the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 11 Mar 2026
Electric fields in and around an auroral arc and the inferred current system from the BROR sounding rocket experiment
Abstract. Magnetosphere-ionosphere interactions play a crucial role in the dynamics of near-Earth space, and the electric field in the vicinity of the auroral arc is one of the major links in these interactions. The electric field in the auroral ionosphere has been measured using various techniques: coherent and incoherent radars from the ground, and in situ measurements using rockets and satellites. The effective approach to studying the auroral electric field is to determine it from observations of the motion of artificial ion clouds released into the ionosphere by a sounding rocket. On 23 March 2023, the Barium Radio and Optical Rocket (BROR) experiment was conducted at the Esrange rocket range, near Kiruna, Sweden. In the experiment, 8 canisters containing a barium-strontium-thermite mixture were released at altitudes between 130 and 240 km. A novelty of the experiment is multi-station narrow-band optical observations of emissions with ALIS_4D. This allows us to reconstruct the 3D distribution of optical phenomena in the ionosphere using a tomography-like technique with spatial and temporal resolutions of ∼500 m and 0.1 s, respectively. The active auroral arc developed inside the area occupied by the barium clouds and intersected with the one of clouds at an altitude of ≈230 km for quite a long time. During this time, the ion cloud experienced various deformations, which we observed and used to determine the electric field as a function of position relative to the auroral arc.
- Preprint
(36201 KB) - Metadata XML
-
Supplement
(11585 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2026-1133', Anonymous Referee #1, 09 Apr 2026
-
AC1: 'Reply on RC1', Tomoe Taki, 01 Jul 2026
The attached PDF file contains our response to RC1, including figures. The text of our response is also provided below.
Thank you very much for your valuable comments and suggestions for improvement.
We will answer your questions below, and we will revise the manuscript reflecting your comments.Abstract
L1 and Abstract: The current abstract does not include any direct results or new findings of this paper itself. It describes the novelty of the experiment, but not the novelty of the findings.This study presents a new method to investigate ionospheric electric field structures. Using tomography, we successfully derived the three components of the ion velocity at each location. This allows us to show from observations that key assumptions in previous studies—such as negligible collisions at the observation altitude and purely horizontal electric fields—are reasonable. Compared to previous approaches, our method estimates the electric field with fewer assumptions.
L4 “The effective approach”: Could the authors clarify in what sense this approach is considered effective?
This approach is effective because it allows us to directly visualize and observe ion motion controlled by the electric field. In-situ rocket and radar observations allow direct measurements of ions, but they are limited to single-point observations. Optical observations enable us to derive three-dimensional velocity fields and obtain their corresponding spatial maps.
L7 “A novelty of the experiment is…”: At L94, the authors state that the difference between ALIS and ALIS_4D is the use of an EMCCD instead of a CCD. Combining these two sentences, does this mean that the only novelty of the experiment, compared to previous studies, is the use of an EMCCD?
Experiments of this type have not been conducted with ALIS. The previous wording may have been misleading, but improvements to ALIS and ALIS_4D are not the main novelty of this study. The novelty of this study is that we reconstruct Ba⁺ ion movement in three dimensions using tomography. This provides a detailed study of ion dynamics with a resolution of 500 m and 0.1 s.
L11 “quite a long time”: How long is this period?
We can observe the clouds by cameras for almost 30 minutes.
Introduction
L15 “current systems”: Could they also explain more why the current system is important for understanding this coupled ionosphere – magnetosphere system?
L15” Various observational approaches”: The introduction discusses rocket, satellite, and ground based measurement approaches. These techniques have different horizontal and vertical observation ranges, but the current text does not describe these differences and their implications.
L20 “were small, less than 10 mV/m”: It may be useful to mention how small these values are compared to the background.
L21 “altitude dependent … were reported”: This sentence appears inconsistent with the statement at L22 that “no significant altitude variation was observed.”
L24 “On the other hand, charge oscillations were observed”: The meaning of this sentence and relation to the previous sentence are unclear. Please rephrase it.
L26 “spacecraft”: This literature appears in the middle of a paragraph introducing rocket studies. How is it related to that context? Spacecraft studies are introduced starting at L37.
L37” attempted”: The authors could add more information to this sentence explaining why satellite measurements were attempted.
L42 “strong”: How strong are these?
L44 “relatively large”: Relative to what, and how large? Please specify.
L45 “… However…”: The reviewer does not understand the implication of this “however” or the logical connection between the preceding and following sentences. Please clarify the intended contrast.
L50 “Strong”: Please explain how strong, and relative to what.
L51 “3 s,”: Is this about optical or EISCAT observation ?
L53 “the two”: Please clarify what “the two” refers to.
L53 “response”: Does this imply any interaction between the radar and the electric field?
L57 “downward field-aligned currents”: Could the authors describe how these currents are related to the electric field?
L58 “background electric field was…canceled by a polarization electric field”: Please explain the mechanism responsible for this.
L 59 “this study”: Please consider rephrasing “this” to avoid possible ambiguity.
L62 “but have relatively limited resolution. However,…”: The reviewer does not understand how these sentences about Clayton et. al., are connected by “but” and “however.” Please rephrase.
L66 “As a result, it was difficult to separate temporal variations from spatial variations based on the observations alone.”: The reviewer does not understand this as a conclusion from the preceding sentence.
L67 “temporal and spatial scales smaller than… not be detected”: This limitation applies to any observation technique. What are intended to mean here?
L68 “barium clouds”: Active ionospheric experiments using chemical releases, including barium, have been conducted for many years. The authors cite Fahleson et al. in the beginning, but additional previous studies and results relevant to the present work could be introduced if there are any.We thank you for your helpful comments. We will address these points and incorporate the revisions into the revised manuscript. We will substantially revise the Introduction in the revised manuscript in response to your comments.
Experiment
L75 “23 March 2023”: The storm events in March 2023 including March 23 have been studied by several authors already. Can authors find any relationship between their study to previous studies? e.g,
https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2023JA032145
https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2023SW003728
As seen in these papers, the EISCAT radars were operating on March 23, and one of their objectives should have been to support the BROR experiment. Were there any EISCAT results that could be relevant to the current work? Also, with regard to previous EISCAT–rocket studies, such as the cited works by Lanchester et al. or Aikio et al.? In particular, ion drift measurements or electric field/FAC estimates derived from the radar data would be highly relevant.EISCAT was operated in a special mode during the BROR experiment. However, because the signal-to-noise ratio was relatively low, it is difficult to make meaningful comparisons of the small-scale dynamics associated with the aurora. It may be possible to present averaged values in the revised manuscript.
L84 “moved under influence of the electric field”: Similar to the comment above, strong thermospheric winds were reported on 23-24 March 2024 storm (see Oyama et al 2024 above). Did the authors estimate the possible influence of thermospheric winds?
Our study accounts for the influence of the neutral wind on the motion of barium ions. The thermospheric wind velocities were determined from the motion of neutral strontium clouds observed by optical instruments. We will compare the neutral wind velocities obtained in our experiment with the papers you mentioned in the revision.
L96 Figure 2: What does the yellow point on the map represent? No description is provided.
The yellow dot on the map indicates the ALIS_4D Nikkaluokta station, but it was not used during the BROR experiment, so we plan to remove it.
2.2 Instruments and Data
L104 “strontium”: Why was only the camera at Esrange used to measure strontium? In addition, were any results from these strontium measurements considered in the analysis from Esrange in Figure 3?Only the ALIS_4D camera at Esrange was operated with the 460.7 nm filter to observe neutral strontium, whereas the other cameras were configured with the 493.4 nm filter to observe ionized barium clouds. This configuration was chosen because the main goal of the BROR experiment was to study the small-scale electrodynamics in the auroral ionosphere through the motion of ion clouds. The initial version of the manuscript did not include a description of the analysis of the Esrange strontium data. In the revised manuscript, we will include approach to estimate the thermospheric neutral wind and to evaluate the possible influence of ion-neutral collisions on the barium ion motion. We will also revise the relevant figures.
L112 “too weak to be detected”: Is this conclusion based on experimentally established results from previous studies, or is it derived from the observations in the present study? In Figure 4, both cameras show isolated round structures near the arc (for the Kiruna camera, one appears near the center of the image). What are these features?The data of the Watec cameras at Kiruna and Tjautjas are the images taken with a band-pass filter for the auroral 557.7 nm emission. The isolated round structures in these images are likely associated with strontium lines at 550.4, 552.1, 553.4 and 554.0 nm. For lower-altitude clouds, barium oxidation competes with ionisation. BaO is characterised by a rich electronic-vibrational spectrum in the visible range and may contribute to cloud images recorded by Watec cameras. Ba+ emission lies outside the wavelength range of the camera filter.
3 Data processing and analysis
L118 “Figure 5”: Please indicate where the fourth cloud is located in all of the Ba related sub figures in Figure 5, and specify how many clouds are present in total.We will revise the figure to make the cloud locations clearer, as illustrated in the Figure R1.
L123 Figure 6: This reconstruction figure seems that some of the early ion clouds reached lower altitudes, close to the auroral altitude around 120 km. If this interpretation is correct, these clouds may be more suitable for analyzing the electric field in the vicinity of the aurora.
The fourth cloud was the only cloud that deformed during the auroral passage, and therefore it was selected for detailed analysis. We were fortunate during the BROR experiment in that an auroral arc traversed a barium ion cloud released at altitudes above 200 km. At these altitudes, the gyrofrequency of Ba+ ions exceeds the collision frequency. The ion motion is controlled predominantly by the ExB drift. It is possible to infer the fine-scale structure of the electric field and current systems associated with the auroral arc. We will clarify this point in the revised manuscript.
L125 “a vertical range of ±10 km around the altitude at which the cloud intensity reaches its maximum.”: Please explain in more detail how the altitude of maximum Ba intensity is determined. Is this done by comparing mapped images using altitude as a parameter from all four stations?
The maximum intensity of barium emission is derived from tomography. Therefore, the vertical position is determined with a spatial resolution of 500 m.
For example, the altitude profile at 18:26:30 is shown in the Figure R2. By horizontally (x–y plane) integrating the reconstructed three-dimensional intensity distribution, we estimate that the fourth cloud is located at an altitude of approximately 222 ± 10 km. Within this range, we identify the point (X, Y, Z) with the maximum intensity. We will clarify this in the revised manuscript.L 133 “we shifted the cloud data to the height at an altitude of 120 km”:
The objective of this projection needs more explanation.
This reconstruction procedure appears to assume that electric field variations can be ignored over the 100 km altitude range between approximately 120 and 230 km. How do the authors justify this assumption in the case of an auroral arc?
Moreover, this projection gives the impression that the analysis does not fully utilize the capabilities of the experimental design. If the reviewer interpreted it correctly, as the tomographic results in Figure 6 show, the total set of ion clouds spans a much larger altitude and horizontal range than the fourth cloud alone, with some clouds extending down toward auroral altitudes. In principle, this experiment could provide not only horizontal motion but also information on vertical deformation/motion.
Also, similar to a comment before, the EISCAT Tromsø radar was operational on that day and should provide plasma parameters, and likely most relevant for this study, altitude variation of plasma parameters including ion drift.It is generally accepted that magnetic field lines in the ionosphere are approximately equipotential down to about 90 km altitude, where electrons become non-magnetised. Although there is evidence for field-aligned electric fields at ionospheric altitudes, these are often considered small compared to the variations in the horizontal electric field components.
Above about 200 km altitude, ions, including heavy barium ions, are mainly driven by the E×B drift and can be used to directly infer electric field variations. At lower altitudes, ion–neutral collisions and neutral winds increasingly affect ion motion, making it difficult to reliably infer the electric field from ion dynamics.
The aim of this study is to investigate the fine structure of electric fields and currents within individual auroral arcs using a specific dataset from the BROR experiment. Additional barium cloud release data are used to examine several phenomena related to the electrodynamics of the auroral ionosphere, such as field-aligned electric fields and the physicochemical behavior of barium in the thermosphere.
Regarding EISCAT Tromsø, the signal-to-noise ratio is not sufficient for a reliable quantitative comparison of ion-drift with the BROR results.L138 “…green points. A blue line…”: The blue lines in Figure 7 do not appear to represent the observed auroral arc accurately, as they differ noticeably from the grayscale structure. Fitting a straight line to an auroral arc does not seem physically justified or effective, especially at 18:30 UT. At 18:25, no auroral arc is visible in the image, but only the blue lines are shown.
The image at 18:25 is cropped to focus on the fourth cloud, but an auroral arc can be seen extending in a curved shape. The blue straight line is drawn based on that configuration.As you pointed out, fitting a possibly curved auroral arc with a straight line may introduce differences in the spatial relationship with the aurora. For comparison, we also show the auroral structure and resulting electric field when fitting the arc using a fifth-order polynomial in the Figure R3. However, the overall results do not change significantly.
L139 “The fourth cloud projected”: The reviewer assumes that the red tracings in Figure 7 represent the projected cloud structures from Figure 6. However, at 18:25 these appear only as very small red points in Figure 7. Based on Figures 6 and 7 alone, it is difficult for the reader to verify whether the red circles in Figure 7 correctly correspond to the fourth cloud shown in Figure 6.
To improve clarity, the authors should consider adding the auroral arc outline also in Figure 6 for comparison, and/or provide the geolocation (latitude, longitude, and altitude) of the fourth cloud. This would allow readers to follow the temporal and spatial development of the fourth cloud more easily.
In addition, as noted earlier, the bright round structures in the auroral images create confusion if they are not related to the barium clouds. Are these bright features indeed unrelated to the BROR cloud releases? A clear explanation is needed to avoid misinterpretation.We would like to clarify that we analyze only the fourth cloud among the multiple clouds. This cloud was released at around 18:25 UT and is the cloud that is tracked in Figure 7. To make the correspondence between Figures 6 and 7 easier to follow, we will revise the figure caption and text to state explicitly that the red contours in Figure 7 are obtained from the tomographic reconstruction of the forth cloud and then projected to the 120-km altitude plane along the magnetic field line. We will also provide additional images for all times shown in Figure 7 as shown in Figure R4.
The bright circular structures seen in the Watec images of the aurora are likely associated with emissions from strontium. They are therefore related to the BROR releases, but they do not represent the Ba⁺ emission traced in this study. The ion cloud emission is not clearly captured by the Watec cameras equipped with the green-line filters.L147 “We selected four points to characterize the cloud deformation”. The reviewer assumes that this paragraph is intended to describe Figure 8, although Figure 8 is never mentioned in the main text.
We will revise the text as follows: We selected four points to characterize the cloud deformation, as illustrated in Figure 8.
L162: “7-s moving average”: High resolution imaging was a key novelty of this experiment. Please explain whether this novelty is preserved when the data are averaged.
The 7-s moving average was applied only to the derived velocity time series, not to the original high-cadence image data or tomographic reconstruction. The 0.1-s observations were used to track the cloud position and to confirm that no significant sub-second or few-second perturbations were present. Thus, the high temporal resolution is still essential for validating the temporal behavior of the cloud motion, even though the final velocity curves are smoothed for display and interpretation. Because the cloud displacement between adjacent time steps is small relative to the spatial grid size, the unsmoothed velocity appears discrete; the moving average was used to show the slowly varying drift more clearly. We will clarify that this smoothing reduces sensitivity to variations shorter than several seconds, but the electric-field structures discussed here occur on longer time scales and are not significantly affected.
L164: “60s, 100 s”: Please specify the times in UT so that readers can compare them directly with the optical figures.
We will update the manuscript to include UT in the manuscript together with the elapsed time, e.g., 60 s (18:26:46) and 100 s (27:26).
L172 “Watec images also include BROR clouds”: This sentence seems to contradict the statement in line 112, creating confusion for the readers. Please explain this point more clearly.
As stated in response to line 139, neutral components are included in the observations. The Watec images were obtained using a 558 nm filter and therefore include emissions from strontium clouds. However, these clouds are not the Ba ion clouds analyzed in this study. We will clarify this distinction in the revised manuscript.
L178 “distance of each point from”: Related to earlier comments, is the projection from 230 km down to 120 km include any geometric effects? For example, can the spatial distance between two points become larger or smaller after the projection? Any such distortion could directly affect the distance/velocity calculation.
At present, the transformation is treated as a simple parallel shift. The barium clouds analyzed in this study are distributed within an area of approximately 200 km × 200 km above Esrange, over which the inclination of the magnetic field lines changes by only about 0.5°. This could introduce a distortion of roughly 1 km. Since this corresponds to only about two grids in the tomographic reconstruction, we consider that the distortion is not large enough to significantly affect the results. We will include this estimate and clarify the possible geometric effect of the projection in the revised manuscript.
L188 “vertical velocity along the z-axis”. Please clarify whether this computation is performed before or after the tomographic reconstruction shown in Figure 6. If it is derived after tomography, then the vertical resolution is limited to approximately 20 km, the it could impact the interpretation of Figure 11.
A more essential question concerns the vertical motion of the barium clouds: how much of the observed descent can be explained by natural gravitational settling, and how much is attributable to electric field–driven motion?
Aslo, in Sergienko et al., 2024, a vertical motion of ions was reported as an unexpected finding. Is there any connection to the current study?The vertical velocity shown in Figure 11 is derived after the tomographic reconstruction. In the reconstruction, the cloud intensity is obtained on a 500-m grid, and the altitude of the maximum intensity is traced as a function of time for the selected representative points. We will clarify this procedure in the revised manuscript and also describe the uncertainty of the altitude determination.
In the original manuscript, the effect of gravity was not taken into account. However, including this effect resulted in better agreement with the observations. In the revised manuscript, we will present these improved results.
The unexpected vertical velocities mentioned in Sergienko et al. (2024) was associated with clouds formed by barium release at altitudes lower than 190 km and was interpreted as possibly indicating a field-aligned electric field. The fourth cloud analyzed in the present paper is located at a higher altitude, around 230 km, and its vertical motion can be explained without invoking such an additional field-aligned electric field. We will clarify the relation to Sergienko et al. (2024) and distinguish the present result from the lower-altitude case discussed there.4 Results and discussion
L193: This section 4 is titled “Results,” but in practice the material presented in Section 3 also contains results. The distinction between the two sections is therefore unclear.In the revised manuscript, we will change the title of this section to a more appropriate one, such as "Discussion" or "Estimation of electric fields".
L199 “for the altitude of approximately 230 km treated in this study,”: One of the most significant methodological questions in this study concerns how the lower F region electric field is being characterized when the mapped auroral structures are located at E region heights. The manuscript does not clearly explain the physical justification for interpreting an electric field at 230 km based on features mapped to ~120 km. The E region is highly collisional compared to lower F-region, so the electrodynamic behavior, and therefore the inferred electric field may be sustainably different.
The magnetic field lines at the ionospheric altitudes are almost equipotential. So the mesoscale (>tens of meters) electric fields at altitudes of 230 km and 120 km are the almost same. The auroral images taken by the Watec cameras are used only for obtaining the geographical position of the auroral arc and its connection to the field-aligned current system.
What is really different at these altitudes is the electrical currents, but they are not the subject of this paper. The vertical motion of the three lowest barium ion clouds created during the BROR experiment indicates the potential presence of the field-aligned electric field at a distance from the auroral arc. This electric field is very interesting as an auroral electrodynamic phenomenon, but it is too small for a noticeable influence on the total electric field structure in the ionosphere.
L 213 “a function of distance from the auroral arc”:
One of the central viewpoints of this manuscript is the characterization of the electric field as a function of distance from the auroral arc, then compared with modeled fields derived from IMAGE data. The manuscript would rather benefit from a clearer explanation of the importance and novelty of measurements themselves relative to previous work. A brief comparison to Marklund (1984) is provided at the end, but this seems insufficient, unless there has been no significant progress in electric field measurements over the past 40 years prior to the BROR experiment.In the revised manuscript, we will incorporate recent studies to more clearly highlight the significance and novelty of this work.
Using tomography, we tracked the three-dimensional motion of barium clouds with a spatial resolution of 500 m. While the overall characteristics of the electric field are consistent with those reported in previous studies, our method enables a more detailed investigation of the electric field structure and its variations.L220/ L238 “Figure 13/14”. How can one interpret the values of this equivalent current in terms of electric field and altitudes? In addition, it would be helpful to show the location of Esrange as well as the auroral arc on the map to aid interpretation. Ideally , Figure 14 can be also overlaid there to provide a complete visualising of the results in order to support conclusion of the manuscript.
Figure 13 shows the heigh-integrated equivalent current inferred from ground-based magnetometers. It should therefore be interpreted as a regional-scale background current system. In this study, we use it only to check the consistency of the large-scale convection direction with the background drift inferred from the barium cloud motion.
In contrast, Figure 14 shows the electric-field pertubation associated with the auroral arc after subtracting the background convection component.
Thus, Figures 13 and 14 represents different spatial scales: Figure 13 describes the regional background electrodynamic condition, whereas Figure 14 describes the arc-related local structure derived from the barium cloud.
We agree that the spatial relation should be made clearer. In the revised manuscript, we will add the location of Esrange and the representative auroral arc position to Figure 13. We will also clarify that Figure 13 represents the regional, height-integrated equivalent current inferred from ground magnetometers, whereas Figure 14 shows the local arc-related electric field in an auroral-arc-relative coordinate system after subtracting the background convection. Because the auroral arc moves with time as shown in Figure R3, directly overlaying the full Figure 14 result on a fixed map would be difficult to interpret; instead, we will improve the figure captions and text to make the relation between the two figures clearer.
-
AC1: 'Reply on RC1', Tomoe Taki, 01 Jul 2026
-
RC2: 'Comment on egusphere-2026-1133', Anonymous Referee #2, 22 May 2026
The article presents observations of barium near a discrete auroral arc. The main purpose of the article appears to be more of a techniques paper that focuses on how the electric field across the auroral arc was determined from interactions with the barium. Imaging tomography was used to derive the drift of the barium clouds, and from which, the electric field was determined, using a simplified model by Haerendal. The derived 1-D (across the arc) electric field is presented as the primary result of the investigation.
While the technique for using barium tracers to determine the electric field in the upper E-region/lower F-region is novel and in particular using the imaging tomography is very new, the scientific results from the investigation are not particularly new. My recommendation is that the paper has more of an emphasis on the novelty of the technique being used to determine the electric field, which is indeed novel. So overall, I do think the paper merits being published because of the use of the technique, but the results are not so exciting in terms of finding variations of electric fields across the arc.
Alternatively, the authors could make the paper more interesting by interpreting the interactions between the barium and the auroral arc. Or second, the authors could frame the electric field observations in a larger context and really point what is new with the observations. Regardless, the paper as it stands right now if it is more focused on the technique is pretty good and these are simply suggestions to consider.
I recommend that the paper have the following minor revisions be undertaken before the paper is acceptable to publication:
1. There should be more citation of other work that has been done on the 1-D electric field. In particular, commenting more on rocket and incoherent scatter radar observations that can be found in Marklund, 1984 (which is cited in the article). Additionally, a good review on the state of the art of auroral electric fields is: 10.1029/2011GM001189 and https://link.springer.com/article/10.1007/s11214-020-0641-7
2. More details need to be provided about how the tomographic imaging method works and how the data are processed so they can be used in the investigation. I found this to be terse, particularly since this seems to be the novel aspect. There is a nice figure, but I think some more information needs to provided and including figures showing how the data are processed.
Citation: https://doi.org/10.5194/egusphere-2026-1133-RC2 -
AC2: 'Reply on RC2', Tomoe Taki, 01 Jul 2026
Thank you very much for your valuable comments and constructive suggestions. We will prepare a revised version based on your feedback.
In particular, we will aim to provide a more detailed description of the introduction and the experimental methods.1. There should be more citation of other work that has been done on the 1-D electric field. In particular, commenting more on rocket and incoherent scatter radar observations that can be found in Marklund, 1984 (which is cited in the article). Additionally, a good review on the state of the art of auroral electric fields is: 10.1029/2011GM001189 and https://link.springer.com/article/10.1007/s11214-020-0641-7
We plan to substantially revise the Introduction to provide a more detailed description of the observational findings and assumptions used in previous studies. We will also incorporate and discuss the papers you kindly suggested.
2. More details need to be provided about how the tomographic imaging method works and how the data are processed so they can be used in the investigation. I found this to be terse, particularly since this seems to be the novel aspect. There is a nice figure, but I think some more information needs to provided and including figures showing how the data are processed.
We will provide a more detailed description of the analysis procedure. We will also expand the explanation of the tomography method in the revised manuscript.
Citation: https://doi.org/10.5194/egusphere-2026-1133-AC2
-
AC2: 'Reply on RC2', Tomoe Taki, 01 Jul 2026
-
RC3: 'Comment on egusphere-2026-1133', John Coxon, 04 Jun 2026
Review of “Electric fields in and around an auroral arc and the inferred current system from the BROR sounding rocket experiment”
I found this article extremely interesting; I am unfamiliar with the concept of launching barium clouds to examine electric fields and found the experiment very compelling. The movie in the supplementary information was particularly cool!
I have one major comment on the methodology herein and one on the literature.
On methodology, the authors say (lines 170–177) that they fit a straight line to the auroral arc and this is shown in Figure 7. However, Figure 7 makes it abundantly clear that this method is fundamentally flawed, especially in the centre-bottom panel. I do not understand why this fitting is necessary: if you have identified the auroral arc (the green points), surely you can just measure the distance from there? If you need to interpolate between the green points, a linear interpolation across datagaps (join-the-dots) would be a better way to achieve this than a linear fit to all the points, from a visual examination. This should be rectified prior to publication.
On literature, the view of the field presented in this paper is sparse, and the post-2000 references are particularly sparse. The authors need to conduct a more thorough literature review before this paper can be published.
I have some other minor comments which should be addressed before this paper is published in Annales Geophysicae and I describe them in detail below.
Science
Lines 18–29: This is interesting, but the authors do not explain well how the different results correspond to one another. How do the following results intersect?
- Fahleson reported fluctuations of 10s mV/m and Marklund reported 100 mv/m electric fields: are these compatible with one another?
- Mozer & Fahleson reported that electric fields do not change when crossing an auroral arc boundary but Burch reported that oppositely directed electric fields are shown on either side of arts: do these contradict one another?
- Burch showed oppositely directed electric fields, Potter showed southward electric fields, and Marklund reported northward electric fields. Are Potter and Marklund’s electric fields on different sides of the auroral arc and therefore consistent with Burch and one another, or not?
Line 29, 41: What is the importance of auroral arcs being pre-midnight or post-midnight?
Lines 64–70: When you say “the observational examples presented here” I initially interpreted that as a comment on the observations in sections 2 onwards, but on reading further, am I right that you’re actually talking about the observations you’ve reported earlier in the introduction? If I am right, I would recommend making this more explicit, and perhaps introducing a paragraph break before “In this study”.
Line 88–89: You mention the word “substorm” for the only time in the text here, and so “indicating that the substorm was still in its early stage” is a non sequitur. First, you should be specific what you mean by “substorm”: I presume you mean “substorm expansion phase onset” but you should be explicit in the text. Second, this begs the question, was the experiment timed to coincide with a substorm or is this a coincidence? If the former, you should explain how you went about identifying a nascent substorm from the ground to time the launch. If the latter, you could instead write “which may indicate that a substorm was beginning at this time”. In either case, you should expand on how you know that it is a substorm.
Figure 2: I cannot see a red field of view corresponding to Esrange, why is this?
Table 1: Is it possible to also quote the magnetic coordinates, for instance, in AACGM?
Figure 2: Given that Araújo et al. (2025a) is a conference talk, I would omit this reference as it is impossible for anyone to refer to it.
Line 116: Explain what a Watec camera is.
Lines 132–134: How did you shift the data along the magnetic field line? Did you use AACGM coordinates? A Tsyganenko model? Something else?
Lines 137–140: It would be worth avoiding using red and green in the same image due to the likelihood of some readers having red-green colourblindness.
Lines 148–149: Why 1/6 of the maximum value?
Lines 150–151: What does “non-consecutive contour lines” mean? I would have interpreted this to mean that there are timestamps where the contour line is absent, but this does not make sense in the subsequent description. Further, why select the most southeastern value rather than the southernmost as previously described?
Line 154: Was the northernmost point really chosen by selecting the easternmost point and then breaking the tie with the northernmost of the ensuing candidates? This seems odd to me: is this a typo? If not, why not describe it as the easternmost point?
Lines 147–154: Why break ties by taking the western/eastern/southernmost? Why not take the average of the candidate points or something similar, or allow every point to contribute to the one-second median described on lines 157–158?
Lines 162–163: I would find it easier to look at the graph of distance than the graph of velocity; it might be worth presenting both?
Lines 172–173: I thought the clouds were too faint to be detected by the Watec images (line 112)? This appears contradictory and should be clarified.
Lines 178–186: Given that you’re combining all the points into a single plot in Figures 10 and 11, it occurs to me that instead of identifying four points per timestamp and plotting those, you could simply just take every point (at some resolution) within the cloud and plot the velocity of the point vs the distance. This would presumably tighten your statistics up.
Line 205–207: Where does the 20% or -sin(12°) figure come from? This should be derived.
Language
When referring to a figure, simply say “shown in Figure 6” rather than “shown in the Figure 6” (for example).
Figure 9: You could move the description of blue/yellow/green lines to before the (a) label and then you don’t have to keep repeating “Line colors correspond to those in (a)”.
Citation: https://doi.org/10.5194/egusphere-2026-1133-RC3 -
AC3: 'Reply on RC3', Tomoe Taki, 01 Jul 2026
Thank you very much for your valuable comments and suggestions for improvement.
We will answer your questions below, and we will revise the manuscript reflecting your comments.
I have one major comment on the methodology herein and one on the literature.
On methodology, the authors say (lines 170–177) that they fit a straight line to the auroral arc and this is shown in Figure 7. However, Figure 7 makes it abundantly clear that this method is fundamentally flawed, especially in the centre-bottom panel. I do not understand why this fitting is necessary: if you have identified the auroral arc (the green points), surely you can just measure the distance from there? If you need to interpolate between the green points, a linear interpolation across datagaps (join-the-dots) would be a better way to achieve this than a linear fit to all the points, from a visual examination. This should be rectified prior to publication.The green points were identified automatically, and in some cases the cloud was misidentified. In particular, it may be difficult to reliably use the green points during periods when the clouds and the arc are located close to each other.
To better account for the curvature of the arc, we plan to replace the current linear fitting with a higher-order polynomial fit.On literature, the view of the field presented in this paper is sparse, and the post-2000 references are particularly sparse. The authors need to conduct a more thorough literature review before this paper can be published.
We will include a broader review of relevant previous studies in the Introduction and we will revise the manuscript to clarify how the previous results are related to each other.
I have some other minor comments which should be addressed before this paper is published in Annales Geophysicae and I describe them in detail below.
Science
Lines 18–29: This is interesting, but the authors do not explain well how the different results correspond to one another. How do the following results intersect?
1. Fahleson reported fluctuations of 10s mV/m and Marklund reported 100 mv/m electric fields: are these compatible with one another?
Larger electric fields have been reported on smaller spatial scales. Marklund measured electric fields with a time resolution of 2 ms. In contrast, Fahleson reported that AC electric fields from 50 Hz to 2 kHz could be observed, but the electric field measurements were obtained at 30 s intervals.
The two results are not necessarily contradictory. The observed differences may reflect electric field variations occurring on different spatial scales within different auroral arcs. Alternatively, the electric fields reported by Fahleson may appear smaller because of the lower temporal resolution of the measurements.2. Mozer & Fahleson reported that electric fields do not change when crossing an auroral arc boundary but Burch reported that oppositely directed electric fields are shown on either side of arts: do these contradict one another?
Various electric field structures have been reported in and around auroral arcs. The total electric field is expected to vary depending on factors such as the potential gradient above the ionosphere and the strength of polarization electric fields within the ionosphere. The two observational results can be considered consistent with each other.
3. Burch showed oppositely directed electric fields, Potter showed southward electric fields, and Marklund reported northward electric fields. Are Potter and Marklund’s electric fields on different sides of the auroral arc and therefore consistent with Burch and one another, or not?
It is proposed that the electric field structures reported by Burch et al. (1976) are produced by field-aligned currents by Marklund (1984). Potter et al. (1970) and Marklund et al. (1982) both suggested that polarization electric fields were generated and played an important role. Potter et al. (1970) observed an auroral arc on the dawn side, while Marklund et al. (1982) observed an auroral arc on the dusk side. Since the background convection electric field is likely to differ between the dawn and dusk sectors, these results are not necessarily contradictory.
Line 29, 41: What is the importance of auroral arcs being pre-midnight or post-midnight?
It is known that the global convection pattern differs between the pre-midnight and post-midnight sectors. These patterns are thought to be influenced by electric potential differences in the magnetosphere along magnetic field lines.
Typically, in the pre-midnight sector, downward field-aligned currents appear equatorward of the auroral oval and upward field-aligned currents poleward of it. In contrast, in the post-midnight sector, the configuration is reversed.Lines 64–70: When you say “the observational examples presented here” I initially interpreted that as a comment on the observations in sections 2 onwards, but on reading further, am I right that you’re actually talking about the observations you’ve reported earlier in the introduction? If I am right, I would recommend making this more explicit, and perhaps introducing a paragraph break before “In this study”.
We will revise the manuscript accordingly.
Line 88–89: You mention the word “substorm” for the only time in the text here, and so “indicating that the substorm was still in its early stage” is a non sequitur. First, you should be specific what you mean by “substorm”: I presume you mean “substorm expansion phase onset” but you should be explicit in the text. Second, this begs the question, was the experiment timed to coincide with a substorm or is this a coincidence? If the former, you should explain how you went about identifying a nascent substorm from the ground to time the launch. If the latter, you could instead write “which may indicate that a substorm was beginning at this time”. In either case, you should expand on how you know that it is a substorm.
Substorm onset happend just after BROR rocket launch is coincident. In revised paper, we will give more attention about the description of general geophysical conditions around the experiment.
Figure 2: I cannot see a red field of view corresponding to Esrange, why is this?
The Esrange camera has a 180-degree field of view, providing full-sky coverage. We will clarify this point in the revised manuscript.
Table 1: Is it possible to also quote the magnetic coordinates, for instance, in AACGM?
The AACGM magnetic coordinates of each station are listed below. The analysis focuses on a very small spatial region above Esrange. Therefore, geographic coordinates may be more suitable than magnetic coordinates for the present analysis.
Table R1: The AACGM magnetic coordinates of ALIS_4D stationsLatitude (degree)
Longitude
(degree)
MLT
(hour)Abisko
65.631
99.285
20.199
Silkkimuotka
65.202
101.463
20.344
Kiruna
65.042
100.297
20.266
Esrange
65.056
100.829
20.302
Tjautjas 64.501
100.296
20.266
Figure 2: Given that Araújo et al. (2025a) is a conference talk, I would omit this reference as it is impossible for anyone to refer to it.
Since reference Araújo et al. (2025b) provides sufficient background, we will remove reference Araújo et al. (2025a).
Line 116: Explain what a Watec camera is.
These cameras are WAT-910 HX camera made by Watec Co.. Band-pass filters centered at 428 nm, 558 nm, and 630 nm were used. In this study, we primarily use the 558 nm data. The exposure time and temporal resolution for the 558 nm channel are 1 s, and the image resolution is 768×576 pixels at Kiruna and 640×480 pixels at Tjautjas.
Lines 132–134: How did you shift the data along the magnetic field line? Did you use AACGM coordinates? A Tsyganenko model? Something else?
In geographical coordinates, the horizontal plane was shifted according to the direction of the magnetic field vector at the Esrange position calculated using the IGRF model.
Lines 137–140: It would be worth avoiding using red and green in the same image due to the likelihood of some readers having red-green colourblindness.
We will revise Figure 2 to improve its accessibility and readability, including for readers with color vision deficiencies. We will also review the other figures and make similar revisions where necessary.
Lines 148–149: Why 1/6 of the maximum value?
The threshold was determined empirically. In addition to 1/6, we also tested values of 1/4 and 1/8. The fifth cloud was located close to and north of the fourth cloud. When a threshold of 1/8 was used, the two features tended to merge, making it difficult to continuously trace the intended structure. In contrast, a threshold of 1/4 resulted in selected points being biased toward the intensity maximum, reducing the spatial extent that could be tracked. Therefore, we adopted a threshold of 1/6, which provided the best balance and allowed the cloud structure to be traced over the widest area.
Lines 150–151: What does “non-consecutive contour lines” mean? I would have interpreted this to mean that there are timestamps where the contour line is absent, but this does not make sense in the subsequent description. Further, why select the most southeastern value rather than the southernmost as previously described?
As the cloud shape changes rapidly, it can sometimes appear to split apart. In other words, multiple local maxima and disconnected contour lines are present. In such cases, candidate points are selected from each connected contour, and the candidate closest to the point selected at the previous time step is adopted.
We have revised the wording to avoid possible misunderstanding. The criterion of selecting the easternmost point is applied only when multiple points exist at the same southernmost latitude. Therefore, describing the selected point as the southernmost point is more accurate than referring to it as the southeastern point.Line 154: Was the northernmost point really chosen by selecting the easternmost point and then breaking the tie with the northernmost of the ensuing candidates? This seems odd to me: is this a typo? If not, why not describe it as the easternmost point?
It was typo. It has been corrected.
Lines 147–154: Why break ties by taking the western/eastern/southernmost? Why not take the average of the candidate points or something similar, or allow every point to contribute to the one-second median described on lines 157–158?
The approach in the manuscript was adopted in order to obtain velocity estimates at multiple locations for a single time step.
Lines 162–163: I would find it easier to look at the graph of distance than the graph of velocity; it might be worth presenting both?
We will add a distance–time graph in the revised manuscript.
Lines 172–173: I thought the clouds were too faint to be detected by the Watec images (line 112)? This appears contradictory and should be clarified.
The data of the Watec cameras at Kiruna and Tjautjas are the images taken with a band-pass filter for the auroral 557.7 nm emission. The isolated round structures in these images are likely associated with strontium lines at 550.4, 552.1, 553.4 and 554.0 nm. For lower-altitude clouds, barium oxidation competes with ionisation. BaO is characterised by a rich electronic-vibrational spectrum in the visible range and may contribute to cloud images recorded by Watec cameras.
Lines 178–186: Given that you’re combining all the points into a single plot in Figures 10 and 11, it occurs to me that instead of identifying four points per timestamp and plotting those, you could simply just take every point (at some resolution) within the cloud and plot the velocity of the point vs the distance. This would presumably tighten your statistics up.
From your comment, we understand that the proposed method extracts all points from the 2D cloud maps obtained by tomography and derives a 2D velocity map for each point.
In this case, it is necessary to accurately determine the source and loss terms. We attempted to include these terms with assumptions, but the results were highly sensitive to small observational errors, leading to large variations in the estimates. As a result, we could not obtain more reasonable results than those from the current method.Line 205–207: Where does the 20% or -sin(12°) figure come from? This should be derived.
The difference is due to the inclination of the magnetic field at Esrange. The formulation should use tan rather than sin. In the revised manuscript, we will separately evaluate the contributions of the E×B drift derived from the magnetic field inclination and gravity, and provide a more detailed description and calculation.
Language
When referring to a figure, simply say “shown in Figure 6” rather than “shown in the Figure 6” (for example).
Figure 9: You could move the description of blue/yellow/green lines to before the (a) label and then you don’t have to keep repeating “Line colors correspond to those in (a)”.We thank you for your helpful comments. We will address these points and incorporate the revisions into the revised manuscript.
Citation: https://doi.org/10.5194/egusphere-2026-1133-AC3
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 622 | 307 | 73 | 1,002 | 92 | 63 | 85 |
- HTML: 622
- PDF: 307
- XML: 73
- Total: 1,002
- Supplement: 92
- BibTeX: 63
- EndNote: 85
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Review of „Electric fields in and around an auroral arc and the inferred current system from the BROR sounding rocket experiment” by Taki et al.
This paper investigates a recent rocket experiment in which barium was released into the auroral ionosphere to measure electric fields by tracking the motion of the ion cloud using optical observations. Optical measurements of an auroral arc at 120 km altitude and of the barium ion cloud at 230 km were used to infer the surrounding electric field system.
The objective of the experiment is interesting, and the data visualization is well presented. However, the applicability of the data fusion between the two optical measurements and the novelty of the results are not clearly demonstrated. In addition, both the abstract and the introduction require improvement to more clearly convey the study’s background and the linkage to main findings. For example, the altitude variation of the electric fields may be of significant importance for this study.
A major concern on the dataset is that the barium cloud is observed at 230 km, whereas the auroral arc is located at 120 km. The authors should explain more clearly how this large altitude separation is compensated for in their analysis. After this can be done convincingly, the novelty of the electric‑field measurements should also be presented more clearly. Another aspect is whether the derived electric fields represent the background field and/or the short‑term fluctuations. Since the method should have high time resolution, it would theoretically capture fluctuations, but the text does not make this clear.
Below are specific comments referenced by line numbers (L). These points should be addressed before the manuscript can be considered for possible publication.
Abstract
L1 and Abstract: The current abstract does not include any direct results or new findings of this paper itself. It describes the novelty of the experiment, but not the novelty of the findings.
L4 “The effective approach”: Could the authors clarify in what sense this approach is considered effective?
L7 “A novelty of the experiment is…”: At L94, the authors state that the difference between ALIS and ALIS_4D is the use of an EMCCD instead of a CCD. Combining these two sentences, does this mean that the only novelty of the experiment, compared to previous studies, is the use of an EMCCD?
L11 “quite a long time”: How long is this period?
Introduction
L15 “current systems”: Could they also explain more why the current system is important for understanding this coupled ionosphere – magnetosphere system?
L15” Various observational approaches”: The introduction discusses rocket, satellite, and ground‑based measurement approaches. These techniques have different horizontal and vertical observation ranges, but the current text does not describe these differences and their implications.
L20 “were small, less than 10 mV/m”: It may be useful to mention how small these values are compared to the background.
L21 “altitude dependent … were reported”: This sentence appears inconsistent with the statement at L22 that “no significant altitude variation was observed.”
L24 “On the other hand, charge oscillations were observed”: The meaning of this sentence and relation to the previous sentence are unclear. Please rephrase it.
L26 “spacecraft”: This literature appears in the middle of a paragraph introducing rocket studies. How is it related to that context? Spacecraft studies are introduced starting at L37.
L37” attempted”: The authors could add more information to this sentence explaining why satellite measurements were attempted.
L42 “strong”: How strong are these?
L44 “relatively large”: Relative to what, and how large? Please specify.
L45 “… However…”: The reviewer does not understand the implication of this “however” or the logical connection between the preceding and following sentences. Please clarify the intended contrast.
L50 “Strong”: Please explain how strong, and relative to what.
L51 “3 s,”: Is this about optical or EISCAT observation ?
L53 “the two”: Please clarify what “the two” refers to.
L53 “response”: Does this imply any interaction between the radar and the electric field?
L57 “downward field-aligned currents”: Could the authors describe how these currents are related to the electric field?
L58 “background electric field was…canceled by a polarization electric field”: Please explain the mechanism responsible for this.
L 59 “this study”: Please consider rephrasing “this” to avoid possible ambiguity.
L62 “but have relatively limited resolution. However,…”: The reviewer does not understand how these sentences about Clayton et. al., are connected by “but” and “however.” Please rephrase.
L66 “As a result, it was difficult to separate temporal variations from spatial variations based on the observations alone.”: The reviewer does not understand this as a conclusion from the preceding sentence.
L67 “temporal and spatial scales smaller than… not be detected”: This limitation applies to any observation technique. What are intended to mean here?
L68 “barium clouds”: Active ionospheric experiments using chemical releases, including barium, have been conducted for many years. The authors cite Fahleson et al. in the beginning, but additional previous studies and results relevant to the present work could be introduced if there are any.
Experiment
L75 “23 March 2023”: The storm events in March 2023 including March 23 have been studied by several authors already. Can authors find any relationship between their study to previous studies? e.g,
https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2023JA032145
https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2023SW003728
As seen in these papers, the EISCAT radars were operating on March 23, and one of their objectives should have been to support the BROR experiment. Were there any EISCAT results that could be relevant to the current work? Also, with regard to previous EISCAT–rocket studies, such as the cited works by Lanchester et al. or Aikio et al.? In particular, ion‑drift measurements or electric field/FAC estimates derived from the radar data would be highly relevant.
L84 “moved under influence of the electric field”: Similar to the comment above, strong thermospheric winds were reported on 23-24 March 2024 storm (see Oyama et al 2024 above). Did the authors estimate the possible influence of thermospheric winds?
L96 Figure 2: What does the yellow point on the map represent? No description is provided.
2.2 Instruments and Data
L104 “strontium”: Why was only the camera at Esrange used to measure strontium? In addition, were any results from these strontium measurements considered in the analysis from Esrange in Figure 3?
L112 “too weak to be detected”: Is this conclusion based on experimentally established results from previous studies, or is it derived from the observations in the present study? In Figure 4, both cameras show isolated round structures near the arc (for the Kiruna camera, one appears near the center of the image). What are these features?
3 Data processing and analysis
L118 “Figure 5”: Please indicate where the fourth cloud is located in all of the Ba‑related sub‑figures in Figure 5, and specify how many clouds are present in total.
L123 Figure 6: This reconstruction figure seems that some of the early ion clouds reached lower altitudes, close to the auroral altitude around 120 km. If this interpretation is correct, these clouds may be more suitable for analyzing the electric field in the vicinity of the aurora.
L125 “a vertical range of ±10 km around the altitude at which the cloud intensity reaches its maximum.”: Please explain in more detail how the altitude of maximum Ba intensity is determined. Is this done by comparing mapped images using altitude as a parameter from all four stations?
L 133 “we shifted the cloud data to the height at an altitude of 120 km”:
The objective of this projection needs more explanation.
This reconstruction procedure appears to assume that electric field variations can be ignored over the 100‑km altitude range between approximately 120 and 230 km. How do the authors justify this assumption in the case of an auroral arc?
Moreover, this projection gives the impression that the analysis does not fully utilize the capabilities of the experimental design. If the reviewer interpreted it correctly, as the tomographic results in Figure 6 show, the total set of ion clouds spans a much larger altitude and horizontal range than the fourth cloud alone, with some clouds extending down toward auroral altitudes. In principle, this experiment could provide not only horizontal motion but also information on vertical deformation/motion.
Also, similar to a comment before, the EISCAT Tromsø radar was operational on that day and should provide plasma parameters, and likely most relevant for this study, altitude variation of plasma parameters including ion‑drift.
L138 “…green points. A blue line…”: The blue lines in Figure 7 do not appear to represent the observed auroral arc accurately, as they differ noticeably from the grayscale structure. Fitting a straight line to an auroral arc does not seem physically justified or effective, especially at 18:30 UT. At 18:25, no auroral arc is visible in the image, but only the blue lines are shown.
L139 “The fourth cloud projected”: The reviewer assumes that the red tracings in Figure 7 represent the projected cloud structures from Figure 6. However, at 18:25 these appear only as very small red points in Figure 7. Based on Figures 6 and 7 alone, it is difficult for the reader to verify whether the red circles in Figure 7 correctly correspond to the fourth cloud shown in Figure 6.
To improve clarity, the authors should consider adding the auroral arc outline also in Figure 6 for comparison, and/or provide the geolocation (latitude, longitude, and altitude) of the fourth cloud. This would allow readers to follow the temporal and spatial development of the fourth cloud more easily.
In addition, as noted earlier, the bright round structures in the auroral images create confusion if they are not related to the barium clouds. Are these bright features indeed unrelated to the BROR cloud releases? A clear explanation is needed to avoid misinterpretation.
L147 “We selected four points to characterize the cloud deformation”. The reviewer assumes that this paragraph is intended to describe Figure 8, although Figure 8 is never mentioned in the main text.
L162: “7-s moving average”: High‑resolution imaging was a key novelty of this experiment. Please explain whether this novelty is preserved when the data are averaged.
L164: “60s, 100 s”: Please specify the times in UT so that readers can compare them directly with the optical figures.
L172 “Watec images also include BROR clouds”: This sentence seems to contradict the statement in line 112, creating confusion for the readers. Please explain this point more clearly.
L178 “distance of each point from”: Related to earlier comments, is the projection from 230 km down to 120 km include any geometric effects? For example, can the spatial distance between two points become larger or smaller after the projection? Any such distortion could directly affect the distance/velocity calculation.
L188 “vertical velocity along the z-axis”. Please clarify whether this computation is performed before or after the tomographic reconstruction shown in Figure 6. If it is derived after tomography, then the vertical resolution is limited to approximately 20 km, the it could impact the interpretation of Figure 11.
A more essential question concerns the vertical motion of the barium clouds: how much of the observed descent can be explained by natural gravitational settling, and how much is attributable to electric‑field–driven motion?
Aslo, in Sergienko et al., 2024, a vertical motion of ions was reported as an unexpected finding. Is there any connection to the current study?
4 Results and discussion
L193: This section 4 is titled “Results,” but in practice the material presented in Section 3 also contains results. The distinction between the two sections is therefore unclear.
L199 “for the altitude of approximately 230 km treated in this study,”: One of the most significant methodological questions in this study concerns how the lower‑F‑region electric field is being characterized when the mapped auroral structures are located at E‑region heights. The manuscript does not clearly explain the physical justification for interpreting an electric field at 230 km based on features mapped to ~120 km. The E‑region is highly collisional compared to lower F-region, so the electrodynamic behavior, and therefore the inferred electric field may be sustainably different.
L 213 “a function of distance from the auroral arc”:
One of the central viewpoints of this manuscript is the characterization of the electric field as a function of distance from the auroral arc, then compared with modeled fields derived from IMAGE data. The manuscript would rather benefit from a clearer explanation of the importance and novelty of measurements themselves relative to previous work. A brief comparison to Marklund (1984) is provided at the end, but this seems insufficient, unless there has been no significant progress in electric‑field measurements over the past 40 years prior to the BROR experiment.
L220/ L238 “Figure 13/14”. How can one interpret the values of this equivalent current in terms of electric field and altitudes? In addition, it would be helpful to show the location of Esrange as well as the auroral arc on the map to aid interpretation. Ideally , Figure 14 can be also overlaid there to provide a complete visualizing of the results in order to support conclusion of the manuscript.