Characterizing Pulse Attenuation of Intra-Cloud and Cloud-to-Ground Lightning with E-Field Signal Measured at Multiple Stations

Round 1

Reviewer 1 Report

General comments:

The submitted paper is a comprehensive and interesting study under estimation of attenuation properties for LEMP in presence of a lossy ground. The paper is well organized and its literature study is completed. It can be a great contribution to the Remote Sensing MDPI, however, some minor changes and improvements depicted below could be made.

1. Abstract: The “b” parameter of y=ax^b was given but some information about “a” parameter should be also included or why “a” is not considered. The same remark for the 3^rd paragraph of Introduction section.
2. Methodology: In this section there is assumed that r>>H, so the attenuation function parameter “b” was not estimated for close range to the lightning channel? The results for close distances to the lighting channel can differ highly because the electrostatic and induction components of lighting electric field can reach higher peak value than the radiated component. Some comment on this and limitations of the proposed method should be given.
3. Equation 4: There is some conflict of variable naming between (3) and (4). From comparison of those two equations the E_p is defined ambiguously. If in (4) attenuation is introduced then E_p should be e.g. E_pa - electric field peak attenuated.
4. Equation 5: The change of name of variable “V” would be beneficial to the reader. “E_pag - electric field peak attenuated with gain taken into account, or E_pm - electric field peak measured” naming could be used because it is still the electric field signal. Moreover the information that this variable is taken from measurement should be given in the text.
5. Equation 5: Parameter “K” should be definded.
6. Last paragraph of 2^nd section: The reason why “the constant coefficient” is not considered in this paper should be given.
7. Last paragraph of 2^nd section: The influence of radius and thickness of circular ring could be estimated for given results i.e. the uncertainty of computed “b” attenuation coefficient should be given e.g. in the table 1.
8. Section 3, paragraph 1^st: the given bottom cutoff frequency of 800 Hz for the system greatly reduces electrostatic component of lighting electric field. Some comment about the influence of this on obtained results could be given.
9. Section 4, 1^st paragraph: “The results are shown in Figure 7” - this statement should be verified because it does not show results but rather the graphical method.
10. Figure 9: There is some collision of variable names between equations given in Fig.9 and the two above equations based on y=ar^-b. The variable “y” is not the same variable in equations and in figure.
11. Table 1: The 10 km width of distance ring seems to be a high distance, some comment about uncertainty of analysis could be done.
12. (3) point of 4^th section: In numerical simulation the height for CG pulse was assumed to be 200 m but the lighting channel should be rather treated as a vertical dipole. Why this height was assumed?
13. (3) point of 4^th section: The center frequency of 40 kHz for CG lightning seems to be a little bit too high. It rather should be 10 kHz. Give some comment on this.

Detailed comments:

1. Introduction, 2^nd paragraph: “lightning EMP” should be changed to “LEMP”
2. Introduction, 3^rd paragraph: Parameters “A” and “D” appearing in E=A/D*10^-aD should be also defined in the text.
3. Whole text: LEMP is impulsive therefore word “amplitude” should be replaced by the “peak value”.
4. Line 6^th in the last paragraph of 2^nd section: HF and JS shouldn’t be swapped?
5. Figure 2: Figure should not end a section.
6. Figure 3 and 6: The information about the range of km scale would be beneficial to the reader.
7. Figure 4: “Waveform of CG lightning strokes” should be changed to “Waveforms of the same CG lighting stroke”. Similar comment to Fig.5.
8. Figure 6: The blue circle should be defined in figure caption.
9. Section 4: Three subsections should be defined within this section for (1), (2) and (3) points given within this section.
10. Figure. 7: The caption of this figure should be verified because there are three panels not only for HF-FN station but also for other combinations of stations.
11. Figure. 7: What is parameter “d” in the figure, please define it in the figure caption.
12. Section 4, paragraph 4^th: The reference for Idone’s work should be done in the text.
13. Section 4, paragraph 5^th: The two equations given in this paragraph should be numbered.
14. Figure 10: The word “HF stations” shouldn’t be changed to “HF station”?

Finally, the paper needs a minor revision, mainly related to additional explanations of some topics according to Reviewer suggestions given in general comments and a slight vocabulary-grammar correction depicted in detailed comments.

Author Response

Please see attachement

Author Response File: Author Response.docx

Reviewer 2 Report

1. On page 5: "When a lightning event is located on an annular ring at the same distance from the HF station, its distance from the JS station is different." And in Fig. 2 the ring is drawn around the JS station?
2. In the method IC discharge is it assumed to be a vertically oriented segment at some height above the ground? If so, what error will be introduced into the methodology by arbitrarily space-oriented IC discharge channels?
3. How can an error in the "separation" of intra-cloud discharges and lightning discharges into the ground affect the proposed methodology?
4. In Figure 7 it is not clear what is d = 5 km?
5. The coefficients b in Fig. 9 and in Fig. 10 do not correlate with the same coefficients in rows 1 and 2 of Table 1.
6. In page 15 probably supposed to mention Figure 12, and not Figure 10?

Author Response

Please see attachment

Author Response File: Author Response.docx

Reviewer 3 Report

This is a review of the paper by W. Wang and B. Zhu entitled "Characterizing pulse attenuation of intra-cloud and cloud-to-ground lightning discharges with E-field signal measured at multiple stations". This paper presents a new method to measure soil conductivity from the measurement of electromagnetic waves emitted by lightning by using the amplitude ratio of the same lightning recorded by two antennas to avoid having to finely characterize the gain of these antennas. The method is detailed and then illustrated on measurements made in August 2019. The conductivities found are compatible with those found by another method and an electromagnetic propagation model. They also highlight that the waves emitted by the CIs do not undergo attenuation related to the conductivity of the soil as is the case for the CGs.

This method is interesting and this study deserves to be published. Nevertheless, clarifications are needed and corrections must be made (some figure legends are not completely appropriate).

Specific comments:

Page 2: change "Claudia R. de Mesquita" to "de Mesquita et al."

Page 2: I think it should read "they used two quantities of EM field magnitude"

Page 2: change "Ivana Kolmašová" to "Kolmašová et al.

Page 2: what do you mean by "the constructed four E-field receivers"? One of the sensors measured the magnetic field.

Page 2: this statement is wrong "all the stations were located beyond 600 km", one station was less than 100 km away and the 3 others between 300 and 600 km.

Page 2: change "by Idone is different" to "by Idone et al. (1993) is different".

Page 3: I think "removed" is better than "declined"

Page 3: a sentence introducing your numerical simulations is missing.

Page 4: this equation (1) is not introduced in reference 12.

Page 4: "the observation distance is relatively far" can you specify the range of distance ?

Page 4: you can write " Denote A(I)= " since you use it so in equation (3).

Page 4: remove the "=" sign after "is"

Page 5: replace "K" with "k" in formula (9)/

Page 5: you can add "relatively to lightning flash" after "by fixing the distance of one station".

Page 5: it appears that compared to what is shown in Figure 2, you have reversed the HF and JS stations.

Page 6: please recall the main performances of JASA: detection efficiency for CG and IC, as well as location accuracy.

Page 6 : "we can manually determine the type of discharge" what are your criteria for identifying and differentiating between CG and IC ?

Page 7, Figure 4 : Is it useful to have 4 digits after the decimal point for the distance ? Give the date of the flashes ('Peaktime') in a readable format or delete it.

Page 8: "we can identify the IC pulses and CG pulses according to the signal waveform, and eliminate some discharging events with relatively weak magnitude. "How many events are you considering (in Figure 6) before and after the selection you make?

Page 9: the legend of Figure 6 does not really describe the figure. The blue circle seems to center on JS. The flashes shown are not only at 200-210 km from HF (the area covered on the map is approximately 900 x 1100 km).

Page 10: the legend to Figure 7 does not describe the Figure.

Page 10: "In the practical calculation, in order to enlarge the sample size as much as possible, we choose the lightning events within 5 km range from HF station and FN, HB, JS station." What is the impact of the location error on filtering the data with a 5 km thickness?

Page 11: The graphs should be repeated by plotting the ratio only between 0 and 2. What is the meaning of a negative ratio? We would like to see the average value per 50 km of distance to see if the ratio changes with it. The legend could be simplified as follows: "Observations of field gain at equal distance for HF-HB station (a); for HF-FN station (b and for HF-JS station (c)."

Page 11: Figure 13 does not exist.

Pages 12 and 13: add a scale with decimal distance values on figures 9, 10 and 11 for better readability.

Page 12: "For the IC events and CG events at range of 205 km (200-210 km) from HF station", it seems on the figure that the ring is centered on the JS station.

Page 12 : "Similarly, for the CG and IC events occurring at a distance of 150 km (140-160 km) from HF station," Idem

Page 12 : "Figure 11 shows the variation of ratio between the E-field magnitude of CG and IC events occurring at a distance of 250 km (245-255 km) from HF station" Idem

Page 14: I do not think it is necessary to capitalize NOT.

Page 14: change "the attenuation index obtained by Idone according" to "the attenuation index obtained by Idone et al. according"

Page 14: change "he could not improve" to "they could not improve"

Page 14: can you give references on COMSOL?

Page 14: The maximum energy of the GCs is measured between 5 and 10 kHz. Justify your choice of 40 kHz and 120 kHz respectively for the GCs and IC.

Page 14: The conductivity values given here are missing a unit.

Page 15: In your discussion of the simulation results you can also add a reference to the following study: Aoki, M., Y. Baba, and V. A. Rakov (2015), FDTD simulation of LEMP propagation over lossy ground: Influence of distance, ground conductivity, and source parameters, J. Geophys. Res. Atmos., 120, 8043-8051, doi:10.1002/2015JD023245.

Page 16: why are the data not available?

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 4 Report

In their paper the authors give a very complete and thorough presentation of their measurements and data. Also the interpretation is thorough. I thus recommend the paper for publication, however only after a few corrections and clarifications have been made.

p8: the authors note that two stations detecting the same pulse from a flash from the same distance for the two stations obtain different amplitudes. This raises two questions:

A) From the text it is not clear if the cross calibration has been applied as discussed at the end of section 2. This should be expressed more clearly in the text.

B) It is not clear what conclusion should be drawn from their finding.

Fig 6: Do indeed all dots in the figure correspond to CG flashes of are there also CC discharges? In their model, eq 1 the authors clearly assume a CG flash, how has it been verified that all pulses included in the present analysis are indeed CG?

Fig 7: the figure caption applies to a single panel only.

Fig 9: Based on the work of ref 16, and also from the derivation by the authors of eq 1 I would expect a dependence of the signal strength

 Y (r)= (a/r) exp(-br)     or      log(y) = log(a/r) – br

Which differs from the formula used by the authors in fig 9. This needs more explanation as it seems it may be easier to plot log(y r) v.s. r.

In the caption of fig 9 the authors should write what is plotted on the y axis and what the units are. This cannot be strength as it may be negative. In the formula displayed in the figure it should be written what x is (including units). Same applies to the following figures that are similar.

Fig 12: state in the caption clearly what is plotted on the y-axis, including units. State also what is meant with x and y in the inset. State the units for sigma. Argue why the distance axis is taken linear while it was log in the previous plots.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 3 Report

see attached file

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 4 Report

The authors have responded to my previous remarks in an adequate way. The paper should be published, it is a nice comprehensive study.

Author Response

非常感谢您的点评。

我已经仔细检查并更正了文本中的拼写。