the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Observations of Electric Fields during two Partial Solar Eclipses at the Geomagnetic Equator
Abstract. This study presents the first coordinated observations of atmospheric electric field (AEF) and ionospheric plasma drifts during the partial solar eclipses of 2 July 2019 and 14 October 2023, observed near the magnetic equator in Lima, Peru. AEF was measured using a field mill, while ionospheric drifts were obtained from radar observations at the Jicamarca Radio Observatory and local magnetometers. The two events displayed contrasting electrodynamic responses: in 2019, AEF variations were ambiguous due to meteorological fluctuations, while in 2023, clearer weather conditions revealed distinct decreases in both surface AEF and ionospheric vertical drift near maximum obscuration. These results demonstrate the variable nature of eclipse-time electrodynamics and emphasize the importance of multi-instrument approaches for understanding atmosphere-ionosphere coupling in low-latitude regions.
- Preprint
(1754 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
CC1: 'Comment on egusphere-2025-3155', Tao Chen, 05 Aug 2025
This manuscript presents observations of atmospheric electric fields (AEF) and ionospheric plasma drifts during two partial solar eclipses over the geomagnetic equator. The topic is relevant and the dual-event comparison across 2019 and 2023 adds some contextual interest. The authors provide detailed descriptions of the instrumentation, site conditions, and multi-instrumental data processing approaches.
It is possibly a result that the eclipses inflence the AEF not so much obvious, but influence Ionosphere largely, Due to less solar light, the ionized rate may much decrease. So, the eclipse may cause some sudden change in ionosphere, in turn the eletric field in ionosphere is alos change.
This manuscript is worlth to be published that the readers will find its contribution concerning the response of ionospheric electric field and some atmospheric electrostatic field to eclipse.
This manuscript have present fluent English writing and logical structure. I suggest directly publish it without revision.
Citation: https://doi.org/10.5194/egusphere-2025-3155-CC1 -
AC2: 'Reply on CC1', Manuel Bravo, 05 Sep 2025
Dear reviewer,
Thank you very much for your response. We hope we have contributed to this topic and met the journal's standards.
Citation: https://doi.org/10.5194/egusphere-2025-3155-AC2
-
AC2: 'Reply on CC1', Manuel Bravo, 05 Sep 2025
-
RC1: 'Comment on egusphere-2025-3155', Krishnamurthy Jeeva, 12 Aug 2025
Observations of Electric fields during two Partial Solar Eclipse at the Geomagnetic Equator
Bravo et. al.
I appreciate the authors for selecting the topic for the investigation. This type of investigations will help to establish the hidden fact on the topic electodynamic coupling between the upper and lower atmosphere and will provide some clues on the link between the spaceweather and terrestrial weather.
Major comments
The authors did not mention about the need of the coordinated observations of Atmospheric Electric field and ionospheric plasma drifts. For example, they could have mentioned about the electrical and electrodyamic coupling between the upper and the lower atmosphere in terms of the spaceweather and terrestrial weather investigations at an appropriate location in the manuscript.
The two observations considered in the present study display contradictory characteristics. The cited references of the past studies, in the similar topic, also do not provide a conclusive result. A few show increase of AEF during the solar eclipse and another few suggest decrease in the same.
The AEF is mostly influenced by the meteorological parameters. To examine the role of the contribution of the signatures of the upper atmosphere over the AEF it is very important to prove that there is no role of the meteorological weather parameters on the AEF. The weather information provided in this paper does not suffice the need.
In Fig. 3 the authors represent the diurnal variation of AEF on a reference day and on the eclipse day. The fluctuation of the data on the day of event is stronger during the hours other than the eclipse duration. Therefore the increase or decrease cannot be attributed to the event.
In Fig. 4 also there are noticeable fluctuations in the diurnal variation of the AEF. However, there is a noticeable change in the AEF after the onset of the eclipse. It is believed to be due to diminishing of the dynamo driven electric fields. To ratify this one needs complimentary experiments such as atmospheric electrical conductivity and air ions measurement. These two parameters are highly sensitive to the meteorological weather parameters. Therefore it is difficult to infer whether this change in AEF is due to the eclipse event.
In the conclusion the authors claim that the atmospheric and ionospheric electric fields during eclipse provides valuable insight into the electrodynamic responses to transient solar forcing, but the results and discussions are not enough to come to this conclusion.
The most discouraging conclusion of this paper is that the authors agree that the large fluctuation of the AEF complicates the study and I agree with the authors.
The signatures observed in the vertical drift is quite convincing as it is taking place in the dynamo region and it is obvious that the solar energy would certainly impact the electrical process.
Minor comments
The figures should be labelled properly. Where ever multiple panels are presented they should be labelled as (a), (b), (c) etc followed by appropriate description.
The time in the X axis in local time is more appropriate than the UT.
Decision
Not recommended for publication.
Citation: https://doi.org/10.5194/egusphere-2025-3155-RC1 -
AC3: 'Reply on RC1', Manuel Bravo, 05 Sep 2025
Comment: The authors did not mention about the need of the coordinated observations of Atmospheric Electric field and ionospheric plasma drifts. For example, they could have mentioned about the electrical and electrodyamic coupling between the upper and the lower atmosphere in terms of the spaceweather and terrestrial weather investigations at an appropriate location in the manuscript.
R: We appreciate this insightful suggestion, which highlights an opportunity to strengthen the manuscript's framing. Indeed, the original introduction briefly noted the novelty of our coordinated AEF and ionospheric drift observations, with references to vertical coupling (Tacza et al., 2016; St-Maurice et al., 2011). To address your point, we will add a dedicated, properly cited paragraph in the revised introduction explicitly discussing the need for such observations in investigating electrodynamic coupling between the upper and lower atmosphere, including linkages to space weather (e.g., solar forcing effects) and terrestrial weather (e.g., meteorological influences on conductivity). This will incorporate additional references to better contextualize the study's contributions.
Comment: The two observations considered in the present study display contradictory characteristics. The cited references of the past studies, in the similar topic, also do not provide a conclusive result. A few show increase of AEF during the solar eclipse and another few suggest decrease in the same.
R: We fully acknowledge the contradictory nature of AEF responses to solar eclipses, as reflected in our results and the literature review. Our pragmatic approach was to highlight this complexity without selectively favoring citations or adjusting results to prior conclusions; instead, we emphasized the intricate connections between solar eclipse phenomena and AEF through two events with opposing ionospheric responses under comparable conditions. This opposition appropriately illustrates the vertical coupling between atmospheric layers. To clarify this, we will revise the discussion section to more explicitly frame these contradictions as evidence of electrodynamic linkages, while calling for further studies to resolve inconsistencies.
Comment: The AEF is mostly influenced by the meteorological parameters. To examine the role of the contribution of the signatures of the upper atmosphere over the AEF it is very important to prove that there is no role of the meteorological weather parameters on the AEF. The weather information provided in this paper does not suffice the need.
R: We agree that meteorological influences on AEF must be rigorously controlled to isolate upper atmospheric signatures, as noted in our methods and discussion (e.g., citing Velazquez et al., 2022). However, the absence of local meteorological data for 2019 limited our analysis compared to 2023; the nearest available station was approximately 200 km away, rendering it unrepresentative due to differing geography. We mitigated this by using GOES-16 satellite imagery to minimize uncertainty regarding adverse weather conditions that could affect local AEF. For 2023, wind data analysis confirmed no significant meteorological perturbations that would obscure potential eclipse responses. To enhance rigor, we will expand the methods section in the revision to detail these limitations and justify our fair-weather criteria, while recommending future integration of proximal meteorological stations for more comprehensive validation.
Comment: In Fig. 3 the authors represent the diurnal variation of AEF on a reference day and on the eclipse day. The fluctuation of the data on the day of event is stronger during the hours other than the eclipse duration. Therefore the increase or decrease cannot be attributed to the event.
In Fig. 4 also there are noticeable fluctuations in the diurnal variation of the AEF. However, there is a noticeable change in the AEF after the onset of the eclipse. It is believed to be due to diminishing of the dynamo driven electric fields. To ratify this one needs complimentary experiments such as atmospheric electrical conductivity and air ions measurement. These two parameters are highly sensitive to the meteorological weather parameters. Therefore it is difficult to infer whether this change in AEF is due to the eclipse event.
R: We recognize that the observed data exhibit suboptimal amplitudes due to installation errors, as mentioned in the methodology. The applied normalization, similar to Tacza et al. (2018), facilitates a qualitative rather than quantitative analysis to mitigate this impact. For the 2023 event in Fig. 4, the AEF response aligns temporally with the eclipse duration, whereas post-event rapid fluctuations are shorter-lived, suggesting they should not be conflated. In contrast, the 2019 data in Fig. 3 show even greater variability, which we explicitly noted. Regarding complementary experiments (e.g., atmospheric electrical conductivity and air ion measurements), these were not feasible due to equipment limitations; we will emphasize this as a challenge for future comparable eclipse events in the revised discussion, while referencing models like Denisenko et al. (2018) to support potential dynamo-driven interpretations.
Comment: In the conclusion the authors claim that the atmospheric and ionospheric electric fields during eclipse provides valuable insight into the electrodynamic responses to transient solar forcing, but the results and discussions are not enough to come to this conclusion.
R: We agree that conclusions must strictly align with the presented evidence, avoiding excessive claims. The main purpose of this paper was to present coordinated observations and acknowledge the complexity of analyzing AEF effects. To illustrate this challenge, we included two eclipse events that showed opposite ionospheric responses. In light of the reviewer’s comment, we will revise the conclusions section, moderating the language: it will now state that ionospheric data provide clear evidence of electrodynamic responses, while AEF inferences are described as preliminary and pending multi-instrumental validation. This adjustment will ensure consistency between results and conclusions, improving the overall clarity of the manuscript.
Comment: The figures should be labelled properly. Where ever multiple panels are presented they should be labelled as (a), (b), (c) etc followed by appropriate description.
R: We concur with this recommendation for improved clarity. In the revised version, we will label all multi-panel figures, and provide detailed descriptions in the captions.
Comment: The time in the X axis in local time is more appropriate than the UT.
R: We agree that local time enhances interpretability for equatorial studies. In the next version, we will update the x-axes in relevant figures to prioritize local time, with universal time noted for reference where necessary.
Citation: https://doi.org/10.5194/egusphere-2025-3155-AC3
-
AC3: 'Reply on RC1', Manuel Bravo, 05 Sep 2025
-
EC1: 'Comment on egusphere-2025-3155', Geeta Vichare, 28 Aug 2025
CC1 was intended as a review.
Citation: https://doi.org/10.5194/egusphere-2025-3155-EC1 -
AC1: 'Reply on EC1', Manuel Bravo, 05 Sep 2025
Ok, understood editor.
Citation: https://doi.org/10.5194/egusphere-2025-3155-AC1
-
AC1: 'Reply on EC1', Manuel Bravo, 05 Sep 2025
Viewed
HTML | XML | Total | BibTeX | EndNote | |
---|---|---|---|---|---|
1,184 | 51 | 14 | 1,249 | 66 | 64 |
- HTML: 1,184
- PDF: 51
- XML: 14
- Total: 1,249
- BibTeX: 66
- EndNote: 64
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1