the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 01 Jul 2025
Characterizing anomalous geomagnetic induction from coastal effects with transfer functions and gradient measurements
Abstract. The occurrence of anomalous subsurface currents in a region is of significant geophysical importance. Several geomagnetic methods have been developed to characterize the effects of geomagnetic induction. Typically, the intensity and direction of the inducing processes are determined using simplified transfer functions that relate corresponding horizontal and vertical geomagnetic components. The geomagnetic field associated with nearby anomalous currents is expected to be inhomogeneous. Additionally, the distortion of the measured geomagnetic field's geometrical structure, as indicated by the magnetic gradient, is a relevant parameter in this context. In this comprehensive study, both methods are applied to simultaneous measurements of geomagnetic induction and magnetic gradient conducted at two geomagnetic repeat stations near the Adriatic Sea. Furthermore, a novel concept for a magnetic gradiometer is introduced. A strong correlation was observed between the direction and intensity of the calculated Parkinson vectors and the strike directions and intensities of the horizontal geomagnetic gradient during periods influenced by the seaside effect. The primary conclusion is that geomagnetic gradient measurements are highly effective for characterizing the effects of geomagnetic induction and/or quasi-stationary subsurface currents.
- Preprint
(1093 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 26 Sep 2025)
-
RC1: 'Comment on egusphere-2025-2562', Anonymous Referee #1, 01 Aug 2025
reply
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-2562/egusphere-2025-2562-RC1-supplement.pdf
-
AC1: 'Reply on RC1', András Csontos, 20 Aug 2025
reply
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-2562/egusphere-2025-2562-AC1-supplement.pdf
-
AC1: 'Reply on RC1', András Csontos, 20 Aug 2025
reply
-
RC2: 'Comment on egusphere-2025-2562', Anne Neska, 25 Aug 2025
reply
Dear Dr. Csontos,
I am writing this in the role of a reviewer.
The submitted manuscript „Characterizing anomalous geomagnetic induction from coastal effects with transfer functions and gradient measurements” by A. Csontos contains two novel theses which are (1) a Declination-Inclination magnetometer measurement can be used to determine the magnetic gradient, (2) the magnetic gradient can be used to characterize the anomalous magnetic fields (as captured in Parkinson arrows) caused by induction, e.g., close to a sea coast.
Thesis (1) assumes a link between possible use of an instrument type that usually is applied in measurements of the main geomagnetic field in observatories and repeat stations, and a quantity that is caused by local to regional geology due to spatially varying magnetic properties of rocks in the Earth’s crust and is usually measured as the difference between two simultaneously operating magnetometers. Establishing this link appears to be an achievement of the Author and it is a good question to which degree this link can be, or is already, accepted by the community. I will not consider this question here but I will assume in the following that this idea of the Author is correct and that his determination of the magnetic gradient in this way is valid.
I will focus on thesis (2). The equivalence of both quantities is demonstrated in the manuscript in Figures 7 (depicting a vector representation of the magnetic gradient at two repeat stations KRBP and SINP close to the Adriatic coast) and Figures 5 (Parkinson induction arrows at KRBP) and 6 (the same at SINP). Real induction arrows are roughly parallel to the azimuth of the gradient, with a 180 degree difference in the direction. So there is a similar pattern in both quantities indeed.
However, I do not agree that this means that a gradient measurement can be used to characterize induction effects. This is an inadmissible generalization and there is a a number of serious methodological problems and unclarities in it as I will indicate in the following.
-
The magnetic gradient measured at surface is a consequence of the distribution of rocks with various magnetic properties in the subsurface. Anomalies of the induced magnetic field are a consequence of inhomogeneously distributed electrical (i.e., conductivity or resistivity) properties of rocks in the subsurface. For the rocks beneath the two stations shown in the manuscript it seems that their resistivity and magnetic properties produce a similar pattern at surface. This is well possible (especially in the tectonic context of the Adriatic and other coasts formed by subduction) but does not mean that it is always like this. In general, a subsurface structure may reveal a conductivity anomaly but no magnetic gradient, and vice versa. Many more measurements like the ones considered here (and in various geological settings) would be necessary to support the link stated by the Author.
-
The magnetic gradient usually is considered to be a quantity that does not depend on time or frequency (if we neglect temporal changes in the subsurface and some magnetic response on strong external field variations) but a static one. In contrast, induction vectors cannot be a property of the static magnetic field. They are functions of frequency or period, and a magnetic variation measurement is necessary to statistically derive the underlying transfer function. The Author acknowledges this in equation 1 (frequency omega), in line 329, and by listing the periods for the arrows in Figs. 5 and 6. It is even visible that the arrow lengths and directions are different for different periods. I want to emphasize that this is not a mistake or scattering of noisy data, it is a normal property of an induction arrow. An induction arrow at one station can easily change its length by the 10-fold and its direction by 90 degree between one minute and one hour periods, to give an example, so it can produce two totally different patterns. The gradient cannot. For this reason a gradient measurement cannot characterize an induction anomaly. The induction phenomenon needs a varying magnetic field at a particular frequency or period. I am not fully confident that the Author is aware of this since the question at which period the induction arrow should be compared to the gradient remains untouched.
-
For each of the considered stations, two sets of induction arrows are presented since two different remote reference stations have been used during transfer function estimation. In general it is good to use remote reference data to suppress effects of local noise, and of course it is good to test this with more than one reference station to ensure correct results, or select the better (less noisy) one – both results are slightly different so this matters. Some final decision about the accepted version of the arrow should have been made in the work where it has been derived. In the present work it is only cited and results for both reference stations are shown without any comment, explanation, discussion, or decision. It is unclear what this is for.
-
According to Table 2 and line 393, time series fragments of just 128 minutes length have been used to derive induction arrows, and the longest period for which arrows have been derived is 64 minutes according to captions of Figs. 5 and 6, i.e. one half of it. This is somewhat strange because usually time series fragments must be much longer than the longest period for which transfer functions are estimated for reasons of statistics. However, since transfer function estimation is not the subject of the present but of some cited work, this issue is difficult to assess.
-
It is not clear what the arrow length in the vector representation of the magnetic gradient means. The scale of it is clearly different from the induction arrow length. Furthermore, gradient arrows are clearly shorter for KRBP than for SINP, whereas induction arrows have comparable lengths for both stations.
Concluding, hypothesis (2) of the Author could not be proven. I am afraid that confusion of static and frequency-dependent magnetic field properties has led to a methodically incorrect approach. Furthermore, the results that support the Author's thesis cannot be generalized so easily, and there is some lack of clarity as to data quality of the induction arrows. I am sorry but I recommend rejection.
Best regards,
Anne Neska
Citation: https://doi.org/10.5194/egusphere-2025-2562-RC2 -
-
EC1: 'Comment on egusphere-2025-2562', Anne Neska, 25 Aug 2025
reply
Dear Author, dear Reviewer,
I want to start my editor comment from sharing how difficult it was to find reviewers for this manuscript. I think this is because it spans even three domains of geomagnetism – instruments, gradients, and induction. I asked colleagues whose great general knowledge I knew, colleagues who have experience in more than one of the domains, and colleagues who I thought might be familiar with the topic due to a similar affiliation as the Author. Six of them did not agree, one gave a feedback that he once reviewed a similar looking work and was not convinced. Hence I much appreciate that Referee #1 accepted the challenge, carefully read the manuscript, and gave a sound opinion on it.
Because of the mentioned difficulties I wrote the second review myself, which is of course a conflict of roles and not a good approach to the editorial process, but it is practiced in some cases and I got the permission of Executive Editors of the Journal for this case.
I confirm the observation of Referee #1 that it is an issue that a significant part of the scientific content and argumentation of the manuscript is situated in other publications. This makes it very difficult to assess part of the argumentation and raises questions as to originality of the work. Also I have serious doubts about methodical correctness of the approach as pointed out in RC 2 (https://doi.org/10.5194/egusphere-2025-2562-RC2). For these reasons I decide against publishing the manuscript.
I am sorry to do so because I have some sympathies for unusual methods caught in the middle. But for being successful with such approaches one has to become – at least to some degree – an expert in more than one domain and this is not easy.
Best wishes,
Anne Neska
Citation: https://doi.org/10.5194/egusphere-2025-2562-EC1
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 267 | 30 | 13 | 310 | 16 | 24 |
- HTML: 267
- PDF: 30
- XML: 13
- Total: 310
- BibTeX: 16
- EndNote: 24
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1