Review of: “The 1538 eruption at Campi Flegrei resurgent caldera: implications for future unrest and eruptive scenarios”, by Rolandi et al.

Overview

This paper contains a great deal of useful information about historical unrest at Campi Flegrei. The information comes from the published literature and from new reconstructions of the volcano’s behaviour. The large amount of information has led to a lack of focus in the main theme and the mixture of old and new data has obscured its novelty. In addition, the conclusions have not obviously been derived from the observations provided. Judicial editing and restructuring of the text would yield a paper in which the logical thread of the analysis and the advance in understanding are made clearer. I have made numerous suggestions below. The number of changes indicates major revision. Even so, the adjustments should be straightforward.

Common themes are:

(1) Reorder the text. One option is:

• Describe observations and current views on ground movement and seismicity before the Monte Nuovo eruption in 1538 and since the unrest that began in 1950. Describe the key features of the subsurface structure of the volcano and connect this to the processes (g., transport of gas and magma) that may be contributing to unrest.
• Describe new reconstructions of events before 1538. There is no need to repeat in detail previous studies – these can be acknowledged by citations in the text. Give just the key points and show how they have been modified by the new studies here.
• Compare the pre-1538 and post-1950 behaviour to investigate whether the same sequence of events can account for both sequences. The different roles of the transport of magma and magmatic fluids could be a key topic.
• Discuss future scenarios based only on the data presented.

(2) Clarify which material comes from previous studies and which is new to this paper. Much of the historical data has appeared elsewhere and does not need to be described at length here. Previous work must be fully cited: this would highlight the novelty of this work.

(3) Remove extraneous topics that do not contribute to the goals and conclusions of the paper (e.g., the account of the Monte Nuovo eruption).

(4) Explain the significance of a resurgence occurring as a block. Repeated pressurization at a common depth (which may happen to be shallow) is expected to produce the observed patterns of ground movement without needing to invoke wholesale movement of bounding faults (see for example Acocella (2019)). Hence, it would be good to have further guidance on the importance of whether or not fault-bounded movement is essential and whether this can constrain the potential for eruption. For instance, might a block model be appropriate only after a critical amount of movement has been achieved (e.g., movements before 3.700 years ago) but not significant otherwise (e.g., movements before 1538 or since 1950)?

(5) Remove the text that is peculiarly negative in tone. In some sections, new material is presented with the apparent aim to demonstrate where others are wrong, rather than to show how it can improve current understanding.

Specific Comments

Abstract

This will need to be adjusted to accommodate the revised text.

Introduction

Lines 32-34. These lines cite previous work on reconstructing historical movements at Campi Flegrei and then state that all their conclusions need to be modified. However, only the work by Parascandola (1943) and Di Vito et al. (2017) is later discussed in detail. Dvorak and Mastrolorenzo (1991) is not mentioned, while work published by Bellucci et al. (2006) is repeated but not cited. The referencing of earlier studies must be more inclusive throughout.

Eruptive history of Campi Flegrei.

Lines 68-100. Reduce text by 30% and focus attention on the movement of a central block. The conventional interpretation in the literature is that the caldera formed about 40,000 years ago, during the eruption of the Campanian Ignimbrite (CI) and that the collapse during the Neapolitan Yellow Tuff (NYT) eruption represents a subsidiary movement. To state that the caldera formed only 15,000 years ago may well be correct, but it is not yet generally accepted. The key point here is that evidence of a resurgent block is only available since the NYT eruption, regardless of whether the caldera formed then or during the CI eruption. To acknowledge the uncertainty, perhaps the text could be adjusted to something like:

                “Campi Flegrei is an active caldera to the west of Naples in southern Italy. About 12-14 km across, its southern third is submerged beneath the Bay of Pozzuoli. Following the most recent – and perhaps only (Rolandi et al., 2020) - episode of caldera formation, some 70 eruptions have occurred across the caldera floor, ranging from the effusion of lava domes to explosive hydro-magmatic (?) eruptions (Di Vito et al., 1999; Smith et al., 2011; Isaia et al., 2015). The most recent eruption occurred in 1538, producing the cone of Monte Nuovo (Di Vito et al., 2016)…”

                For completeness, add citations to previous work on resurgence at Campi Flegrei, including: Luongo et al. (1991), Orsi et al. (1996, 1999) and Acocella (2010).

Subsidence and uplift before the 1538 eruption

Lines 103-368. This key section presents an exhaustive evaluation of historical observations on ground movement at Campi Flegrei. However, it is not clear which observations are new and which are referring to previous studies. To highlight the new material, it would be helpful to have:

1. A paragraph identifing previous interpretations, by bringing forward Fig. 13 a and b (and related text) to provide a starting reference. I would add also the trends proposed by Dvorak & Mastrolorenzo (1991), Bellucci et al. (2006) and Morhange et al. (1999, 2006).
2. A clear statement at the end of the section that specifies the novel results from this study. For example, how new are the trends for the via Herculea (Fig. 3) and Serapeo (Figs 6 & 13c)?

Lines 109-110. I would avoid claiming that this paper is “correcting misrepresentations or erroneous reconstructions”. This is a value judgement and may be viewed by readers as hostile. Why not say something like “the new data allow more detailed (?) reconstructions than have previously been possible and, as a result, provide tighter constraints on the mechanisms driving unrest.” Such an approach conveys the positive idea that this paper in building on previous work, rather than correcting it.

Lines 128-171. Only a single source (Parascandola, 1943) has been cited for the descriptions here. If this is correct, then what advances have been made by the current paper? If this is not correct, can the authors identify the sources of their new data?

In Fig. 3, the different trends in ground position shown by the continuous and dashed lines need to explained in the caption; similarly, what do the numbers 1-10 indicate? Which measurements have been published before (e.g. by Parascandola (1943)?) and which are new to this paper? If the original data have been moved to Supplementary Material, why not add a table here that gives short descriptions?

Line 197 (and Line 116). It may help to include a subheading (not necessarily numbered) that the following text has switched observations of ground movement from Via Herculea to the Serapeo. If so, an equivalent subheading for Via Herculea could be inserted around Line 116.

Lines 238-265. These lines describe Fig. 7. The text is virtually identical to that in Bellucci et al. (2006, pages 149-50), for which Rolandi is co-author. There is no need to repeat the text at the same level of detail: it can be shortened with reference to Bellucci et al. (2006), which ought to have been cited in the first place. 

The numbered sources in Fig. 6 need to be identified and not relegated to Supplementary Material. Why not add these to the table already recommended to accompany Fig. 3? What is new about this plot? How does is differ from previous measurements? How are the dates on the right of the graph related to the trends? In particular, nine dates are given from 1430 to 1538, yet only three points are shown defining the uplift between c. 1430 and 1538: why have more dates been shown than the number of points?

                An important omission is the interpretation of ground movement by Bellucci et al. (2006) and which the present authors have used in subsequent publications (e.g., Troise et al. (2007, 2019)). It includes a possible uplift around 700-800 AD, following archaeological evidence presented by Morhange et al. (1999). Although Morhange et al. (2006) later removed this uplift without explanation, no clear evidence has been presented to resolve whether or not the uplift occurred. Any new data in Fig. 6 are thus especially significant because they offer an opportunity to clarify this point. Indeed, Fig. 6 shows horizontal arrows to highlight two measurements that happen to show continued subsidence across the relevant time interval. Is this deliberate or a coincidence? An explanation is necessary here.

Lines 295-312. Please add references to previous studies in the main text: don’t hide them in figure captions. Can the authors specify where they have added their “own reworking” and so highlight the advances in this paper?

Lines 318-328. The authors present evidence why they believe some inferences about ground movement made by Di Vito et al. (2016) need to be revised. Can they suggest which observations led Di Vito et al. (2016) to estimate different values? As for their comments on Claims (2) and (3), it is notable that the trend shown in Fig. 6 from the 12th Century to 1538 supports that presented by Bellucci et al. (2006, Fig. 7), which should be referenced to avoid giving the incorrect impression that all the data are new to this paper.

Lines 336-338. The inference of a maximum subsidence in the 1400s is not new to this paper. Dvorak & Mastrolorenzo (1991) and Bellucci et al. (2006) reached the same conclusions: at best, this paper confirms their results (although the similarity with Bellucci et al. (2006) may also indicate that some results are being repeated, especially since the lead author here was a co-author on the earlier paper).

Lines 341-346. The occurrence of seismicity is out of place. I would move the observation to a general discussion after the sections that describe the seismicity more fully. The new paragraph could then begin directly with “Our findings...” (Line 346).

Line 358-9. As above, the comment on seismicity can be moved to a later discussion.

Lines 366-8. The comment about the “anomalous” pre-eruptive uplift at the site of the Monte Nuovo eruption appears without context. I suggest mentioning this behaviour at the start of the reconstruction and then explain why it can be neglected when investigating ground movement across the caldera.

Ground movements after the 1538 eruption

Lines 370-428. This section describes subsidence between 1538 and c. 1950. To emphasize the relevance, it may help to compare rates of subsidence with those before 1538 and, perhaps, draw conclusions about whether conditions in the crust before the 1430-1538 uplift might have been similar to those before 1950, hence further supporting the idea that the two sequences may have been driven by similar subsurface processes.

Schematic model for the preparatory phases of the 1538 eruption

Lines 448-455. I would adjust the text to argue that the resurgent-block model is consistent with observed ground movement and seismicity, but is not the only possible interpretation. Numerous models have accounted for the observations without involving block movement (among others, Berrino et al., 1984; Bianchi et al., 1987; Amoruso et al., 2008, 2014; Woo & Kilburn, 2010). The authors could then explain the significance of whether or not block movement is significant to determining uplift before 138 and since 1950 (bringing forward text from Lines 472-478).

Lines 455-471. I would move the description of subsurface structure to the start of the paper, as part of the background context.

Lines 475-483. Could the authors clarify the contradiction in proposed substructures? The authors argue that tuff occupies the upper 1.5-2 km of the crust (Lines 474, 482 and 499). How does this contradict the “layer of loose pyroclastics from recent eruptions” (Lines 476-7)? How “loose” will these deposits really be 1.5 km below the surface? Will they behave differently from a tuff in any significant way? Presumably the upper levels do contain deposits from recent eruptions too.

Lines 491-507. The text would be easier to follow by starting with field data that support the presence of a thermometamorphic horizon (e.g., borehole and gravity data (Rosi & Sbrana, 1987)) and fluid filled rock at shallow depth (doesn’t this correspond to Campi Flegrei’s hydrothermal system? If so, just say this and cite suitable papers) and then speculate on its origin. I don’t see that the discussion on Lines 491-498 adds anything to the argument. It could be omitted, perhaps moving the citation to Vinciguerra et al. (2006) elsewhere.

Lines 519-522. Why a mush? The authors have argued that renewed uplift may have occurred during c. 1400-1538 and since 1950. If driven by magma intrusions, then presumably the maximum volume of magma intruded corresponds to the amounts associated with these uplifts. Would not the amount involved have fully solidified by now?

                For example, the slowest rate of heat loss is expected to be by conduction. The thickness of a solidified layer is on the order of (4kt)^1/2, where the thermal diffusivity k is ~ 4 x 10^-7 m² s^-1, and time t is in seconds. For a single sheet-like body, the thickness will thus be about 7 m in 1 year, 22 m in 10 years, 35 m in 100 years and 80 m in 500 years. Since cooling occurs across the upper and lower boundaries, the total thickness solidified will be about twice these values. If the surface area of the sheet is ~10 km² (e.g., a circular sheet c. 2 km in radius), then the corresponding minimum volumes of magma involved are 0.15, 0.45, 0.7 and 1.6 km³. In others words, an intruded volume of 1.6 km³ before 1538 may have completely solidified before the return to uplift in 1950. At the other extreme, an intrusion of as much as 0.6 km³ in 1984 would have solidified before the return to uplift in 2004.

                Although these calculations are approximate, they suggest that it may be difficult to preserve a mush layer at depths of 2-4 km over the required time intervals; indeed, the authors suggest as much when discussing the uplift since 1950 in Lines 594-631. If this is correct, can the authors address the apparent contradiction between mush and full solidification and, if necessary, propose an alternative scenario that does not depend on the existence of mush at such depths?

Lines 529-534. Subsidence has been occurring between 3,700 years ago and 1538. Can the authors offer some independent checks for consistency? For example, if the rate has been similar to that in historical time (1.5-2 m per century), then ground level 3,700 years ago must have been some 50-60 m higher than in 1538. Is this consistent with the reconstruction of prior resurgence?

Lines 537-546. Is the level of detail necessary here? None of the specified processes can be demonstrated to have operated at Campi Flegrei. I would simply postulate that a mush zone exists – perhaps referencing more direct evidence from the seismic-tomographic analysis by Zollo et al. (2008), who argued that the crust at these levels contain dispersed patches of molten rock. Additional papers can be left as citations.

Lines 546-551. I found the text confusing. Do the authors mean that the rapid uplift between 1430 and 1538 could have been caused by the release of gas? If so, why not just state this in a single sentence?

                It may be of interest that Mormone et al. (2011) have presented evidence from melt inclusions that for deeper magma entering the “8-km” reservoir (mush zone?).

Lines 552-583. The description jumps back and forth between different levels in the crust. I would restructure and shorten the text – perhaps following progress upwards from the “8-km” reservoir.

Lines 552-554 could form part of a revised introduction to the Campi Flegrei’s subsurface structure. Links to Yellowstone can be removed, or moved to a later discussion that compares in a single section the authors’ model of Campi Flegrei with other volcanoes.

Lines 557-561. Can the authors connect their description of saturated rock to the hydrothermal system that (according to the literature) exists at similar depths?

Lines 567-8. These lines need to be adjusted. Kilburn et al. (2023) proposed common depths for magma intrusion and gas accumulation below the thermo-metamorphic horizon at depths of c. 3 km (and not beneath a shallower cap-rock). I would also suggest changing how the comment is presented. For example: “We consider that magmatic gases may not necessarily be restricted to below the thermo-metamorphic horizon (Kilburn et al., 2023), but may instead accumulate at shallower levels beneath the “summit” lava at a depth of c. 2.5 km.” In general, I would avoid statements that previous work is “wrong” or “incorrect”. After all, the authors have not demonstrated this: they have instead proposed an alternative view for consideration (which may or may not turn out to be an improvement).

                Another constraint to consider is whether the proposed depth of 2.5 km for gas (and pressure?) accumulation is consistent with depths closer to 3-3.5 km estimated by geodetic modelling of uplifts in 1982-84 and since 2004 (e.g., Amoruso et al., 2008, 2014; Amoruso & Crescentini, 2022).

Line 567. The term “summit” lava is distracting. Why not refer to the “lava level at 1.5 km” (or something similar) to connect with the lava mentioned in Line 560-1. Moreover, could this unit be an intrusion rather than a “lava”, which implies that it flowed over the surface?

Lines 572-645. The ideas described are interesting, but have jumped to processes operating today, rather than before 1538, which is the title of the subsection. I suggest reorganising the text to appear as a separate section focussing on processes operating today, followed a separate discussion comparing the pre-1538 and modern behaviour (see potential connection to Lines 838-856 below)..

Lines 594-635. The authors describe the intrusion of magmatic sills (presumably at the depths of 2.5-4 km (Line 602)?) and solidification of these sills within 20 years (Line 610). This conclusion is based on previous work. It appears to contradict Lines 546-551, which argue that gas transfer is the mechanism driving unrest. The connection between gas release and magmatic intrusion needs to be clarified – and succinctly: to be honest, I think the description could easily be reduced by 40% (which would force the authors to focus on essential features). Fig. 16 is attractive, but essentially shows collections of arrows indicating an upward transfer of pressure. Which arrows refer to gas transfer and which to magma movement? I would redraw simplified figures that specify the features described in the text. There is no need here to rely on speculative generic models, such as that proposed by Bachmann & Bergantz (2006): the figure can be based solely on the evidence presented in this paper.

Line 585. See comment on Lines 567-8: Kilburn et al. (2023) placed the zone of pressurization below the thermo-metamorphic horizon.

Lines 587-593 can be omitted*. All they do is introduce what is about to written next. Go straight to the next paragraph. (*Or perhaps a shorter version can be moved to the conclusions?).

The eruption of 1538

Lines 647-719. Why is this section included? The eruption of Monte Nuovo doesn’t contribute to the rest of the paper? It is really a review of previous work and does not present new information relevant to the main arguments of this paper. (N.B. In Fig. 19, Wohletz’s figure of magma-water interaction and eruptive explosivity has previously been applied to Campi Flegrei. I can’t remember the authors or journal, but believe it was published in the late 1980s or 1990s.)

The seismicity before and after the 1538 eruption

Lines 731-822. The connection between earthquake magnitude and intensity is extremely interesting. Again, though, I think too much detail has been added. It breaks the structure of the paper and loses sight of the key objectives. Figures 18 and 19 are the key results and the text could be shortened so that they become more prominent. Thus I would (1) move the methodology to supplementary material; and (2) tabulate the sequence of events, because they have already been described comprehensively by Guidoboni & Ciuccarelli (2011). I expect the whole section could be reduced to two figures, one table and half a page of text.

Post-eruption seismicity

Lines 825-835. How relevant is this section? It hasn’t clearly been connected to understanding the processes operating beneath Campi Flegrei.

Comparison of precursory phase of the 1538 eruption with current unrest

Lines 838-856. If the authors could present a graph showing the seismic behaviour during unrest since 1950, this paragraph could then be moved to the end of the revised description of pre-1538 seismicity (Figs 18 & 19).

Lines 858-918. The alternative scenarios are a good idea, but not clearly connected with the picture of gas and magma transport the authors have been developing. This is a missed opportunity.

                The discussion of future vent location and final rapid ascent of magma (Lines 871-894) is interesting, but does not naturally follow from the previous text. It thus seems arbitrary and can be removed.

                The same holds for Lines 897-918. This section passes comment on previous work – especially Kilburn et al. (2023) – without evidence or serious analysis. If the authors wish to propose alternative views, they should do so in a separate paper that presents an in-depth analysis; in any case, these views have nothing to do with the present paper and are not supported by the data in the preceding text. [For the record, the statements in Lines 904-914 are simply wrong: (1) the seismic time series did change in 2015-17 and, as we know now, evolved towards a full seismic crisis; (2) the corrected uplift in 2015 had reached its 1984 level; and (3) major rupturing did begin in August-September 2023.]  

                It would be more interesting to see how the revised analysis of events before 1538 might better constrain what is happening today. As I understand it, the authors' pre-1538 reconstruction suggests early uplift driven by gas transfer, followed by magmatic intrusion and eruption (comparison with the model of Bodnar et al. (2007) and Lima et al. (2009) may be valuable here). If this is correct, can the authors identify comparable stages today – or is the modern sequence significantly different in some way?  Given the large amount of work in preparing the reconstruction, it seems a pity not to take full advantage of the new results.

Figures

Most of the figures are relevant and of high quality.

Figures 3 & 6. The time axis is labelled as “Centuries”. This can cause confusion, because the century may be taken to be the preceding hundred years (for instance, in standard English usage, the fourth century refers to 1300-1399). I would replace the times with actual years (as done in Fig. 13).

Fig. 15. Consider changing the colours to make the shapes outlining zones clearer on the map (especially the red of the caldera outline).

Please check language. Some terms remain in Italian: e.g., in Figs 3, 6 and 13, time “AC” should be “AD” (use either the combinations BC and AD, or BCE and CE (“Before Common Era” and “Common Era”), and in Fig. 13 “duttile” should be “ductile”.

Language

The English is generally good, but contains grammatical oddities that could be spotted with help from a native English speaker. I would avoid phrases that describe interpretations as “likely”. Their use may be mistaken as attempts to support uncertain interpretations without sufficient evidence. They are not necessary here.

References

Please check that all references are in alphabetical order (e.g., Lines 1013-1024 need to to be brought forward), and also some of the dates that have a number missing (e.g., Line 1010). Also, check that citations in text are presented in chronological order.

Recommended Additional References

Acocella (2010) Bull Volcanol 72:623–638.

Acocella (2019) Front Earth Sci 7:173, doi: 10.3389/feart.2019.00173

Amoruso et al. (2008) Earth Planet Sci Lett 272:181–188.

Amoruso et al. (2014) J Geophy. Res 119:858–879.

Amoruso & Crescentini (2022) Remote Sens 14:5698. doi.org/10.3390/rs14225698.

Berrino et al. (1984) Bull Volcanol 47:187–200.

Bianchi et al. (1987) J Geophys Res 92:14,139–14,150, doi:10.1029/JB092iB13p14139.

Isaia et al. (2015) Geol Soc Am Bull,  doi:10.1130/B31183.1.

Luongo et al. (1991) J Volcanol Geotherm Res 45:161-172.

Morhange et al. (1999) Phys Chem Earth 24:349-354.

Morhange et al. (2006) Geology 34:93-96.

Mormone et al. (2011) Chem Geol 287:66–80.

Orsi et al. (1996) J Volcanol Geotherm Res 74:179-214.

Orsi et al. (1999) J Volcanol Geotherm Res 91:415–451.

Troise et al. (2007) Geophys Res Lett 34, L03301, doi:10.1029/2006GL028545.

Zollo et al. (2008) Geophys Res Lett 35:L12306, doi:10.1029/2008GL034242.

This is a very interesting paper, in which historical data are thoroughly revised to propose a revisited chronology of the ground (vertical) displacement over a long period time preceding  the Mte Nuovo eruption (1538 CE) at Campi Flegrei (CF). I appreciate the historical background on which the study is founded, particularly the analysis of the relative position of the Roman artifact in Via Herculaea, which puts constarints to the maximum  uplift.

First of all, the new reconstruction is interesting because it shows a steeper but rather continuous ground uplift prior to eruption. This steep uplift actually looks like the one observed in recent times, during the 1982-84 unrest episode.  Here is my first comment: this similarity is not discussed and perhaps it should be in the double-scenario section. Worth noting is the long duration (abot 100 years prior to eruption) of such a steep uplift, with implications also for uplift rates observed during the current crisis and the 1982-84 one.

Secondly, the new reconstruction of the pre-1538 uplift and seismicity  highlights the two and half scenarios that should be expected. I say "... and half" because the phreatic eruption is considered somehow apart in case the system does not reach the threshold for a magmatic eruption. This is correct and it should be a scenario "per se", which should be more emphasized by the Authors. In my view we do not know of many phreatic events at CF simply because they were soon erased by the following magmatic-explosive activity. However, phreatic eruptions must have occurred and likely few of them have represented some kind of peak activity for longtime, before the culmination in an magmatic-explosive eruption. The phreatic scenario deserves a lot of attention for the current evolution of the CF unrest.

Thirdly, the new reconstruction offers a semi-quantitative criterium to establish the evolution of seismicity as uplift continues: the M 5 event from historical sources is put into the evolutionary context of the 1538 CE eruption. Also worth noting, although from a qualitative point of view, the role of thermal cracking which accompanies the steep uplift.

Although the Vinciguerra et al paper is a central one, note that thermal cracking was also invoked in previous papers, particularly Chiodini et al. (2015, EPSL) and Moretti et al. (2018, SciRep), in which thermal cracking was approached from different perspectives. Also past papers from the same Authors of this study mentioned it.

Of particular interest is also the role attributed to a resurgent block (see also below), which is the same as the one nowadays active, and which in literature was already described. I wonder whether the resurgent block which the Authors refer to corresponds to that in Capuano et al. (2013) : just see their figure 6 and the text part in which its descriptions starts by saying : "We refer to the South-western deeper sector as the ‘undeformed-to-subsiding portion of the Pozzuoli Bay,’’ and to the North-eastern shallower sector as the ‘‘resurgent portion of the caldera’ (Figure 6)". The descriptions which follows of the resurgent block refines previous findings by Orsi et al. on  JVGR (1999; vol. 91, pp 415-451).

Let me remark that in this paper the cause of unrest at CF - also the current one, if I uderstood well - is attributed to some hybrid mechanism of heat and fluid release from a magma batch ascending from the deep regional body (Zollo et l., 2008 GRL) and percolating through the caldera bottom. How significant is this view for the advancement of the current scientific debate around the causes of unrest at CF ?

Finally, I appreciated the reconstruction of the low-energy pyroclastic flow that reached the Pozzuoli temple (Serapeum). It is in line wth the rationale of the study and its title, but it seems a bit disconnected from the main text flow. This is just about text organization and perhaps the Authors could harmonize the whole thing by recalling in the Conclusions that the spectrum of phenomena which result from this study and  apply to the eventual evolution of the present-day unrest include 1) increasing seismic activity and M 5 events , 2) phreatic eruption and 3) pyroclastic flows reaching Pozzuoli in case of an eruption like the 1538 one.

The paper requires an attentive English revision and also more attention to previous work, which should be properly cited. I understand that there is a bunch of literature, but this can be improved.

Some minor points:

- With ref. to Troise et al. 2019, at line 863 the Authors say "The same authors calculated that such a sill intrusion should have solidified, in form of mush, after about 20 years, i.e. around 2003. Closely rleated are the statements also reported at lines 609-10 and 518-519. Actually, rapid solidification of a thin sill like mahgma in the order of 10 to 20 years was put in evidence by Woo and Kilburn (2010 JGR), Moretti et (2013 EPSL) and it was modelled by Moretti, Troise et al. (2018 Sci Rep).

- At line 849 the M 4.2 event on October 2023 is reported as the maximum one recorded at the time of writing, but somewhere in the Introductory part the Authors refer to a M 4.4 occurred in 2024