Registration of Atmospheric-Electric Effects from Volcanic Clouds on the Kamchatka Peninsula (Russia)

Round 1

Reviewer 1 Report

There are three main conclusions to this paper.  One is that field-mills can be useful in  identifying ash-clouds as they drift, with possible application to air safety.  The second is that there are "globules" of differing charge that pass overhead as the cloud drifts and that this charge inhomogeneity can be seen. The first two conclusions are nice and the paper supports them well.

 

The third conclusion is that volcanoes emit "conventional" ice/water mechanism lightning like a "dirty thunderstorm".  While the paper provides data in Figures 7 and 8 that supports this conclusion, this third conclusion is already well established in references that the authors already cite.  The only bit that might be new is that WWLN was used to identify the conventional lightning as opposed to faster instruments or networks like GLD360 or home-build slow-antennae.  In fact I think other studies have also use WWLN to see volcanic lightning.  In this case, the third conclusion is not original, and should

be phrased as "we confirm the results of others", rather than being a new conclusion.

 

This paper has some nice results, but they make the reader work too hard to check them.  Figures 3 and Figure 4 are both nice results which are purported to show Ez anomalies caused by a drifting charged ash cloud. 

I agree that the "anomalies" pointed out look like they could be from a volcanic cloud (they do not really look like thunderstorm anomalies, for example). However the result would be strengthened by some other argument that these variations really do correspond with an ash cloud.  Since the authors have video observations, they could cite in table 2 in how many cases their reported anomalies correspond with visual observations of an ash cloud over the sensor. Another way to convince the reader that these anomalies are from an ash cloud is via wind speed, distance and time of eruption. For example, in figure 3 they also show a sounding indicating upper level wind speeds and directions. The authors would strengthen their case if they would clearly state (near Figure 3) the distance and direction between the volcanic eruption site and each of their sensors.  (They do show a map, and this sort of information is scattered around the paper, but it would be good to put it all in a single paragraph.

For example, in Figure 4 there are purported ash clouds that all seem to arrive about 90 minutes after an eruption.  Why not show that for these three examples (at least) the time of arrival is consistent with upper level wind speeds, distance and direction to the measuring stations?

 

On lines 188-191 is one of the more significant scientific statements in the paper.  They do not support this conclusion that some explosions are magma fragmentation and some are gas phase.  Perhaps experts in this new field already know that magma charges negative (for example) while the lighter particles charge positive (that's just a guess ... I don't know this myself).  If this is the case they should cite references supporting this -- or are they merely speculating.   If they can't support this, these lines should be deleted.

 

Throughout the paper, the authors use the term "electromagnetic pulses".  There are lots of complexities in lightning waveforms, including pulses at frequencies from 10~Hz to 1~GHz (though I think WWLN only works to about 20 kHz).  The general term used to refer to them is "sferics".  I was going to suggest the authors replace every use of "EP" with the word sferic, or sferics.  On the other hand, it seems that they mean something in particular by “electromagnetic pulse” because they count them.  See my next comment.

 

Figure 7ab and Figure 8ab show EP activity vs. time.  The figures makes sense in terms of prior publications. Some of them (e.g. 7a1, 7a4 and 7a5) even seem to show an initial compact "vent discharge" period followed by a broader convective "plume discharge" peak. However, at no point in the paper do the authors define an algorithm for EP pulses. This is very instrument dependent.  Perhaps it is

a standard WWLN product.  Regardless, this algorithm either needs to be explained in the Instrumentation section or a citation to a methods paper for WWLN (that defines “pulses”) be provided.

 

Around line 74 the authors talks about EF-4 field mills in the instrumentation section. As an experimentalist myself, I support bragging about ones equipment, but I request more detail.  A motor part number or specification might be helpful (as appropriate motors for field mills can be difficult to find).

Also, they mention "low electrical energy consumption".  This is a useless statement if they do not specify the consumption.  Do they mean one watt?  100 milliWatts, why not quote voltage and current? [I must state that I reviewed this paper on a long airplane flight, and so I did not have internet access to check the reference on the EF4.]

 

Also -- they should specify that the EF-4 is only the analog section, the digital section appears to be an E-24 (and they don't specify the power consumption there.  So ... if going to talk about the EF-4 – then provide more detail, or point at reference 4 sooner if that has all the detail).

 

Finally, on line 82, they refer to a "Thiner board".  I have never heard of a "Thiner board".  Do they simply mean "motherboard?".  I ask the authors to recheck reference 4 and see if this is an error.

 

===== TYPOS, UNCLEAR STATEMENTS AND AWKWARD/UNGRAMMATICAL ENGLISH =====

 

This paper represents a large data set of electric-field anomalies caused by

volcanic ash clouds and as such is worthy of publication.  The English

remains a bit shaky, but if the authors will take most or all of my suggestions,

their paper will be easier to read and the results better appreciated.

 

Line number vs. awkward expression, grammatical errors, or use of incorrect or unclear vocabulary.  For each problem I suggest a solution.

 

 

20 During the explosive ---> During explosive

 

22 in the result  --->  as a result

 

26 particles falls near a ---> particles fall near a

 

29 Control of electric field ---> Measurement of electric field

 

32 radio signal direction finding at the frequency of 63 MHz occurring from volcanic ---> radio source locations (centered at 63 MHz) of impulses emitted by volcanic lightning

(The authors are referring to a lightning mapping array.  Unlike the crossed coils that they ALSO use, it's not really correct to call it a "direction finder", and their language could be clearer in general.)

 

39 I am not a volcanologist, but I do not know what the "auto-model" section of

a volcanic plume is, and I suspect that they have translated a Russian word incorrectly.  I recommend rephrasing as follows:

length of 1-7 km were observed on the auto-model section (Near-vent Lightning);

length of 1-7 km. (Near-vent lightning);

 

40 buoyancy section ---> buoyancy stage (or buoyancy phase)

 

58 Instrumentation and observation method --> Remove it ... redundant.

 

59-60  This sentence is very confusing.  I believe

the authors want to say this:

By measuring AEF Ez near the Northern group of volcanoes, we hope to detect the passage of eruptive clouds over the neighborhood settlements.

 

74 tow ranges --> two ranges

 

87 "is intended for recording weak signals" ---> "is NOT intended for recording weak signals"

 

95 agreed cable line ---> dedicated cable

 

97 I have never heard of a "correct eight" directivity pattern.  I do believe that the authors are referring to a fairly standard crossed-coil magnetic induction detector that allows one to simultaneously measure two orthogonal components of magnetic field (and thus to, indeed, have a single-site direction finder).  They should probably cite another article (for example there are articles about the National Lightning Detection Network in the USA that explain the crossed coil method ... and I think these authors used similar techniques.)

 

106 "destroyed volcanic constriction" ... This language is too incorrect to understand ... they must try again ... or remove the phrase entirely..

 

107 basis ---> base

 

139 "Records examples of ..." This sentence is unnecessary, confusing, incorrect English and should simply be deleted.  (THey may have meant "Recorded examples of" ... but the rest of the sentence is still incomplete and unnecessary.

 

157 "Evolutionized" --> Evolved

Also ... what do they mean a "shock wave evolved into an infrared wave?".  Are they trying to say that near the volcano the wave was measured as a pressure disturbance which was too small to be seen at greater distance and so was tracked by infrared sensors because it was slightly warmer than the ambient air? I know my description is longer but at least it is clear ... and I really am not sure what they were trying to say in the first place.

 

184 What is a "bay-like" form?? 

Also ... What is the difference between a "bay-like form of both polarities" and a "bipolar signal form".   I think they are trying to describe Figure 4.  It would be clearer to replace this "bay-like" sentence with: Ash clouds sometimes produced unipolar negative field-changes (see 4a), and sometimes positive field changes (4b). Infrequently, they produced bipolar field changes (like figure 3b and 4c).

 

193 lateralextent --> lateral extent 

 

205, 213, 214 "volcanic globule" and "high frequency" appear enclosed in << >> rather than standard quotations. Is this an anomalous preference of MDPI journals, or are the authors << >> when they should be ""? If this is an idiosyncracy of the paper, please fix it

 

339 Figgure --> Figure

Author Response

Thank you for your review.

??: The third conclusion is that volcanoes emit "conventional" ice/water mechanism lightning like a "dirty thunderstorm".  While the paper provides data in Figures 7 and 8 that supports this conclusion, this third conclusion is already well established in references that the authors already cite.  The only bit that might be new is that WWLN was used to identify the conventional lightning as opposed to faster instruments or networks like GLD360 or home-build slow-antennae.  In fact I think other studies have also use WWLN to see volcanic lightning.  In this case, the third conclusion is not original, and should be phrased as "we confirm the results of others", rather than being a new conclusion.

!!: Your comment has been taken into account, we have changed the conclusion.

??: This paper has some nice results, but they make the reader work too hard to check them.  Figures 3 and Figure 4 are both nice results which are purported to show Ez anomalies caused by a drifting charged ash cloud. I agree that the "anomalies" pointed out look like they could be from a volcanic cloud (they do not really look like thunderstorm anomalies, for example). However the result would be strengthened by some other argument that these variations really do correspond with an ash cloud.  Since the authors have video observations, they could cite in table 2 in how many cases their reported anomalies correspond with visual observations of an ash cloud over the sensor. Another way to convince the reader that these anomalies are from an ash cloud is via wind speed, distance and time of eruption. For example, in figure 3 they also show a sounding indicating upper level wind speeds and directions. The authors would strengthen their case if they would clearly state (near Figure 3) the distance and direction between the volcanic eruption site and each of their sensors.  (They do show a map, and this sort of information is scattered around the paper, but it would be good to put it all in a single paragraph.

For example, in Figure 4 there are purported ash clouds that all seem to arrive about 90 minutes after an eruption.  Why not show that for these three examples (at least) the time of arrival is consistent with upper level wind speeds, distance and direction to the measuring stations?

!!: The fact is that for all cases there is video confirmation. Also, for all cases, there are satellite images confirming the passage of the eruptive cloud above the sensor, which are available at the provided links. Suggestions have been added explicitly indicating the propagation speed and residence time of the eruptive clouds at the observation points.

??: On lines 188-191 is one of the more significant scientific statements in the paper.  They do not support this conclusion that some explosions are magma fragmentation and some are gas phase.  Perhaps experts in this new field already know that magma charges negative (for example) while the lighter particles charge positive (that's just a guess ... I don't know this myself).  If this is the case they should cite references supporting this -- or are they merely speculating.   If they can't support this, these lines should be deleted.

!!: The link to work is given: Van Eaton, A.R., Schneider, D.J., Smith, C.M. et al. Did ice-charging generate volcanic lightning during the 2016–2017 eruption of Bogoslof volcano, Alaska?. Bull Volcanol 82, 24 (2020). https://doi.org/10.1007/s00445-019-1350-5

??: Throughout the paper, the authors use the term "electromagnetic pulses".  There are lots of complexities in lightning waveforms, including pulses at frequencies from 10~Hz to 1~GHz (though I think WWLN only works to about 20 kHz).  The general term used to refer to them is "sferics".  I was going to suggest the authors replace every use of "EP" with the word sferic, or sferics.  On the other hand, it seems that they mean something in particular by “electromagnetic pulse” because they count them.  See my next comment.

Figure 7ab and Figure 8ab show EP activity vs. time.  The figures makes sense in terms of prior publications. Some of them (e.g. 7a1, 7a4 and 7a5) even seem to show an initial compact "vent discharge" period followed by a broader convective "plume discharge" peak. However, at no point in the paper do the authors define an algorithm for EP pulses. This is very instrument dependent.  Perhaps it is a standard WWLN product.  Regardless, this algorithm either needs to be explained in the Instrumentation section or a citation to a methods paper for WWLN (that defines “pulses”) be provided.

 

!!: WWLLN only records high current discharges. To construct Figures 7a and 8a, b, the VLF radio direction finder developed at IKIR FEB RAS was used. This device is described in detail in Druzhin G.I., Pukhov V.M., Sannikov D.V., Malkin E.I. .. VLF lightning direction finder. Bulletin of KRAESC. Physical and Mathematical Sciences, 2019, Vol. 2, No 27, 95-104.

 

??: Around line 74 the authors talks about EF-4 field mills in the instrumentation section. As an experimentalist myself, I support bragging about ones equipment, but I request more detail.  A motor part number or specification might be helpful (as appropriate motors for field mills can be difficult to find).

Also, they mention "low electrical energy consumption".  This is a useless statement if they do not specify the consumption.  Do they mean one watt?  100 milliWatts, why not quote voltage and current? [I must state that I reviewed this paper on a long airplane flight, and so I did not have internet access to check the reference on the EF4.]

Also -- they should specify that the EF-4 is only the analog section, the digital section appears to be an E-24 (and they don't specify the power consumption there.  So ... if going to talk about the EF-4 – then provide more detail, or point at reference 4 sooner if that has all the detail).

 

!!: We decided to exclude a detailed description of the measurement equipment. But I will answer your questions.
Since 90% of consumption is mechanics, the total power will be determined mainly by the electric motor. Therefore, the Maxson Flat 32 precision engine with a maximum power of 6 watts was selected. The lightweight design of the mechanical part allows the engine to work almost without load, which in turn reduces the power consumption to ~ 3 W with a supply voltage of 10-14 V.

??: Finally, on line 82, they refer to a "Thiner board".  I have never heard of a "Thiner board".  Do they simply mean "motherboard?".  I ask the authors to recheck reference 4 and see if this is an error.

!!: Thiner Board that is a microcomputer powered by Asus.

??: 20 During the explosive ---> During explosive

!!: Corrected

??: 22 in the result  --->  as a result

!!: Corrected

??: 26 particles falls near a ---> particles fall near a

!!: Corrected

 

??:  29 Control of electric field ---> Measurement of electric field

32 radio signal direction finding at the frequency of 63 MHz occurring from volcanic ---> radio source locations (centered at 63 MHz) of impulses emitted by volcanic lightning (The authors are referring to a lightning mapping array.  Unlike the crossed coils that they ALSO use, it's not really correct to call it a "direction finder", and their language could be clearer in general.)

39 I am not a volcanologist, but I do not know what the "auto-model" section of a volcanic plume is, and I suspect that they have translated a Russian word incorrectly.  I recommend rephrasing as follows: length of 1-7 km were observed on the auto-model section (Near-vent Lightning); length of 1-7 km. (Near-vent lightning);

40 buoyancy section ---> buoyancy stage (or buoyancy phase)

58 Instrumentation and observation method --> Remove it ... redundant.

59-60  This sentence is very confusing.  I believe the authors want to say this: By measuring AEF Ez near the Northern group of volcanoes, we hope to detect the passage of eruptive clouds over the neighborhood settlements.

74 tow ranges --> two ranges

87 "is intended for recording weak signals" ---> "is NOT intended for recording weak signals"

!!: The introduction has been rewritten.

 

95 agreed cable line ---> dedicated cable

!!: Corrected

 

97 I have never heard of a "correct eight" directivity pattern.  I do believe that the authors are referring to a fairly standard crossed-coil magnetic induction detector that allows one to simultaneously measure two orthogonal components of magnetic field (and thus to, indeed, have a single-site direction finder).  They should probably cite another article (for example there are articles about the National Lightning Detection Network in the USA that explain the crossed coil method ... and I think these authors used similar techniques.)

!!: Corrected

 

106 "destroyed volcanic constriction" ... This language is too incorrect to understand ... they must try again ... or remove the phrase entirely.

!!: Corrected

 

107 basis ---> base

!!: Corrected

 

139 "Records examples of ..." This sentence is unnecessary, confusing, incorrect English and should simply be deleted.  (THey may have meant "Recorded examples of" ... but the rest of the sentence is still incomplete and unnecessary.

!!: Corrected

 

157 "Evolutionized" --> Evolved

Also ... what do they mean a "shock wave evolved into an infrared wave?".  Are they trying to say that near the volcano the wave was measured as a pressure disturbance which was too small to be seen at greater distance and so was tracked by infrared sensors because it was slightly warmer than the ambient air? I know my description is longer but at least it is clear ... and I really am not sure what they were trying to say in the first place.

!!: Translation error, not infrared, correct infrasonic.

 

 

184 What is a "bay-like" form?? 

Also ... What is the difference between a "bay-like form of both polarities" and a "bipolar signal form".   I think they are trying to describe Figure 4.  It would be clearer to replace this "bay-like" sentence with: Ash clouds sometimes produced unipolar negative field-changes (see 4a), and sometimes positive field changes (4b). Infrequently, they produced bipolar field changes (like figure 3b and 4c).

!!: Thank you, we agreed with your wording.

 

193 lateralextent --> lateral extent 

!!: Corrected

 

??: 205, 213, 214 "volcanic globule" and "high frequency" appear enclosed in << >> rather than standard quotations. Is this an anomalous preference of MDPI journals, or are the authors << >> when they should be ""? If this is an idiosyncracy of the paper, please fix it

!!: Corrected

 

??: 339 Figgure --> Figure

!!: Corrected

Reviewer 2 Report

The objective of the paper is to describe atmospheric and electric effects associated with volcanic clouds expelled by Kamchatka volcanoes (Shiveluch, Bezimianny, Ebeko). Authors base their study on three geophysical networks like electric field mills (vertical electric field in the atmosphere), seismometers and a VLF station used as goniometer. In addition, they exploit the World Wide Lightning Location Network for the recognition of high energetic lightning discharges during emission of ash clouds.

Authors confirm that multi-phases of electric variations are observed during explosive activity, ash clouds and their displacements due to atmospheric winds. They assume that a first stage is linked to the formation of an eruptive column and a second one to the emergence, development and displacement of electric charges with ash plumes and/or aerosols.  A simulation of such electric signals is then proposed.

The paper is written is a so concise way that it is rather difficult to understand the meaning of the text. The article would gain in clarity and in conviction if several descriptions are detailed and clarified. Sometimes assumptions are done while they are not explained.  In fine, the paper must be clarified and most of paragraphs should be written in a way that they become easily understandable. I suggest that the paper should be improved for a second submission.

References 2, 8, 12, 25 could not be found in the core of the paper.

What is the difference of information given by the field mills and the WWLLN network (except for the distance and the amplitude of the signal)?  The field mills should also records regional lightning?

Where are the WWLLN stations located?

 

General comments:

1. Introduction

L30. “In near-ground layer”. It simply means that the device is set at the ground level?

L32. “results of radio signal direction finding”. Could it be clarified? What was the recording device and how was it used?

L32-53. These paragraphs seem to refer only to records of the electric signals recorded by the WWLLN network. If so, it has to be introduced at the beginning.

L36-42. This summary would be much more efficient with a companion graph. “lightning channel” should be defined for the benefit of any reader.

L43. The meaning of “volcanic lightning” used by the authors is not defined and may generate misunderstanding. It has to be defined.

L45-46. It seems quite “normal” that EP occurrence is linked to the eruption process. But this sentence is non informative.  Moreover, do the two stages of EP records away of a volcano correspond to the passage of ashes, volcanic plumes or aerosols over a field mill, a WWLLN station?

L52-53. Please inform on the records which will be used: electric mills, WWLLN network?

 

2. Results

L58. Remove “Instrumentation and observation method”.

L60-63. It should be placed at the beginning of the paragraph. Figure 1a should be called.

L63-65. When Ebeko volcano started to be active? It is different from the beginning of the records.

It is unclear if (1) AEF Ez was measured with one field mill and if, in total, 3 field mills were installed.

L68. Which sites, those on Ebeko or all of them (Ebeko, KLY, KZY, KBG)?

L71-72. How infrared measurements were done? Explain their use in the paper?

L73. Detail the observations made by satellites.

L74. Please clarify EF 4-records (two and not tow).

L75-78. Not very useful for the reader as it is written.

L79-80. It is obscure for the reader who cannot be convinced at all on the procedure.

What do the authors mean by “adjust all the complex nodes”? E-24 is a 24 bits converter?

L87-88. Unclear. Please rewrite.

L90-92. Do the authors suggest that VLF direction finder Karymshina is similar to a station of the WWLLN station?  Please indicate the station of the figure. Coordinates are useless in the paper.

A VLF direction finder is not yet defined; so, indicate that you will do it “just below”.

L93-95. It is unclear. Are multi-turn loops two orthogonal and vertical loops for recording horizontal components of the electric field, while the whip antenna records the vertical field?

L95. “agreed” has no meaning.

L96-97. Not informative.

L97-98. Very unclear. Rewrite.

L98-99. It should be much better and clearer to write how authors operate.

L101-102. Title should be in bold and small characters.

L106. Scales should be indicated in figure 1.

L107. What does “constriction” means?

L109 and 114. Homogenise the writing: km.

L111-112. The type of explosive activity should be indicated and would be more useful that the general sentence.

 L18-126. For this paper the type of explosive activity is of the uppermost importance. So, the eruptions which are used for the description of the electric field records must be described, as well as the description of ash emissions.

L127-128. Title should be in bold and small characters.

L129-132. Too vague. Please clarify the idea. Indicate the exact number of field mills which were installed and used.

L134-135. Not understandable.

Note: In the following paragraphs, most of the Internet links are not working or they are not reaching the right page. Some are in Russian.

L139-140. This sentence should not be written as it is, because some signals seem to appear before the ashes fly over the recording station. If it not true, please point out this observation.

L140-141. This time laps should be indicated in the summary!

L142. Table 1 is wrongly formatted. Please check and correct.

L144-145. It is badly written. The eruption occurred at 16:26. And later, stations KLY and KZY recorded a signal.

L148. What is Klyuchi? The name of the eruptive or ash transfer event?

L150. Sentence is unclear. Could the authors give the direction of the displacement, it would help the reader.

L152-153. Table. Polarity is not defined. The time arrival of the electric signal is not mentioned. It should be important.

L155. Do the “explosive earthquake” at SMK indicate the beginning of the seismic crisis which will be followed by the volcanic activity (other EQs, ash explosions, etc) or is it corresponding to the exact time occurrence of the formation of the first ash column? 10 mn is not corresponding to an earthquake. The meaning is most probably the time and the 10 mn duration of the first expel of ash column?

L157-158. It would have been interesting to show the time propagation of the air shock.

L162-163. Please report KLY and KZY on Figure 1b.

L165-166. The sentence is vague. Indicate the direction.

L167-168. Not understandable. Why ash large particles should be negatively charged?

L171. How q was computed? What are the parameters?

L173. June 14, not 4. Signal arrived at KZY 4 hours after reaching KLY. What were the wind direction and the speed?

L174. Explain “anomaly back edge front”.

L175. Could the authors explain carefully why some parts of ash clouds (if so) are negatively charges while they are positively charges later (observation at KZY)? This result should be supported by references, laboratory experiments, or ...

L179. Ebeko, not Ebeco.

L182-183. These cases are very important. Do the authors could write a detail paragraph on that point? Do they have an interpretation? What were the parameters for these cases?

L188-191. Please indicate that it is the subject of the next paragraph!

Or explain why negative, positive and dipolar electric field anomalies are observed?

L193.Lateral extent.

L194. Thickness in height.

L207. Indicate that authors refer to figure 5. 0 in time should be indicated in the horizontal scale. Scale should be every hour.

L207-208. Badly written. The eruption started at 10:17, not 10:18. And the first arrival time of an electric signal is at 10:45, 28 minutes later.

L212. 12:03 and 13:10. Refer to points 3 and 5 in figure 5.

L214. How 78 minutes are found? Not clear; 3 to 6? 3 to 4? ...

L215. The signal was ...

L219-220. How the authors can know the volcanic plume structure? “VG of different” time durations (not size).

L221-222. “induced by it”. What does it means? Sentence is very unclear. Please rewrite.

L223. It is not true from figure 5. Signal appears at point 1; 28 mn later.

L223-225. Not clear rewrite and introduce time markers.

L225-227. Unclear. Rewrite. How a turbulent low-power driblet outflow can be defined and proved and seen in the observations? Etc.

L227-230. It is an interpretation. What is the basis of this interpretation?

L231. Not clear.

L233. What is the meaning of transition to EC globular structure ...?

L235. Reference coordinates is not defined. x, y, z would be the distances from the volcano crater to the observation point.

L236. Is the height thickness of the plume is included in Dg?

L240. Not clear: “VG minimum offset from the observation site”. Should be at?

Y should be equal to 0, because perpendicular to the axis motion.

Figure 6. How is computed and drawn the optimized approximation?

L243. What is “it”?

L243-244. Very unclear.

L251. Define “the base” and “globule double height”. To what is it corresponding?

L252. Why Rmin takes into account y, which was supposed before equal to 0. Authors consider the x-axis in the propagation of the ash cloud/aerosol and the AEF field? V is in the direction of the propagation, x. z has not been clearly defined yet.

L253+254. Not understandable (relatively to the observation  ...).

Equation 4: Why  Ez2πεo ? Not explained.

L256. Not understandable.

L259. Could the authors explain “with respect to parameter”?

L261-264. Authors are assuming that the level of charges Q is directly linked to the amount of ash. Is it sure? It could depend of the density of ash in the cloud; if ashes are dispersed, Q could be lower, for instance. It would also depend on the meteorological conditions in the cloud (it was written before in the paper).

L266-267. Format.

L270. Link is not valid.

L271. It is not sure that the emission of ash clouds, ash plumes and aerosols canbe assimilated to magma fragmentation process which is much appropriate to what happens in a magma reservoir.

L273. Do the authors call explosive earthquake the first earthquake which translates the emission of ashes? Generally earthquakes are rock rupture at depth which is different from these “particular earthquakes” which are nearby the surface of a volcano which breaks materials nearby the surface.

L277-278. Please explain “radiation source in THIS direction”.

L279. The way to write  EP activity - N (pulse/min) is not clear.

Table 3. What are the ground-based sites?

L284. Thunderstorm is not the appropriate term.

L289. Writing is unclear.

Could this paragraph related to a figure?

L290-291. How is it known that is related to thermal and turbulent stream formation? How is it specific to other phases of ash emissions?

L295-299. It would be fine to present some graphs. Figure 7a is not called.

Legend of Figure 7. “Azimuth of 25.6 ...” is unclear. “and the squares ...” Are they red rectangulars?

L301-303. Sentence is unclear. Where is recorded the seismic signal? Etc.

L304-305. EP should propagate in any direction. But the main stream should be in the direction of the displacement of the ashes in the atmosphere. Is it the meaning?

Where is the corresponding table?

Figure 8. Define clearly the azimuth! 16:05 and 17.11 should be indicated in (a) and (b).

L317. It is not shown in the paper.

L319. It is not clear. Authors are able to compute an approximate value of the charge, no more. But, it is already a good result.

L318-322. Concerning the eruptive cloud motion, it is still not clear. To assume such result, authors should have installed stations all around the volcano and compare the results (those which are presented in this paper for a few particular stations). The main result of the study is to be able to estimate the charge content in the dispersed clouds transported by atmospheric winds.

L332-335. Paragraph would be rewritten, because it is unclear.

Author Response

Thank you for your review.

 

??: What is the difference of information given by the field mills and the WWLLN network (except for the distance and the amplitude of the signal)? The field mills should also records regional lightning?

!!: Field mills record electric field potential gradient values with the sampling rate of 1 kHz followed by the averaging up to 1 Hz. Thus, fast processes of lightning stroke cannot be recorded.

 

??: Where are the WWLLN stations located?

!!: Web site link: https://wwlln.net/

 

??: L30. “In near-ground layer”. It simply means that the device is set at the ground level?

!!: The device is mounted at the height from 1 m to 5 m from the ground level.

 

??: L32. “results of radio signal direction finding”. Could it be clarified? What was the recording device and how was it used?

!!: Direction finding is location of lightning strokes in space. The paper by R. J. Thomas et. al. “Electrical Activity During the 2006 Mount St. Augustine Volcanic Eruptions” does not describe the sensor construction, it describes only lightning stroke location method.

 

??:L32-53. These paragraphs seem to refer only to records of the electric signals recorded by the WWLLN network. If so, it has to be introduced at the beginning.

!!: We changed the Introduction.

 

??:L36-42. This summary would be much more efficient with a companion graph. “lightning channel” should be defined for the benefit of any reader.

!!: Lightning channel is understood as a path along which lightning main discharge current flows.

 

??: L43. The meaning of “volcanic lightning” used by the authors is not defined and may generate misunderstanding. It has to be defined.

!!:Volcanic lightning refers to any electromagnetic dischrges eminating from active caldera or from or within the volcanic plume. These lightning events can look like notmal cloud lightning, but are related not to weather phenomena, but to the dust, tephra, water vapor and other trace gasses emitted by the volcano.

 

??: L45-46. It seems quite “normal” that EP occurrence is linked to the eruption process. But this sentence is non informative. Moreover, do the two stages of EP records away of a volcano correspond to the passage of ashes, volcanic plumes or aerosols over a field mill, a WWLLN station?

!!: Here we mean that the intensity of strokes during the first stage corresponds to volcano eruption intensity. Intensity of the second stage of volcanic lightning correlates with the total volume of erupted material.

 

??:L52-53. Please inform on the records which will be used: electric mills, WWLLN network?

!!: Corrected

 

??:L58. Remove “Instrumentation and observation method”.

!!: Corrected

 

??:L60-63. It should be placed at the beginning of the paragraph. Figure 1a should be called.

!!: Corrected

 

??:L63-65. When Ebeko volcano started to be active? It is different from the beginning of the records.

!!: Ebeko volcano became active in October 2018 and from that time AEF E_z was continuously recorded at Severo-Kuril’sk (SKR), 7 km from the volcano crater.

 

??: L68. It is unclear if (1) AEF Ez was measured with one field mill and if, in total, 3 field mills were installed. Which sites, those on Ebeko or all of them (Ebeko, KLY, KZY, KBG)?

!!: It is assumed that field mills are installed at all the sites denoted by ^о in Fig. 1а. A VLF direction finder and WWLLN station are installed at Paratunka.

 

??: L71-72. How infrared measurements were done? Explain their use in the paper?

!!: There was a mistake during the translation. There should be not infrared but infrasound measurements which are carried out by microbarometers installed at the sites. Microbarometers are necessary to record shock waves occurring during eruptions.

 

??:L73. Detail the observations made by satellites.

!!: Satellite images are necessary just to confirm the presence of volcanic activity, that is why we think that it is not necessary to describe this type of observations.

 

??:L74. Please clarify EF 4-records (two and not tow).

!!: EF-4 device is operating in two measurement ranges, thus we mentioned two.

 

??:L75-78. Not very useful for the reader as it is written.

!!: This paragraph characterizes EF-4 unit as a small and autonomous and explains why just it was chosen to organize the network.

 

??:L79-80. It is obscure for the reader who cannot be convinced at all on the procedure.

!!: The paragraph was changes. But it is necessary to show the operativity of data acquisition.

 

??:What do the authors mean by “adjust all the complex nodes”? E-24 is a 24 bits converter?

!!: The name of the unit is E-24, it was produced by L-Card.

 

??:L87-88. Unclear. Please rewrite.

!!: Was rewritten.

 

??:L90-92. Do the authors suggest that VLF direction finder Karymshina is similar to a station of the WWLLN station? Please indicate the station of the figure. Coordinates are useless in the paper.

!!: The VLF direction finder is described more in detail. In this map scale, the direction finder location will coincide with Paratunka marker PRT.

 

??: L93-95. It is unclear. Are multi-turn loops two orthogonal and vertical loops for recording horizontal components of the electric field, while the whip antenna records the vertical field?

!!: Changed the paragraph.

 

??: L95. “agreed” has no meaning.

!!: Agreed cable line allows us to send a signal without distortions that makes it possible to recover the electromagnetic field pattern. That is why agreed makes sense.

 

??: L96-97. Not informative.

!!: As long as the device is not an industrial unit, a reader should have some notion about its structure.

 

??:L97-98. Very unclear. Rewrite.

!!: Was rewritten.

 

??:L98-99. It should be much better and clearer to write how authors operate.

!!: The device and the technique of application was developed by the authors and are not widely known that is why a reader should know how the equipment is working.

 

??: L101-102. Title should be in bold and small characters.

!!: Corrected

 

??: L106. Scales should be indicated in figure 1.

!!: Fig. 1а is schematic as it is written in the caption. Fig. 1 b contains a coordinate grid as long as the figure is a map.

 

??:L107. What does “constriction” means?

!!: Misprint, changed to “construction”

 

??:L109 and 114. Homogenise the writing: km.

!!: Corrected

 

??:L111-112. The type of explosive activity should be indicated and would be more useful that the general sentence.

!!: The paragraph was changed.

 

??: L18-126. For this paper the type of explosive activity is of the uppermost importance. So, the eruptions which are used for the description of the electric field records must be described, as well as the description of ash emissions.

!!: The paragraph was changed.

 

??:L127-128. Title should be in bold and small characters.

!!: Corrected

 

??: L129-132. Too vague. Please clarify the idea. Indicate the exact number of field mills which were installed and used.

!!: The exact number of electrostatic sensors is mentioned in Section 2.1. L129-132 was changed.

 

??:L134-135. Not understandable.

!!: L134-138 were removed from this section of the paper as long as this paragraph refers to the observation method and was moved to the section 2.1. Instrumentation and observation method.

 

??: Note: In the following paragraphs, most of the Internet links are not working or they are not reaching the right page. Some are in Russian.

!!: We corrected the inactive references. The KB GS RAS site has only Russian version. It is impossible to remove this link as we use the data available on this site.

 

??:L139-140. This sentence should not be written as it is, because some signals seem to appear before the ashes fly over the recording station. If it not true, please point out this observation.

!!: Inaccurate translation, this sentence is an introductory for the following paragraph.

 

??:L140-141. This time laps should be indicated in the summary!

!!: Here we indicate the whole period of observations, Table 1 shows the exact dates for the events.

 

??:L142. Table 1 is wrongly formatted. Please check and correct.

!!: In the third case, deviation is bipolar, that is why both experimental values are shown. Horizontal borders were added into the Table.

 

??:L144-145. It is badly written. The eruption occurred at 16:26. And later, stations KLY and KZY recorded a signal.

!!: The paragraph was changed.

 

??:L148. What is Klyuchi? The name of the eruptive or ash transfer event?

!!: We ment KLY site located in Klyuchi village.

 

??:L150. Sentence is unclear. Could the authors give the direction of the displacement, it would help the reader.

!!: Direction was given.

 

??: L152-153. Table. Polarity is not defined. The time arrival of the electric signal is not mentioned. It should be important.

!!: Polarity was indicated by a sign in the column polarity (+,-, ±), and the arrival time is shown in Fig. 3.

 

??:L155. Do the “explosive earthquake” at SMK indicate the beginning of the seismic crisis which will be followed by the volcanic activity (other EQs, ash explosions, etc) or is it corresponding to the exact time occurrence of the formation of the first ash column? 10 mn is not corresponding to an earthquake. The meaning is most probably the time and the 10 mn duration of the first expel of ash column?

!!: Explosive earthquake accompanies ash-gas flow ejection into the atmosphere.

 

??: L157-158. It would have been interesting to show the time propagation of the air shock.

!!: Table 2 shows the characteristics of the shock wave recorded at KLY. Fig. 7b indicates the times of shock wave arrival to KLY.

 

??:L162-163. Please report KLY and KZY on Figure 1b.

!!: The figure was changed.

 

??:L165-166. The sentence is vague. Indicate the direction.

!!: The direction is summed from redial spreading and drift under wind effect. The authors think that indication of the direction is not necessary as long as there are data from balloon sounding and consecutive satellite images from which the direction is clear.

 

??: L167-168. Not understandable. Why ash large particles should be negatively charged?

!!: When particles of different sizes but similar density are settling in a mixture, first large fraction settle out, than smaller ones. The particles with the sizes of less than 10 При осаждении из взвеси частиц, разных размеров но одинаковой плотности сначала выпадает крупная фракция затем более мелкая. Частицы же с размером менее 10 µm may stay in the atmosphere for ages. When the EC came to KLY, a negative anomaly was recorded at the station, that is why we made such a conclusion.

 

??:L171. How q was computed? What are the parameters?

!!: It was calculated in the paper 5. A more detailed model of the calculation, different from that in paper 5, was described in the subsection Fine aero-electric structure of an eruptive cloud.

 

??:L173. June 14, not 4. Signal arrived at KZY 4 hours after reaching KLY. What were the wind direction and the speed?

!!: Indeed, it should be June 14. The wind direction and velocity are taken from balloon sounding data illustrated in Fig. Fig. 2g. Fig. 3 shows the time of beginning of AEF E_z anomaly at KLY and KZY, at 17:40 and 21:33. The difference between the beginning of the eruption and the time of anomaly beginning is 1 hour and 14 minutes and 5 hours and 7 minutes, respectively.

 

??:L174. Explain “anomaly back edge front”.

!!: We changed it. Based on the shape of the trailing edge of the anomaly, it can be assumed that it could arise due to the passage of a cloud with a positive charge.

 

??:L175. Could the authors explain carefully why some parts of ash clouds (if so) are negatively charges while they are positively charges later (observation at KZY)? This result should be supported by references, laboratory experiments, or ...

!!: Based on the form of back front of the anomaly at KZY with Ez.max.≈ 1.0 kV/m (Fig. 3 с) we can assume that it might be generated by the passage of a cloud with positive charge [Cherneva et al., 2007]. Evidently, when propagating at long distances during eolian differentiation, EC develops into an aerosol cloud in which the main components are volcanic gases (up to 96-98 % H2O) with insignificant content of very fine ash. At this stage of EC evolution, positive total charge, apparently, prevails in its aero-electric structure.

 

??:L179. Ebeko, not Ebeco.

!!: Corrected.

 

??:L182-183. These cases are very important. Do the authors could write a detail paragraph on that point? Do they have an interpretation? What were the parameters for these cases?

!!: During the indicated period two events were recorded when an eruptive cloud passed over the SKR observation site and close to it but response in AEF E_Z was not observed. Statistics of such events is very small. We offer to remove L182-183 from the manuscript.

 

??:L188-191. Please indicate that it is the subject of the next paragraph!

!!: We’ve indicated it is in the next paragraph.

 

??:L193.Lateral extent.

!!: Corrected.

 

??:L194. Thickness in height.

!!: Corrected.

 

??:L207. Indicate that authors refer to figure 5. 0 in time should be indicated in the horizontal scale. Scale should be every hour.

!!: Corrected.

 

??:L207-208. Badly written. The eruption started at 10:17, not 10:18. And the first arrival time of an electric signal is at 10:45, 28 minutes later.

!!: The beginning of the eruption with the accuracy up to seconds is 10:17:55, that is almost 10:18 (the beginning of the eruption was indicated in Fig. 7). Yes, the first response occurred 28 minutes later. The cloud was only developing. In other words the cloud did not reach the station yet but there as a response. We made to such a conclusion as long as this time period was characterized by fair weather conditions.

 

??:L212. 12:03 and 13:10. Refer to points 3 and 5 in figure 5.

!!: Corrected.

 

??:L214. How 78 minutes are found? Not clear; 3 to 6? 3 to 4? ...

!!: Anomaly beginning corresponds to moment 3, moment 1 (10:45) is 78 minutes before it. It was indicated in the text.

 

??:L215. The signal was ...

!!: Corrected.

 

??: L219-220. How the authors can know the volcanic plume structure? “VG of different” time durations (not size).

!!: See many articles on this subject including

_{Bulletin of Volcanology (2020) 82: 24 https://doi.org/10.1007/s00445-019-1350-5}

_Or

https://www.sciencedirect.com/science/article/pii/S0377027318304591?via%3Dihub

or

https://agupubs.onlinelibrary.wiley.com/doi/full/10.1002/2016GL068076

 

 

??:L223. It is not true from figure 5. Signal appears at point 1; 28 mn later.

??:L223-225. Not clear rewrite and introduce time markers.

??:L225-227. Unclear. Rewrite. How a turbulent low-power driblet outflow can be defined and proved and seen in the observations? Etc.

??:L227-230. It is an interpretation. What is the basis of this interpretation?

!!: The authors formulated the paragraph incorrectly (L219-L230). Volcanic globule recording duration is bounded to the seismic signal but the delay during propagation. The paragraph was changes and we added the description of atmosphere parameters, wind velocity and direction in the zone of eruptive cloud propagation.

 

??:L231. Not clear.

!!: Changed.

 

??:L233. What is the meaning of transition to EC globular structure ...?

!!: Point charge has infinitely small volume and, thus, charge density is the infinitely large that never happens. In this case field intensity near the source is infinitely great that is senseless physically. That is why, it is usually assumed that the charge is distributed evenly in some compact limited volume close to a spherical form. In this case, expression (1) is fair if field measurement point is out of the region of charge location that usually corresponds to experiment conditions. In case of eruptive clouds such assumption is reasonable as every such a cloud is generated by gas babble in magma before its eruption. During the following transformation of such a charge into generated vertexes, each vertex structure can also be considered as a region of increased pyroclastic density with increased charge density.

 

??:L235. Reference coordinates is not defined. x, y, z would be the distances from the volcano crater to the observation point.

!!: We indtroduced a model coordinate system at the beginning of which field detector is located. The model coordinate system can be related to a physical system after introduction of additional coupling.

 

??:L236. Is the height thickness of the plume is included in Dg?

!!: Plume parameters may be associated with model parameters by introduction of additional coupling. At this time such coupling was not introduced.

 

??:L240. Not clear: “VG minimum offset from the observation site”. Should be at?

!!: Paragraph L238-L241 was changed.

 

??:Y should be equal to 0, because perpendicular to the axis motion.

!!: Y=0 only in rare cases when the trajectory passes exactly over an observation site.

 

??:Figure 6. How is computed and drawn the optimized approximation?

!! The figure was moved to line 255, in that paragraph the method for estimation of optimized approximation is described briefly.

 

??:L243. What is “it”?

??:L243-244. Very unclear.

!!: We changed it.

 

??:L251. Define “the base” and “globule double height”. To what is it corresponding?

!!: It is really unclear. The correct version is:

In this case, the VG field is approximated by the field of a symmetric dipole with the same charge but with the separation equal to the globule double height.

 

??:L252. Why Rmin takes into account y, which was supposed before equal to 0. Authors consider the x-axis in the propagation of the ash cloud/aerosol and the AEF field? V is in the direction of the propagation, x. z has not been clearly defined yet.

!!: Because in general case, the cloud does not pass over the observation site, please, find in previous answers.

 

??:L253+254. Not understandable (relatively to the observation ...).

!!: We changed it.

 

??:Equation 4: Why Ez2πεo ? Not explained.

!!: When typing the formulas, we made some mistakes. It was corrected.

 

??:L256. Not understandable.

!!: We changed it.

 

??:L259. Could the authors explain “with respect to parameter”?

!!: Parameter gamma is missing.

 

??:L261-264. Authors are assuming that the level of charges Q is directly linked to the amount of ash. Is it sure? It could depend of the density of ash in the cloud; if ashes are dispersed, Q could be lower, for instance. It would also depend on the meteorological conditions in the cloud (it was written before in the paper).

!!: This conclusion was confirmed by direct optical observations at KLY station by the rate of ash cloud dissipation.

 

??: L266-267. Format.

!!: Corrected.

 

??:L270. Link is not valid.

!!: Corrected.

 

 

L271. It is not sure that the emission of ash clouds, ash plumes and aerosols can be assimilated to magma fragmentation process which is much appropriate to what happens in a magma reservoir.

!!: Current research questions the significance of plume charge separation caused solely by fracturing tephra and triboelectric charging of dust. Actually most important plume lightning may come from ice formed from the huge volume of water vapor expelled by the volcano. See https://doi.org/10.1007/s00445-019-1350-5

 

L273. Do the authors call explosive earthquake the first earthquake which translates the emission of ashes? Generally earthquakes are rock rupture at depth which is different from these “particular earthquakes” which are nearby the surface of a volcano which breaks materials nearby the surface.

!!: The paper makes the reference to the paper (Zobin V.M. Volcano Seismology: An Introduction. In Encyclopedia of Complexity and Systems Science.; Meyers R. (eds); Springer: Berlin, 2017; 559 p.) which describes the model which relates high-frequency seismoacoustic radiation and magma body rapture.

 

??:L277-278. Please explain “radiation source in THIS direction”.

!!: The paragraph was changed.

 

L279. The way to write EP activity - N (pulse/min) is not clear.

!!: Changed «EP activity» with «EP counting rate»

 

??:Table 3. What are the ground-based sites?

!!: Corrected.

 

??:L284. Thunderstorm is not the appropriate term.

!!: You are right, the term «electrical storm» is better.

 

L289. Writing is unclear.

!!: Changed.

 

??:Could this paragraph related to a figure?

!!: This paragraph is related to Fig. 7.

 

L290-291. How is it known that is related to thermal and turbulent stream formation? How is it specific to other phases of ash emissions?

!!: This conclusion was grounded.

 

L295-299. It would be fine to present some graphs. Figure 7a is not called.

!!: We made references to Fig. 7.

 

??Legend of Figure 7. “Azimuth of 25.6 ...” is unclear. “and the squares ...” Are they red rectangulars?

!!: Corrected.

 

??:L301-303. Sentence is unclear. Where is recorded the seismic signal? Etc.

!!: The paragraph was changed.

 

L304-305. EP should propagate in any direction. But the main stream should be in the direction of the displacement of the ashes in the atmosphere. Is it the meaning?

!!: The sense is that EMR flux was recorded by the VLF direction finder from the direction to Bezymyanniy volcano.

 

??: Where is the corresponding table?

!!: We added a Table with the characteristics of Bezymyanniy eruptions.

 

??: Figure 8. Define clearly the azimuth! 16:05 and 17.11 should be indicated in (a) and (b).

!!: Azimuth to Bezymyanniy, Siveluch and Kluchevskoy volcanoes coincide with the accuracy to tens and is 25,6^о. Figures show the time..

 

??: L317. It is not shown in the paper.

!!: The paragraph was changed.

 

L319. It is not clear. Authors are able to compute an approximate value of the charge, no more. But, it is already a good result.

!!: Solution of a reverse problem considered in section “Fine aero-electric structure of an eruptive cloud.” will allow us to recover the trajectory of motion and spatial location of a cloud.

 

L318-322. Concerning the eruptive cloud motion, it is still not clear. To assume such result, authors should have installed stations all around the volcano and compare the results (those which are presented in this paper for a few particular stations). The main result of the study is to be able to estimate the charge content in the dispersed clouds transported by atmospheric winds.

!!: Solving the inverse problem posed in the "Fine aero-electric structure of an eruptive cloud" using data obtained from two stations (KLY and KZY), we can determine the velocity vector and the trajectory.

 

??: L332-335. Paragraph would be rewritten, because it is unclear.

!!: We removed that paragraph.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

This second version is a much more appropriate and improved version.

Most of the comments have been taken into account, and any reader can now understand the content and appreciate the the results.

It remains one point to solve. Equations 2 and 3 should be compatible.

 

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

Review of Registration of atmospheric-electric effects from volcanic clouds on the Kamchatka Peninsula (Russia), by Firstov et al., submitted to Atmosphere

This study documents electrified volcanic plumes from several eruptions of Shiveluch, Bezymianny, and Ebeko volcanoes ranging from 2014­–2019. The data originate from three different types of sensors. #1 is a network of (seven?) electrostatic fluxmeters—also known as field mills—set up around Shiveluch volcano. These sensors measure the vertical electric field in the atmosphere, and can provide an indication of charge buildup from electrified volcanic clouds. #2 is a global network of long-range sensors from the World Wide Lightning Location Network, which can detect high-current volcanic lightning. And #3 is a single, VLF direction finder, which detects electromagnetic pulses at close range.

From the time-series observations and discussion presented, it is clear that the electrostatic fluxmeters and VLF direction finders are capable of detecting electrical activity associated with the volcanic explosions, which assists volcano monitoring by confirming the presence of hazardous ash clouds in the atmosphere. That alone is a valuable observation. However, it is not clear how the observations and models improve response efforts by, for example, helping with earlier or more definitive detection, or characterizing the severity of the volcanic hazards. Several sections allude to the ability to distinguish between ash-rich and ash-poor volcanic clouds (Lns. 163, 262), or stages of eruption development (lns. 8-10) but details are not adequately supported. The study would be greatly improved by providing a deeper analysis of these very interesting electrical observations. Some suggestions below:

Incorporate plume heights into the analysis where they are available. The studies of Arason et al. (2011) and Van Eaton et al. (2020) showed that moderate-sized volcanic plumes in Iceland and Alaska only produced detectable lightning once they rose above the local –20oC freezing level of the atmosphere. Is this also the case for the Kamchatkan eruptions studied? Could it partly explain why some of the eruptions produced WWLLN-detected lightning and others did not?

 

Consider the time-delay of electrical activity compared to the start time of the eruptions, where possible (perhaps added as a table). Did the fluxmeters detect an electric field at exactly the moment of eruption (based on webcam or other tool), or was there a time delay of minutes to tens of minutes? How do these time delays compare with WWLLN lightning detections?

 

Consider the ash content of the volcanic plumes studied. Some ash fallout data is provided in Table 1, which seems like it could be quite useful. Can this combine with other observations (webcam/satellite) to understand how the electrical activity changes with high vs. low ash content plumes? The study of Nicoll et al. (2019) could provide a useful comparison with an ash-free gas plume (which carried a charge, but did not produce lightning).

Finally, the manuscript structure and clarity of the writing need some careful attention, in addition to the English grammar and word choice. It would greatly improve the readability of the paper if nearly ALL of the acronyms were removed (except station names). Perhaps only 2 or 3 are really necessary – the rest make the text difficult to follow.

Overall, I think this study describes an extraordinary dataset and has a lot of potential. I would strongly support its publication after addressing the scientific comments above.

Other detailed suggestions:

Ln. 1-12. Abstract: Include a few lines carefully outlining the 3 methods of electrical measurements used in the study, and which volcanoes/eruptions/time periods were examined. The ‘stages’ of eruption described in the abstract are not fully described or supported in the main text.

Ln. 6. Use a consistent spelling of Bezymianny and Shiveluch throughout the paper. Also use consistent date/time formats through the paper, figures, and tables.

Ln. 21. Replace with “the heaviest fraction of ashthe largest particlesfallsnear a volcano, and the light fractionsmaller ash particlesand aerosols propagate..”

Ln. 25-26. Replace “finely dispersed ash” with “fine-grained ash” throughout paper

Ln. 31-36. The updated terms for these three classes of discharges are: (1) continual radio frequency impulses (“vent discharges”), (2) near-vent lightning, and (3) plume lightning. Please refer to Fig. 4 of Behnke et al. (2013) for a good overview.

Fig. 1. Text is too small to read in part A. Is Tolbachik included? Provide more detail in the caption regarding the meaning of each symbol used.

Table 1. Why does Table 1 appear in the section “Instrumentation and observation method”? It would be helpful to separate the methods and results sections. It may also be valuable to create one master table describing all of the eruptions examined in this study, each with a unique identifier.  In caption of table 1, specify the ash collection site and distance/direction from source. Is it labeled on the location map in Fig.1?

Ln. 124. Replace “almost round EC” with “a nearly circular umbrella cloud”

Fig. 2. Give locations of cameras (lat/lon/distance from source) and label on Fig.1 if appropriate

Fig. 3. For each of the time-series figures, include the volcano, date/time of eruption, and clearly label the start of eruptive activity on the figure.

Table 1. Specify which instrument(s) and location(s) recorded these data.

Ln. 168. Characterized by significant lateralextent

Ln. 174. Replace “velocity fluctuations of its outflow from the crater” with “changes in the mass flux at source”

Pg. 8. Replace “lava streams” with “lava flows” throughout paper

Ln. 209-220. Moving this methods section earlier in the paper (combined with first methods section) would give the paper a more logical structure and flow.

Ln. 219. Please specify that this VLF direction finder is (presumably) separate from the WWLLN sensor network. Can you provide more details on this type of sensor?

 Ln. 240-259. This section discusses different stages of electrical activity, but is missing some critical detail. How do the electromagnetic pulses change across the stages of activity? Is it consistent across different eruptions? And what are some possible reasons for that change? Is there any observational data to support a change in the dynamics of the eruption plume?

Ln. 252. “cases 1 and 3” – as mentioned earlier, it would help to create unique identifiers for each eruption. There are several “cases 1 and 3”

Fig. 7. Specify the volcano/time/date of each case in the time series

Fig. 8. Align parts A +C and B+D on top of each other so the graphs are easier to read and link to the correct eruption

Ln. 262. What is the evidence for some eruption clouds being less ash-rich than others?

Ln. 269-270. Mathematical modeling of the charge field “showed acceptable results” but it was not clear why the modeling was important, or what the implications were for understanding the electrical phenomena.

Lns. 281-287. Is this saying that WWLLN lightning only appeared when the plume reached its maximum height? If so, that is very interesting and deserves further discussion. Would be interesting to compare with Fig. 5 of Van Eaton et al. (2020), showing that one of the Bogoslof eruption plumes did not create detectable lightning until it rose above the freezing level.

 

References

Arason P, Bennett AJ, Burgin LE (2011) Charge mechanism of volcanic lightning revealed during the 2010 eruption of Eyjafjallajökull. J. Geophys. Res. 116:B00C03 doi: 10.1029/2011jb008651

Behnke SA, Thomas RJ, McNutt SR, Schneider DJ, Krehbiel PR, Rison W, Edens HE (2013) Observations of volcanic lightning during the 2009 eruption of Redoubt Volcano. J. Volcanol. Geotherm. Res. 259:214-234 doi: 10.1016/j.jvolgeores.2011.12.010

Nicoll K, Airey M, Cimarelli C, Bennett A, Harrison G, Gaudin D, Aplin K, Koh KL, Knuever M, Marlton G (2019) First In Situ Observations of Gaseous Volcanic Plume Electrification. Geophys. Res. Lett. 46(6):3532-3539 doi: 10.1029/2019gl082211

Van Eaton AR, Schneider DJ, Smith CM, Haney MM, Lyons JJ, Said R, Fee D, Holzworth RH, Mastin LG (in press) Did ice-charging generate volcanic lightning during the 2016–2017 eruption of Bogoslof volcano, Alaska? Bull. Volcanol. doi: 10.1007/s00445-019-1350-5

Author Response

Dear reviewer,

 

I really appreciate your review and the remarks.

 

As for the modeling, we quite agree with you the main idea was not formulated clearly. We changed the last paragraph. I would like to clarify that modeling allows us to estimate the total charge more accurately that, in its turn, correlates with the total amount of ash in an eruptive cloud (256-262) and the ash amount is estimated on the basis of explosive earthquake parameters. According to electromagnetic pulse duration (or their amplitude-frequency portrait) we can judge about volcanic thunderstorm stage or eruption stage development. Atmosphere temperature gradient at the moments of Sheveluch eruptions described in Table 3 almost coincided and is illustrated in Fig. 2g (red). However, WWLLN recorded different number of lightning strokes. In three cases no current strokes were observed. In contrast to the abovesaid, Bezymyanniy eruption supports the conclusions described in the papers which you mentioned. It is likely to be associated with a large amount of water steam in the erupted material. However, it is impossible to confirm this assumption at this stage of investigation.

Fig. 3 illustrates a graph of dependence of AEF Ez on time. It is clear from the figure that the response occurs in the readings only during an eruptive cloud passage. At the same time VLF direction finder shows simultaneous beginning of the eruption and increase of the counting rage from the azimuth corresponding to Shiveluch volcano direction.

Detailed remarks:

We have moved the section describing the instrumentation for electromagnetic radiation observations. Electromagnetic pulses really change during eruption development. We've added a paragraph describing these changes. We have also made other changes according to the remarks.

Thank you for the review.

Author Response File: Author Response.pdf

Reviewer 2 Report

The results introduced seem valid and probably to be new to some extent.

However, some additional explanation is needed in order to tell accept or not.

- In the introduction the explanation for the volcano is abundant, however,  the review of volcano lightning is not sufficient. There are not a few researches on this kind of electrical discharges related to volcano eruption. Strongly suggest to enhance the review of similar researches or similar subject researches in order to clarify the originality and new finings of the present pare.

- In line 139 it is noted that “EZ began to decrease to -6 kV/m, and then sharply increased to +5 kV/m” while in line 100 “The applied electrostatic fluxmeter EF-4 records AEF Ez  in the range up to 4 kV /m”. Are those consistent?

- line 146: Please explain how to estimate “40 C”. Is it written in [4]?

- Figure 3: What is reason of spiky fluctuations at KZY, which are not seen at KLY?

- line 158: Is there any ideas to explain why only bay-like form was recorded?

- Fig 4: Seemingly the fluctuation is not small even for the period of non-eruption. How much is the typical amplitude of this kind?

- Fig 5: I would suggest to point the times of eruptions, namely, 12:04 and 13:10 with arrows.

- Fig 5: How did you confirm “powerful outburst of pyroclastic material” or “vast lava streams”?

- Fig 6: Is this type of result world’s first in volcano research? In that case please mention it explicitly.

- It should be noted explicitly in the text that the some sentences explain Figure 8

(There is no term “Figure 8” in the text.)

- It’s very ambiguous about the new findings in Figure 7 and 8. Pleas mention them explicitly in the text.

- If author confirmed the EP of WWLLN was not caused by thunderstorm(s), please remark it explicitly.Is there any possibility that the eruption induced the development of thunderstorm?

Typos?

- Table 1: DATA > DATE

- line 151: EO might be EC?

- ” It is clear in Figure 5a that «higher-frequency» oscillations with the amplitude of 10V /m“  ——>   Figure 5b?

Author Response

Dear reviewer,

 

I really appreciate your review and the remarks.

 

At the present moment, it is not quite possible to extend the review of our investigations but we promise to do it in detail in the next paper.

When describing electrostatic fluxmeters EF-4, we made an omission. In fact, the devices are capable of operating in two ranges with different sensitivity (within ± 4 kV/m and within ± 7 kV/m).

In the paper [4], charge was estimated within point charge dipole model according to formula 1 contained in our paper. However, such an estimate is very rough that is why we performed the modeling described in this paper.

The cause of electric field oscillations at KZY was the passing eruptive cloud propagating at tropopause height. Based on the satellite data, we can state with confidence that the distance from the cloud to KLY settlement was ~ 30 km that was enough for it not to influence the instrumentation. The lower cloud in this eruption was formed from pyroclastic flows that is why it did not reach tropopause height and was divided with the main cloud. Owing to the lower height it lost its mass and charge more effectively. Thus, it could not have strong impact on the instrumentation located at KZY.

In fair weather conditions, electric field intensity oscillations are small and are determined mainly by diurnal variations.

Powerful pyroclastic flows were confirmed visually and then by an expedition group.

As far as the authors know, approximation of eruptive cloud electric field by the method described in the paper has been carried out for the first time. The modeling was performed in order to estimate the cloud total charge more accurately.

During the described time periods, there was no lightning activity of atmospheric origin over the whole territory of Kamchatka peninsula.

 

Thank you for the review.

Author Response File: Author Response.pdf