conservation of angular momentum that imply a faster Earth rotation and a stronger dipole field.
Note finally that several other studies suggested that on longer time scales (Ga) the control of the geomagnetic reversal frequency (and occurrence of superchrons), is influenced (or driven) by the heat transfert in the mantle) (e.g. Olson and Amit, 2015, 2019 Frontiers in Earth Science; Franco et al. 2019, Nature Sci. Reports).
Answer 7: Theoretical calculations suggest that the formation and retreat of ice sheets are sufficient to trigger geomagnetic reversals (Olausson and Svenonius, 1975; Doake, 1977; Takehiro and Yozo, 2013). However, the phase relationship between these two factors remains unresolved. Our study on Cenozoic ice volume and geomagnetic reversal frequency indicates that periods of high ice volume correspond to increased geomagnetic reversal events. We have identified this pattern, though the specific mechanisms still require further validation through additional model results.
Several other studies have suggested that, on longer timescales (Ga), the frequency of geomagnetic reversals—and the occurrence of superchrons—are influenced (or driven) by heat transfer in the mantle. As outlined in Answer 5, while ice volume has a significant impact on the geomagnetic field during the Cenozoic, prior to this period, factors such as plate distribution and Earth's orbit differed substantially, and in the absence of ice sheets, other factors become more pronounced in influencing the geomagnetic field.
Citation: https://doi.org/10.5194/esd-2024-43-AC1
-
AC1: 'Reply on RC1', Jiasheng Chen, 27 Mar 2025
-
RC2: 'Comment on esd-2024-43', M.J. Dekkers, 02 Feb 2025
Review Chen “Cenozoic Ice Volume as a Driver of Geomagnetic Events”
For Earth System Dynamics
The manuscript discusses a potential relation between ice volume and what is referred to as ‘geomagnetic events’, with a focus on short duration events lasting < 0.1 Myr (whether this would be short remains an unanswered question). The current lay-out suffers from several serious issues listed below (random order).
The idea seems to be inspired from Doake (1977) which is not a recent publication on this issue. Potential newer ideas (agreeing or disagreeing with Doake) should be reviewed. Referencing stops at the turn of the 21st century, no newer relevant work on the topic?
The data quality should be assessed; at present they are taken at face value from old sources. This applies at least to the geomagnetic events time series.
Are the ice volume and geomagnetic data placed on the same time scale? For frequency analysis this probably does not matter too much but here data seem to be plotted vs time.
The number of short geomagnetic events shorter 0.03 Myr (excursions or reversal excursions) may be grossly underestimated in the geomagnetic time series. Before 2 Ma or so, the existing data base (as off 2025) is very incomplete.
The 0.1 Myr ice age period applies only after the Mid-Pleistocene Transition (1.2-0.6 Ma). It may vary in older periods.
How the data were processed and with which software package(s) remains unclear.
There is no assessment of the contrast between icehouse and greenhouse state. Visually there is a similar amount of variability in Figure 1h. The claim that minimal geomagnetic variation would occur during the icehouse-greenhouse transition is not assessed. It reads as merely postulated.
What about the Jurassic (greenhouse) as an example? Reversal frequency is higher than 10x per Myr; this is full reversals, no information is available about excursions. If ice volume is a driver, what is the reason for geomagnetic variability during greenhouse times?
The implications/inferences should be assessed with the most meaningful data sets (why are the selected references the best?) and not against a few seemingly randomly picked references.
With the geomagnetic event data base being incomplete (short events are bound to be under-represented) inferences on links between future climate and geomagnetic field variability are on thin ice.
Future global warming is sometimes equated to heading to a uni-polar glacial state but whether completely ice-free conditions will be reached is an open question.
Citation: https://doi.org/10.5194/esd-2024-43-RC2 -
AC2: 'Reply on RC2', Jiasheng Chen, 27 Mar 2025
Comment 1: The manuscript discusses a potential relation between ice volume and what is referred to as ‘geomagnetic events’, with a focus on short duration events lasting < 0.1 Myr (whether this would be short remains an unanswered question). The current lay-out suffers from several serious issues listed below (random order).
The idea seems to be inspired from Doake (1977) which is not a recent publication on this issue. Potential newer ideas (agreeing or disagreeing with Doake) should be reviewed. Referencing stops at the turn of the 21st century, no newer relevant work on the topic?
Answer 1: In response to the reviewer's suggestion, we have updated the references. In addition to the theoretical calculations by Olausson and Svenonius (1975) and Doake (1977), the study by Takehiro and Yozo (2013) also conducted magnetohydrodynamic geodynamo simulations. All of these studies demonstrate that changes in ice volume and Earth's rotation have a significant impact on the geomagnetic field.
Comment 2: The data quality should be assessed; at present they are taken at face value from old sources. This applies at least to the geomagnetic events time series.
Are the ice volume and geomagnetic data placed on the same time scale? For frequency analysis this probably does not matter too much but here data seem to be plotted vs time.
The number of short geomagnetic events shorter 0.03 Myr (excursions or reversal excursions) may be grossly underestimated in the geomagnetic time series. Before 2 Ma or so, the existing data base (as off 2025) is very incomplete.
The 0.1 Myr ice age period applies only after the Mid-Pleistocene Transition (1.2-0.6 Ma). It may vary in older periods.
How the data were processed and with which software package(s) remains unclear.
Answer 2: All geomagnetic records are derived from magnetic anomalies preserved in mid-ocean ridge basalts. Using these records as a basis for reconstructing geomagnetic field variations minimizes potential climatic interference.
The Geomagnetic Polarity Time Scale (GPTS) was initially established by Cande and Kent (1992) based on oceanic magnetic anomalies. The record was later updated by Gee and Kent (2007), with significant revisions focusing on the M-series of the Mesozoic. While our study primarily concentrates on the Cenozoic, we have adopted the Gee and Kent (2007) model. Cryptochrons, defined as magnetic anomalies with a duration of less than 0.03 Myr, were excluded from the Geomagnetic Polarity Time Scale when it was first established by Cande and Kent (1992). However, these cryptochrons are listed separately in the literature. Currently, no updated studies on these cryptochrons exist. Therefore, the data on geomagnetic field changes is exclusively drawn from Gee and Kent (2007) and Cande and Kent (1992).
Ice volume data has a resolution of 1 Myr, and the frequency of geomagnetic events is also calculated with a 1 Myr time step. The comparison between the two datasets is conducted on the same temporal scale, which satisfies the resolution requirements for our analysis.
It is possible that the number of cryptochrons before 2 Ma is underestimated, but no new data can confirm this. The study uses only mid-ocean ridge basalt records for cryptochrons, and no updated data sources are available.
The eccentricity cycle of 0.1 Myr becomes significant only after the Mid-Pleistocene Transition in the ice volume records. Reviewer 1 also raised concerns about the connection between ice volume and geomagnetic field changes on the orbital scale. Consequently, we have removed related studies on orbital-scale changes and focused on the relationship between ice volume and geomagnetic field variations on the million-year scale.
The calculation of the frequency of geomagnetic events involves counting geomagnetic events within a sliding window. This process can be handled in Excel, with the main procedure outlined below:Previous studies have primarily used the frequency of geomagnetic reversals (Biggin et al., 2012; Pétrélis et al., 2011) to reflect changes in Earth's magnetic field. In contrast, we also consider incomplete reversals and employ a similar methodology to calculate the frequency of geomagnetic events (FGE). It was determined using a moving window approach with a 2 Myr window width and 1 Myr increments. Three groups of FGE were categorized using a cutoff duration of 0.03 Myr (Fig. 1a–c). This duration serves as the threshold that distinguishes these chrons from cryptochrons. For example, the calculation of FGE>0.03 involves listing the chrons from the GPTS, each of which has a duration greater than 0.03 Myr. Each chron begins and ends with a polarity reversal or geomagnetic event. These geomagnetic events are then arranged chronologically, and the moving window method is applied to compute the FGE within each window. This approach is similarly applied to all other FGE calculations.
Comment 3: There is no assessment of the contrast between icehouse and greenhouse state. Visually there is a similar amount of variability in Figure 1h. The claim that minimal geomagnetic variation would occur during the icehouse-greenhouse transition is not assessed. It reads as merely postulated.
What about the Jurassic (greenhouse) as an example? Reversal frequency is higher than 10x per Myr; this is full reversals, no information is available about excursions. If ice volume is a driver, what is the reason for geomagnetic variability during greenhouse times?
Answer 3: Reviewer 1 raised concerns regarding the relationship between geomagnetic field variations and ice volume prior to 49 Ma. Around 50 Ma, the collision between the Indian plate and the Eurasian continent, coupled with a chaotic orbital transition in the Solar System (Zeebe and Lourens, 2019), resulted in significant geological and orbital changes that distinguish the pre-50 Ma period from subsequent epochs.
Geomagnetic field variations are influenced by various factors, including plate tectonics (Pétrélis et al., 2011), mantle convection processes (Biggin et al., 2012; Franco et al., 2019), and others. Our findings primarily highlight the connection between Cenozoic ice volume and geomagnetic field changes, offering a novel perspective for understanding geomagnetic variations during this period. In the absence of ice volume changes, other factors, such as those mentioned, likely exert a more prominent influence on the Earth's magnetic field.
Comment 4:The implications/inferences should be assessed with the most meaningful data sets (why are the selected references the best?) and not against a few seemingly randomly picked references.
With the geomagnetic event data base being incomplete (short events are bound to be under-represented) inferences on links between future climate and geomagnetic field variability are on thin ice.
Future global warming is sometimes equated to heading to a uni-polar glacial state but whether completely ice-free conditions will be reached is an open question.
Answer 4: As explained in Answer 2, the data have been clarified. Other references have also been updated, including additional studies on the impact of ice volume on the geomagnetic field and other factors influencing geomagnetic variations, as cited in Answer 3. The revised manuscript now exclusively focuses on the relationship between ice volume and geomagnetic field variations on the Cenozoic million-year timescale. Future climate warming involves shorter timescales, and the geomagnetic response to such changes cannot be inferred from the relationship observed over million-year scales. Accordingly, relevant sections have been removed in the updated manuscript.
Citation: https://doi.org/10.5194/esd-2024-43-AC2
-
AC2: 'Reply on RC2', Jiasheng Chen, 27 Mar 2025
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 113 | 21 | 8 | 142 | 6 | 5 |
- HTML: 113
- PDF: 21
- XML: 8
- Total: 142
- BibTeX: 6
- EndNote: 5
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1