General comment

The manuscript presents an interesting study on the paleoseismology of the Fiandaca Fault. The investigation is aimed at extending back to the past the rich dataset of historical earthquake ruptures already documented in the literature since mid-1800s. The manuscript is very long and containes a lot of information not needed for the work; this makes it not easy to be read and somewhat dispersive.

The main issue I see is that the manuscript wants to cover topics not relevant for the target stated in the title. In my opinion the valuable contribute related to the paleoseismological investigation is diminished by the attempt to enlarge their conclusions to the whole Etna domain. This introduces a series of speculations not really supported by the study, which are not significant for the work.

I recommend to focus on the analysed fault emphasizing both the paleoseismological and morphological point of views. To do this, I should avoid other not relevant considerations that confuse the reader.

Reply to General comment: We are grateful to the reviewer for this comment. We will modify the text in order to improve the manuscript. The considerations about the entire Etna domain are just a comparison between our results about the throw rates along the Fiandaca Fault and the other faults with data that have mostly already been published. Anyway, we will follow your suggestion and we will focus on the paleoseismological and morphological characterization of the Fiandaca Fault.

In the following points I indicated some general considerations, whereas in the revised manuscript you will find specific comments.

1. Introduction

In general the authors quote all the references available in the literature; this produces long citations continuously interrupting the text; hard to follow. Many of these references are dated and already cited in more recent papers; please keep only those significant for framing the work (also for other chapters).

Reply to Introduction: Thanks for the suggestion. We will remove some citations.

2. Geological setting

I consider this chapter not really useful for the aims of the work, at least in this form. Many data and information can be synthetized and referred to well-known papers describing in a complete way the geological and seismotectonic features of the Etna region. Also the description of the volcanic activity is useless and looks like a volcanological scholar book. Finally, the description of flank instability and the faults involved in this process can be reduced and aimed at understanding the general geodynamic framework of the area.

Reply to Geological setting: Thanks for the suggestions, we will reduce this chapter, removing part of the volcanic activity.

3 Geologic features around the Fiandaca Fault

This is an important part of the paper useful for interpreting the paleoseismological and morphological analyses. I think that the authors could also advantage of a recent revision of the local geology made by Branca (see in Azzaro et al., 2022). Regarding the paragraph 3.2 (Morphotectonic and historical seismicity), I recommend not to add new fault names compared to the previous literature since this produces uncertainty and misleading in the readers, mainly professionals who work in the area and use the scientific literature.

Reply to Geological features around the Fiandaca Fault: We are grateful to the reviewer for the advice. After a new check of Azzaro et al. (2022) paper we have observed that the geology is the same after Branca et al. (2011) and Branca and Abate (2019). Regarding the fault name we just used the name ‘Fleri’ for the NW-striking segment of the fault, as defined also in the ITHACA database (ITHACA (ITaly HAzard from CApable faulting, A database of active capable faults of the Italian territory, Version December 2019, ISPRA Geological Survey of Italy, http://sgi2.isprambiente.it/ithacaweb/Mappatura.aspx.

5 Trenching site results

This chapter is the core of the manuscript. It contains the essential information to reconstruct the past faulting history but some passages are not clear to me. For example, FIA 1 N-wall1, FIA 1 S-wall2 (F4-F5) show dislocations that can be interpreted ambiguously: typically in volcanic terrains the surface of the lava flows is very rough and is frequently broken by primary large fractures due to cooling and residual movement. This results in uneven horizons. How do the authors exclude this?

Reply to Trenching site result: We thank the reviewer for this comment. As described in the manuscript, the faults F4 and F5 are uncertain (dashed in the figures) because they can be just the rough surface of the lava flow, as also suggested by the reviewer. For this reason, we did not consider them as event horizons and for the restoration.

6 Trench restoration and paleoseismic sequence (1)

At line 401 you affirm that "the highlighted sequence of earthquakes strongly suggests an increase in fault activation over time". I strongly disagree with you. You cannot compare a very detailed historical dataset produced by observers in a densely inhabited area with faulting evidence observed in a trench. Fractures can jump the site explored by the trench so the record of event horizons is not complete.

Reply to Trench restoration and paleoseismic sequence: Yes, it is true, thanks for noticing that. The word ‘strongly’ is not correct in this case, we will change this part in the revised version of the manuscript. However, comparisons between paleoseismic events and historical earthquakes have always been done, also obtaining strong correlations. For example, see Rockwell et al. (2000) regarding a segment of the North Anatolia Fault.

*Rockwell, T., Barka, A., Dawson, T. et al. Paleoseismology of the Gazikoy-Saros segment of the North Anatolia fault, northwestern Turkey: Comparison of the historical and paleoseismic records, implications of regional seismic hazard, and models of earthquake recurrence. Journal of Seismology 5, 433–448 (2001). https://doi.org/10.1023/A:1011435927983.

6 Trench restoration and paleoseismic sequence (2)

Furthermore, in our specific case most of the more recent event represented in Fig. 12 ruptured only minor patches of the Fiandaca Fault, while only the largest ones (1894, 2018) ruptured the whole structure. So in any case you could observe the evidence of this smaller events in the trench. This is a critical point on which you base the idea of the throw rate increase over time.

I consider this assumption too strong for including even other faults in your hypothesis.

Reply to Trench restoration and paleoseismic sequence (2): We thank the reviewer for this comment. Yes, it is would be possible to observe minor events in the trench, as we know by the historical data also the minor events can produce throw of 20 cm, as occurred in 1984 during the earthquake that ruptured the NNW-SSE segment of the Fiandaca fault. Therefore, it is possible that some of the cumulated displacement in the trench may be associated to other events. However, we just assume that the most probably event is the strongest historic one.The most important novelty obtained by the paleoseismic history is the medieval coseismic surface faulting event. Of course, the evidence in the trenches alone is not enough to suggest an increase in the throw rate; however, this paleoseismic evidence coupled with the results of our morphotectonic analysis show this increase. Furthermore, the 1984 throw value suggests that also minor events can contribute to the scarp formation.

7 Throw rates analysis and flank dynamics (1)

Paragraph 7.1 is an important part of the manuscript since it tries to link the results of the trenches with the geological evolution of the fault. I also suggest to consider in the discussion findings coming from the papers by Romagnoli et al. (2021*), Tortorici et al. (2021**) and Azzaro et al. (2022), who investigated the morphological evolution of the Fiandaca Fault vs lava flow coverage.

*Romagnoli, G., Pavano, F., Tortorici, G., & Catalano, S. (2021). The 2018 Mount Etna earthquake (Mw 4.9): Depicting a natural model of a composite fault system from coseismic surface breaks. Tectonics, 40(5), e2020TC006286.

**Tortorici, G., Pavano, F., Romagnoli, G., & Catalano, S. (2021). The effect of recent resurfacing in volcanic areas on the distribution of co-seismic ground deformation due to strike-slip earthquakes: New insights from the 12/26/2018 seismic event at Mt. Etna. Journal of Structural Geology, 145, 104308.

Reply to Throw rates analysis and flank dynamics: We thank the reviewer for this comment. This is an important aspect of the paper, Azzaro et al. (2022) consider the NNW-SSE segment of the Fiandaca fault not hidden because shows morphological evidence. Some lava flows show a flexure in their sectors intercepted by the fault, but without clearly changing its surface slope angle. For instance, where the Linera Fault trace (LIN in the Fig. 1 attached) cross the "mg” lava surface, the latter shows a clear change on the surface slope angle, specifically there is an increase. This flexure in the Linera-Santa Tecla fault means that the "mg” lava flow covered an old scarp without completely resetting the previous surface morphology (see Fig. 1 attached). In the case of the Fiandaca Fault instead, the "sd” lava flow shows a flexure and in some cases the slope angle is reduced in the subsided block of the fault (see profile in the supplementary material and Fig. 2 attached). This morphological evidence appears to represent an initial stage of deformation on the surface of a lava flow that completely covered the fault scarp. In fact, the flexure may be related to the recent deformation acting with a trishear mechanism (Fig. 3 attached), which produces a deformation similar to the one reported in Romagnoli et al. (2021). Tortorici et al. (2021) state that the fault plane is not yet propagated to the surface where recent lava cover occurs; we agree with this interpretation. This does not mean that there is not deformation on the surface, in this case a trishear mechanism is active (see Fig. 3 attached). In addition, if we consider lower movement rates, for example 1-2 mm/yr for the Fiandaca Fault (see Azzaro et al., 2012), the “sd" lava flow would have had a vertical throw similar to the current one. Consequently, the 1329 lava flow, which is approximately 15 m thick, would have erased the old morphological evidence of the fault anyway. Therefore, the throws on the 1329 lava flow are real. Based on these considerations, throws on"sd" lava may also be considered real. We will improve this part in the manuscript, and we will also add a new figure.

7 Throw rates analysis and flank dynamics (2)

Paragraphs 7.2 and 7.3 are secondary in the framework of the manuscript. There is a lot of work but the basic hypothesis is wrong or too weak in my opinion, and it would require much more data and analyses in a dedicated paper, not in this same paleoseismological study. 

Reply to Throw rates analysis and flank dynamics (2): Thanks for the comment. This part would be the most important novelty of the work but we will follow your advice. We will explore this in a new dedicated paper.

7 Throw rates analysis and flank dynamics (3)

Conversely, I suggest to move in this part the materials related to the morphological profiles and related tables of the Fiandaca Fault, actually put in the electronic appendix. So the reader will easily find any information in the text.

Reply to Throw rates analysis and flank dynamics (2): Thanks for the advice. We agree and we will move the table in the main text.

8 Conclusions and future perspectives

Conclusions are repetitive. While it is obvious repeating the results of investigation, it would be also useful to comment critical issues and uncertainties of the paleoseismological investigation in a volcanic environment. This methodological approach is hard to apply here, but it has a significant potential to contribute for reconstructing the seismic history of a fault. Only basing on a significant dataset of paleoseismological data relevant to different faults (not yet existing), it will be possible to infer insights into geodynamic processes acting in the Etna region.

Reply to Conclusions and future perspectives: We are grateful to the reviewer for this comment. We will add a discussion about the paleoseismological trench investigation in a volcanic environment. The increase in late Holocene to historical throw rate has been already described in the published literature. We were only suggesting that this increase can be related to more frequent sliding event because of the ongoing flank instability process. As mentioned above, we will explore the hypothesis of the flank instability increase in a dedicated paper with more data.

Dear Editor,

we are grateful for your revision and comments. We will modify the paper removing the discussions about the flank instability that will be the focus of a new paper. Moreover, after modifying and deleting sections of the article, the abbreviations refer only to the investigated fault, and other faults are rarely mentioned.

Specific comments

Line 13: Missing statement why this earthquake is important

Thanks for the suggestion, we added why the earthquake is important.

Line 17: "strongly" - here and elsewhere: please avoid the use of such semi-quantitative terms that seem to push the reader into a certain direction.

Thanks for the suggestion, we will remove them.

Line 18: verify ==> test

Done

Line 27: why focus on basaltic volcanoes here? Which limitations exist? delete "considerable"; "highly dynamic" is too vague. What do you mean?

We are grateful to the Editor for this comment. The basaltic volcanoes, as Mt. Etna, can be very active, so the lava flows, due to the high resurface process, can easily cover fault scarp. We modified this sentence to be clearer.

Line 30: delete "significant"

Done

Line 32: "Not much has been done" ==> colloquial. Please rephrase

Done

Line 38: rates ==> rate; specify "over longer time windows"

Thanks for noticing. In this case it is a shorter time window. Moreover, we will add also morphotectonic alongside paleoseismology.

Line 39 following - much of this has been said above

Thanks for the comment, we will remove the repeating parts (Lines 39-40).

Line 40: specify "these geological processes"

This part has been removed as you suggested in the previous comment.

Line 42: which methodologies do you refer to?

This part has been removed as you suggested in the previous comment.

Line 43: "disentangling purely tectonic from volcanic-induced deformation" ==> this seems to be a stretch. What is the difference between these two and volcano-tectonic events? I don't think that you can disentangle these events, as they are closely linked, and chicken and egg is not clear. i.e., a volcanic event can cause a tectonic event, and vice versa.

Thanks for the comment. We agree and we will remove the sentence.

Line 47: "volcano behaviour" is too vague. please specify

We will remove all the sentence as suggested by reviewer 1.

Line 50: "verify" ==> test

Done

Line 53: specify "significant effects"

Done

Line 56: delete "careful" (should go without saying ;) )

Done

Line 64: "entire volcano flank dynamics". What does "entire" mean. Your study is smaller than that. 

Following this and reviewer 1 comments, we will remove the comparison with the entire volcano.

Line 70: sort abbreviations alphabetically

Done, except for the Fiandaca Fault which is the first.

Line 79: split sentence in two

Done

Line 79: 1-3 mm: based on what data?

The data are obtained from paleoshorelines, we will remove this sentence because it is not relevant in the revised version of the manuscript.

Line 87: feed ==> fed

Done

Line 98: andwith ==> and with

Done

Line 101: 1865 CE, respectively

Done

Line 113: "is interested" ==> change wording

Done, changed with “affected”

Line 129: Regardless the TFS meaning ==> change wording

Done

Line 131: as it is further ==> as further

Done

Line 135 & 176: delete "significant"

Done

Line 183: split sentence

Done

Line 186: S ==> southern; does the termination rupture seismically and creep, or is it fully aseismic, as opposed to other parts of the fault?

The southern termination shows both coseismic and aseismic, fault creep ruptures

Line 199: numbering incorrect

Fixed

Line 201: delete "clear"

Done

Line 211, 212: W-most ==> western most

Done

Line 226: "coded" rephrase

Done

Line 227: delete "meticulously"

Done

Line 230: delete "deposit"

Done

Line 235-236: sentence can be deleted. In case you want to keep the reference, use a similar sentence as first sentence to this paragraph.

Thanks for the suggestion, it has been moved to the introduction.

Line 243: rephrase

Done

Section 4.3: should be one paragraph, references missing after "... published recently"

References are not needed in the new version of the manuscript, so we will remove the sentence. We moved this section to one paragraph.

Line 298: 3 ==> three

Done

Line 349: constraint ==> constrain

Done

Line 355: what kind of "erosive event"? couldn't this be gradual? How much material has been removed?

We do not think that this erosive event removed a lot of material, but the surface has been exposed until the accumulation of new material probably coming by some ash fallout deposits mixed with local colluvium, where the new soil subsequently formed.

Line 359: delete "some" and "generally"

Done

Line 374: how much material has been eroded?

In accordance with Coltelli et al. (1998) the thickness of the fallout deposit was about 10-30 cm in the area, so not much.

Line 384: W ==> west

Done

Line 391: to ==> with

Done

Line 393: period between

Done

Fig. 12: add magnitudes

Thank you for the suggestion. Magnitudes and intensities added to the figure for earthquakes following the yar 1875. Previous earthquakes are poorly known and info about intensity and magnitude are not quite reliable.

Line 401: delete "strongly"

Done

Section 7.1: you have to describe more explicitly how you arrive at the throw rates. Consider moving table A1 to the main text, so it becomes more transparent what you do

Thanks for the suggestions. We will extensively improve this part in the revised manuscript as replied to reviewer 1 (see also figures attached to "Reply on RC1"). Table A1 will be moved to the main text.

Fig. 13: you show a power law trend. If you were to project this trend to today or even the near future, there would be an unrealistic amount of throw. This trend can be flawed by factors such as incomplete observations or strain distribution across multiple faults. Power law trends are always suspicious of bias, as we know from sedimentation or erosion rates. You have to discuss this more extensively. 

Thank you for this advice. We will remove this part and eventually we will extensively discuss this trend in a dedicated new paper.

Line 434: gradually ==> gradual

Done

Line 446: what do you mean by "undergoing recent evolution and expansion"

The deformation has moved recently toward the east. However, we will remove this part from the manuscript.

Line 449: delete "significant"

Done

Line 450: this growth of the edifice and it's rise to it's maximum size in the past 20 kyrs may cause reactions until today, but does not explain why a power law increase of throw rates should occur so recent. I think this is an over-interpretation of the data

We are grateful to the Editor for this comment. We will remove this part and we will discuss this in a dedicated new paper.

Lines 468 following: move to conclusions

This part will be removed from the final version of the manuscript.

Line 475: specify "change in volcanic behaviour"

Thanks for the comment but this part will be removed from the final version of the manuscript.

Line 488: "interested" rephrase

Thanks for the comment but this part will be removed from the final version of the manuscript.

Line 490 following: split sentence

Thanks for the comment but this part will be removed from the final version of the manuscript.

Line 493: grow == growth

Thanks for the comment but this part will be removed from the final version of the manuscript.

Line 514: "if the observed trend will persist" ==> that's impossible. This power law trend would mean an incredible amount of throw already happening now (approximately 73 mm/yr). 

We are grateful to the Editor for this comment. We agree. We do not desire generating unjustified alarmism. In fact, we do not believe that there will be an increase of this rate in a short period of time and probably will not even be there in the future, unless particular changes in the volcanic system lead to increasingly explosive activity that may triggers new collapses. Anyway, this part will be removed from the final version of the manuscript and better discussed in a separated companion paper.

Line 515: "consequences in the associated hazards" is too vague. please specify

Thanks for the comment but this part will be removed from the final version of the manuscript.

Lines 545 -560: these three paragraphs are mostly speculative, vague, and unnecessarily conjure up crisis scenarios. Based on a re-evaluation of the trend they should be down-toned. 

Thank you for this comment. We do not desire conjuring up disaster scenarios, this is not the purpose of the paper. Furthermore, we all agree that a catastrophic scenario at the moment is not possible at Mt. Etna. However, we will remove this part from the final version of the manuscript and we will better discuss this in a separated new paper.