This document is the response to the reviews of the paper titled “Paleoseismic history of the intermountain Rieti Basin (Central Apennines, Italy)”, which we submitted to the journal NHESS.

We wish to thank the reviewer Gerald Roberts for the positive feedback and for taking the time to provide improvements to our text. Here, we provide a point-by-point answer to all the comments raised by the reviewers. Original comments are shown in plain text, while answers are in italics.

The revised manuscript will be eventually uploaded in the journal system.

Reviewer’s report on Livio et al. “Paleoseismic history of the intermountain Rieti Basin (Central Apennines, Italy)” (Prof. Gerald Roberts - 1st July 2025)

The manuscript presents new insights regarding the paleoseismic history of the Rieti Basin in Central Italy, derived from the excavation of 17 paleoseismic trenches.

These results will be incredibly valuable, both locally for the seismic assessment of the Apennines, but also in a general sense to aid the understanding of active faults worldwide. I was particularly impressed by the number of trenches, the number of AMS dates, and the detailed trench logs and photos. The descriptions of the trenches and the stratigraphy was also very detailed. The maps of the sites were also of high quality. The observations are put into context of the Rieti basin faults in a clear way. As such, I suggest accepting this manuscript with minor changes.

Thanks for the very detailed review, Below, we address your comments and describe the changes we will implement in the text.

My comments as follows:

Overall, I was left wondering about how these observations fit in with the regional context of extension across the Apennines. Would you like to add some text on this? For example, how does the timing and magnitude of slip you have observed relate to that on other neighboring faults? For example, were other faults on the SW flank of the Apennines also slipping when the Rieti basin experienced fault slip? I think it would be a missed opportunity not to say something about the regional context of the slip you have documented.

     Thanks for this comment; it is certainly an important point, and we missed this. We will add some text about this in the discussion before the “disclaimer”, near Line 502; as you suggest below, it is better to start the discussion with positive results. In fact, on nearby Quaternary faults located near the SW flank of the Central Apennines, we do have evidence of similar surface faulting events that occurred in the same period as those observed in the Rieti basin (e.g., Leonessa Fault: Mildon et al., 2022; Fiamignano Fault: Beck et al., 2018; Fucino Fault: Gori et al., 2017; Liri Fault: Maceroni et al., 2022). In particular, slip histories recovered from the 36Cl data using Bayesian MCMC modelling show that the Leonessa Fault, Fiamignano Fault, Fucino Fault, and Liri Fault exhibit a period of slip rate acceleration during the 7 to 5 kyr BP and 2.5 to 1.5 Kyr BP time windows (Roberts et al., 2025). The Rieti Basin faults generated the Earthquake Sequence 1 and 2 based on the paleoseismic analyses described in the manuscript during this same interval.

Refs to be added:

Gori, S., E. Falcucci, F. Galadini, M. Moro, M. Saroli and E. Ceccaroni, 2017, Geoarchaeology and paleoseismology blend to define the Fucino active normal fault slip history, central Italy. Quaternary International 451 (2017): 114-128.

Maceroni, D., Dixit Dominus, G., Gori, S., Falcucci, E., Galadini, F., Moro, M., & Saroli, M. (2022). First evidence of the Late Pleistocene—Holocene activity of the Roveto Valley Fault (Central Apennines, Italy). Frontiers in Earth Science, 10, 1018737.

The text from lines 560 to 630 is well written and clear. However, reporting the ages in “BCE” means readers need to convert the ages in years BP. Can you also add the ages as “years BP”? This would make it much easier for most readers.

Thanks for pointing this out. Considering that the recognized paleoearthquakes span a considerable time window and are also during historical times, we choose to use the CE/BCE date format. Nonetheless, we understand that, especially for older dates and for a direct comparison with other results obtained in the Central Apennines, with different techniques, we will add the BP format in the text of Section 5.2 and table 2 as well.

Please try to provide the amount slip for each of the events, even if this is a minimum estimate. Your approach is common amongst palaeoseismic trenching papers, where the magnitude of slip is commonly challenging to interpret, and it is common for nothing to be reported (I know you do report it for two of the events). However, it would be helpful to know what you think about the slip for all the events. If you have no constraints on the slip (e.g. because dated material just fills a fissure) state this. If you have some idea (e.g. colluvial wedge has a vertical extent of X cm), please state the value of X. If you can identify piercing points please state the exact value of offset (e.g. vertical offset, and some estimate of the fault dip with error bars). I feel it is better for you to say what you have observed about possible offsets per event rather than someone trying to interpret this from your trench logs at a later date – it is you who has the best chance of producing the best estimates, so it is a shame not to provide this.

Thanks for this note. Yes, it is indeed important to report all the possible available information related to the identified paleoevents. We will add our best estimate for the slip for each event; even if a minimum. We will clearly state if no estimation is available for specific causes. We will add the estimation in Table 2 and take care to clearly describe the constraints on the estimated slip per event in the trench description,

I also think that somewhere in your manuscript you need to define exactly what you mean by a “cluster” (e.g. a spatial cluster? Or temporal cluster where the implied slip rate exceeds the long-term slip-rate, or both? Or some other way of defining it?). You mention this word “cluster” on Line 634, yet you do not define what you mean by this. Also, if these are temporal clusters, as implied by the text on lines 635-640 (e.g. Line 640 mentions a period of quiescence”, implying the slip is clustered in time), how do you define this? For example, how do you assess the relative weightings of the (i) average recurrence interval over the long-term, (ii) the intervals defined by the preferred ages of the events (e.g. the aperiodicity), whilst considering (iii) the uncertainties defined by the analytical uncertainties on the 14C ages and the "brackets" they define? In other words, if you are going to talk about clustering, you need to add some text about how you define clustering in your particular case.

Thanks for pointing this out. Indeed, the term “cluster” is misleading. We do not observe a significant temporal clustering in the events. Analysis of the earthquake aperiodicity suggests a moderately regular recurrence pattern for the events A-D (aperiodicity ca. 0.53). After including also Events E and F, the calculated aperiodicity is close to 1.12 (i.e., a highly irregular recurrence pattern). Nonetheless, the stratigraphic record for the 20 - 10 ka interval can be considerably incomplete.

After these considerations, we substituted the term “cluster” with sequences to avoid confusion. Here, we intend to communicate a sequence as a series of characteristic earthquakes rupturing adjacent fault segments or adjacent faults over a short time period. So, it is mainly referring to a spatial criterion rather than a temporal one. We further explore the possible spatial interaction among the basin-bounding faults by running some Coulomb stress transfer models (for full details, see below the answer to the comment at line 655).

Line 502 – It might be better to start the discussion with a paragraph that states the main findings, and then go on to set the scene for the coming discussion by outlining the main points that will be covered. The reader will then know what is to come next. Starting with disclaimers is a little dull and off-putting. Instead start with positive outcomes.

     Yes, we agree. We will change the text accordingly.

Line 521 – you say it is “impossible” to find suitable sites, but what you really mean is that it is challenging and beyond your current capabilities. Please change the text.

Agreed. Text changed to “challenging”.

Line 535 – The text “underwent slope recession during the Last Glacial, are now very limited” is very vague. Imagine the reader who is not familiar with the Apennines; will they understand what this means? Re-phrase to describe the precise interplay between tectonic slip-rates and the change in erosion rate that occurred at the end of the last glacial maximum, perhaps even stating values in mm/yr, because such values are available in the literature (e.g. Tucker et al. 2011, JGR), and it is your duty to point this out to readers who do not know the Apennines.

We are grateful to the reviewer for this comment. Indeed, the interplay between tectonic slip-rates and erosion rates in the Apennines of Central Italy, particularly on carbonate bedrock fault scarps, is a well-studied example of how climatic and tectonic processes interact on geologic timescales. At the end of the Last Glacial Maximum (LGM), significant changes in surface processes—especially erosion—impacted how fault scarps are preserved and measured, with implications for estimating long-term slip rates.

During the last glaciation, the high elevations of the Central Apennines (reaching up to 2900 m) hosted mountain valley glaciers, as evidenced by the presence of moraines and other glacial landforms. In areas beyond the extent of glaciation, periglacial conditions prevailed. On mountain slopes carved in pure carbonate bedrock, rapid erosion and sedimentation (on the slopes the deposition of thick talus deposits – i.e., Grèzes litées or éboulis ordonnés) at rates higher than 0.2–0.4 mm/year (Tucker et al., 2011) led to the formation of alluvial fans emerging from ice-free mountain valleys and slopes. These rates exceeded the typical fault throw rates (e.g., Roberts and Michetti, 2004), as demonstrated by fan surfaces and colluvial slopes on fault hanging walls that are graded to the adjacent footwall bedrock slopes.

As the glaciers retreated, the reestablishment of temperate vegetation helped stabilize both the alluvial fans and surrounding slopes, while stream discharges declined. This transition resulted in the smooth hillsides that characterize landscapes shaped by former periglacial activity. Fan surfaces, bedrock slopes, and moraines are often covered by a thin (approximately 0.5–1.0 m) layer of soil enriched in organic material and, in some places, volcanic components, reflecting deposition during and after glacial retreat.

Elsewhere, the end of glaciation is marked by frontal moraines overlain by fluvial outwash or deposits from meltwater lakes. These sediments contain palaeovegetation and volcanic ash from nearby eruptions, and often provide material suitable for radiocarbon dating and tephrochronology. A large dataset of such dates allows for the determination of both absolute and relative ages of glacial and periglacial features and facilitates correlations with climate records from Tyrrhenian Sea cores and other oceanic and continental archives. The final major phase of glacial retreat occurred around 18–16 thousand years ago, coinciding with a significant shift in δ¹⁸O values observed in Tyrrhenian Sea cores and other marine records, confirming a major climatic transition.

Presently, normal fault scarps carved in pure carbonate bedrock cut through these glaciation-related landscapes. In many locations, these scarps expose Mesozoic carbonate platform bedrock in their footwalls and exhibit minimal degradation.

However, the marly-limestone bedrock from the Meso-Cenozoic Umbria-Marche-Sabina pelagic environment of the Central Apennines that is characteristic in the Rieti basin behaves differently from pure carbonate platform bedrock of the Abruzzi Mesozoic facies. In our study, we mainly investigated bedrock fault scarps carved in the marly limestone of the Scaglia Formation. Modern erosion rates on the Scaglia Formation are higher than those observed on the Abruzzi carbonate platform (0.016 ± 0.005 mm/yr; Tucker et al., 2011). In fact, prominent bedrock fault scarps in the Umbria-Marche-Sabina region (in the Rieti, for instance, Leonessa and Norcia basins) are only visible where pure carbonate formations such as Calcare Massiccio and Maiolica are outcropping. This point is relevant and was discussed by Blumetti et al. 1993.

We will add this discussion at the beginning of para 5.1.

Refs to be added:

Tucker, G. E., McCoy, S. W., Whittaker, A. C., Roberts, G. P., Lancaster, S. T., & Phillips, R. (2011). Geomorphic significance of postglacial bedrock scarps on normal‐fault footwalls. Journal of Geophysical Research: Earth Surface, 116, F01022, 1-14, doi:10.1029/2010JF001861, 2011

Line 545 – you imply that studying scarps is challenging due to human effects, but then show you can get results. So, don’t you think you are overstating the effect of humans on ruining scarps? This will be confusing and possibly misleading for readers unfamiliar with the Apennines. If one does a good job of picking sites one can avoid human effects. This is nothing new and you show it very well with your work. Even in sites with less forest and a lower value for the population density through time one still has to work hard to find suitable sites. So, please clarify on this point. We don’t want people being put-off from working on the faults in the Apennines because they gain the false impression that it is impossible to get results due to human degradation of the sites.

We thank the reviewer for this comment. We agree; this problem can be avoided if it is possible to select trenching sites not influenced by human activity. However, our study faced logistical constraints because we primarily focused on capable fault detailed mapping for post-emergency reconstruction microzoning; we were forced to trench fault scarps clearly affected by human activity. We will make this clearer in the revised text. We think we are not overstating the anthropic effects on scarp morphology preservation. We just emphasize the need for properly taking into account human impact on fault scarp erosion and sedimentation, especially  at sites clearly modified by agriculture and human land use.

Line 555 – Please state some values in mm/yr for slip-rates on the faults you mention to prove your claim that they are “similar”.

Ok, we will add published values of postglacial slip-rates for the Monte Morrone Fault (0.20-0.40 mm/yr; Puliti et al., 2024), Paganica Fault (0.25-0.30 mm/yr; Cinti et al., 2011), Montereale Fault (0.30-0.40 mm/yr; Cinti et al., 2018), and Monte Vettore Fault (minimum 0.26-0.38 mm/yr; Cinti et al., 2019). For the Mt. Vettore Fault more recent slip rates estimates are 0.8-1 mm/yr (Puliti et al 2020); 0.7-1.2 mm/yr (based on 36Cl exposure datring of bedrock fault plane; Pousse Beltran et al. 2022); ranging from a minimum of 0.4 to a maximum of 1.3 mm/yr (Galli et al., 2019).

References:

Puliti, I., Pizzi, A., Benedetti, L., Di Domenica, A., & Fleury, J. (2020). Comparing slip distribution of an active fault system at various timescales: Insights for the evolution of the Mt. Vettore‐Mt. Bove fault system in Central Apennines. Tectonics, 39(9), e2020TC006200.

Puliti, I., Benedetti, L., Pizzi, A., Fleury, J., Francescone, M., Guillou, V., & Team, A. (2024). Evidence for a constant slip rate over the last∼ 40 ka along the Mt. Morrone Fault System in Central Apennines. Tectonics, 43(3), e2023TC007871.

Cinti, F. R., Civico, R., Blumetti, A. M., Chiarini, E., La Posta, E., Pantosti, D., Papasodaro, F., Smedile, A., De Martini, P. M., and Villani, F.: Evidence for surface faulting earthquakes on the Montereale fault system (Abruzzi Apennines, central Italy), Tectonics, 37, 2758–2776, https://doi.org/10.1029/2018TC005054, 2018.

Cinti, F. R., De Martini, P. M., Pantosti, D., Baize, S., Smedile, A., Villani, F., et al. (2019). 22‐kyr‐long record of surface faulting along the source of the 30 October 2016 earthquake (central Apennines, Italy), from integrated paleoseismic data sets. Journal of Geophysical Research: Solid Earth, 124. https://doi.org/10.1029/2019JB017757 

Cinti, F. R., Pantosti, D., De Martini, P. M., Pucci, S., Civico, R., Pierdominici, S., and Patera, A.: Evidence for surface faulting events along the Paganica fault prior to the 6 April 2009 L'Aquila earthquake (central Italy), J. Geophys. Res.-Sol. Ea., 116, B7, https://doi.org/10.1029/2010JB007998, 2011.

Pousse-Beltran, L., Benedetti, L., Fleury, J., Boncio, P., Guillou, V., Pace, B., ... & Aster Team. (2022). 36Cl exposure dating of glacial features to constrain the slip rate along the Mt. Vettore Fault (Central Apennines, Italy). Geomorphology, 412, 108302.

Galli, P., Galderisi, A., Peronace, E., Giaccio, B., Hajdas, I., Messina, P., ... & Polpetta, F. (2019). The awakening of the dormant Mount Vettore fault (2016 central Italy earthquake, Mw 6.6): Paleoseismic clues on its millennial silences. Tectonics, 38(2), 687-705.

Line 565 and Figure 15 – It is not very easy to move between the “black-boxes” on Figure 15 and the events and ages on Table 2. Please improve this. In fact, I found Figure 15a quite poor and hard to understand. I had to make my own figure to try to understand the timings of events using the data in the table. I don’t think the “fading colours” are clear in their meaning, e.g. purple to white and dark-green to white, as you have not defined what “white” means. Please use solid symbols (something like error bars on a cross-plot?). I also think the “red” blocks showing the 14C dates are unclear in their meaning, so please use a symbol with the analytical error bars and the calibrated age as a data point (so a data point and error bars). Overall, this Figure is your key result, and its present design makes it hard for the reader to extract the results – as I say I had to try to make my own figure. Please improve this because this is what will improve the citation of your paper. If people don’t understand the figures and hence the paper, they will not cite it.

....

Figure 15b is good for seeing the spatial pattern of clustering, but the text on line 635 implies a pattern of temporal clustering, which I feel is not adequately displayed on Figure 15b. Can you provide a figure that shows the temporal pattern of clustering?

Thanks for pointing this out. Yes, Figure 15a needs to be definitely improved. We will follow your suggestions by using clearer graphical symbology in a simpler chart. Also, we will improve the graphical rendering of the timeline with the proposed earthquakes. We agree that this figure could be the real core of the paper and deserves to be improved.

Line 650 – I agree they look like release faults. However, you don’t need structural inheritance to form these. If there is evidence for pre-existing structures in this specific example, please tell us what the evidence is. If there is no evidence, don’t claim this.

Inheritance is not needed to develop release faults, as you correctly state. On the contrary, release faults are typically much shallower than the main basin-bounding fault and their role is expected to diminish at depth. Inheritance has been invoked here for the consideration that in the Central Apennines faults can be segmented into short faults with considerable slip gradients along strike. Indeed, the recent work by Caportorti and Muraro (2025) showed several examples of Quaternary faults that re-activated at depth previous Miocene normal faults and, at surface, display a fault trace close to coincident to the Miocene ones (e.g., the Norcia, Mt. Boragine, Leonessa and Mt. Marine faults – see Capotorti and Muraro and Supplementary Material of their work for a comparison). The faults bounding the Rieti Basin are located close to inherited rift-related faults, as well. To the west, the basin bounding fault runs almost along the inherited Sabina Paleofault (Galluzzo and Santantonio, 2002). To the east, the main basin-bounding fault is running at the margin of a series of aligned Mesozoic structural highs that (i.e., the so-called pelagic Carbonate Platforms - PCP; specifically, the Lisciano, Mt. Rosato and Polino ones; Capotorti and Muraro, 2024). These have already been interpreted as being related to a secondary structural high in the hanging wall of the Sabina paleofault (see Galluzzo and Santantonio, 2002, their Figure 30).  We will add this specific explanation in the text to make it clearer.

Line 655 – Have you conducted a stress transfer calculation, or are you speculating? It might be worth doing a stress transfer calculation to make sure you are correct. After all, the value of stress on the receiver fault is influenced by the dip and strike of the fault, and the depth considered, as well as its position relative to the ruptured fault.

Some models have been run to test the hypothesis. Under very simple assumptions and boundary conditions, given a simple geometry of the modeled faults, we observed that it is the along-strike slip distribution on the eastern border fault that primarily determines the CSS on the receiving faults. In the attached pdf file, you can check the results from the CSS modeling. Northern and southern boundary faults can be slightly loaded in the shallowest sectors, by the movement of the eastern border fault. In particular, the simulation of a slip event like the D earthquake we uncovered through paleoseismology approaches is slightly charging the southern border fault that consistently showed evidence for a contemporary movement.

We do not want to discuss in detail the CSS modeling in this work; rather, we prefer to investigate these results in a future companion paper that can specifically address this issue.

Line 658 – You say they are not compatible with a single phase of extension, but they are compatible if you believe they are release faults (which you say a few lines earlier). I suggest removing the “hardly compatible” phrase as it contradicts your earlier claim of release faults (also this is incorrect English – it is either compatible or incompatible – “hardly compatible” is not correct English).

 Thanks for pointing this out. Yes, indeed, we were referring to a simple Andersonian model of faulting under tectonics. Nonetheless, we will change it as suggested.

Line 660 – why not cite a paper on the 1983 Borah peak ruptures where ruptures fanned out at the NW end onto the Willow Creek Hills at a relatively high angle to the main fault. This supports what you claim.

 Thanks for the suggestion. Yes, indeed, we’ll add the Borah Peak 1983 earthquake as an additional example.

Line 654 – change the word “descent” - incorrect English. “originate”?

Text will be changed as suggested.

On Figures where you show the “vertical offset”, e.g. Figure 3, please state what feature is offset. I presume it is the topographic slope, so I also presume that this is the vertical offset across the topographic scarp formed after the slopes stabilized. However, I don’t think you have stated this, so I am not sure. Would you like to comment on how the magnitude of slip in the trenches relates to the “vertical offset” on the profile? 

Thanks to Gerald Roberts for this detail. The offset feature is the topographic slope in the scarp footwall, as presumed by the reviewer. This footwall slope is carved in the marly limestone Scaglia Bianca Formation; the fault juxtaposes Scaglia Bianca Formation against the Villafranchian lacustrine clay and silt deposits. The trench site is on the Southern border of the Apoleggia village. Due to agricultural impact, the sedimentation of slope deposits at this site was quite relevant during historical times, as demonstrated by the age of colluvial deposits in the fault hanging wall. Therefore, in Figure 3 the 5.2 m vertical offset is calculated by projecting the topographic slope in the footwall below the thickness of historical slope deposits. In fact, ERT profiles clearly show the position of the top of the Pleistocene lacustrine deposits, which is parallel to the footwall slope and can be regarded as a proxy for the glacial topographic slope. Therefore, at this site the scarp height is lower than the vertical postglacial fault offset.

We follow the same approach also for the Cantalice scarp, where the footwall bedrock is in the same Scaglia Bianca Formation, and the hanging wall is in the Villafranchian lacustrine deposits; the human impact on slope deposits is similar, based on dating of hanging wall samples collected in the trenches, and ERT profiles show a similar geometry for the top of Pleistocene lacustrine deposits and glacial slope. We will add this text in the discussion.

On Figures the photos are in places “greyed-out” and covered by an interpretation. Is it possible to provide the un-interpreted photos as well, perhaps in an Appendix?

     Thanks for the suggestion. Yes, We will provide an uninterpreted photomosaic of each trench wall in the Supplementary Material.

On Figure 7 please change the text “unit is truncated, to the top, by an erosive surface – Bedrock” That is obviously incorrect English and does not state the field relationship precisely or correctly. It is not "to the top".

               Thanks for pointing this out. We’ll correct the text.

On Figure 9, there is mention of slickensides on unit 12 – do you have any measurements of the slip vector to show if these faults are dip-slip or oblique slip?

In this case, the slickensides are a common characteristic of vertisols, and typical of clay soils subject to mass movements or vertisification phenomena. They are not indicators of kinematics.

The graphics on the inset ERT plot for Figure 11 are poorly quality (pixellated) – can you improve the resolution of this figure?

We improved the quality of Figure 11 but we cannot further improve the resolution of the enlargement of the ERT profile. We underscore that this is only a graphical limitation, and it does not change the resulting interpretation in any way.

Figure 13 is hard to visualize the context of the image. Can you provide a context photo to show the aspect of the slope?

Thanks for the suggestion. We have updated the image by adding a new photo showing the local morphological setting.

Overall, I enjoyed the paper and found it quite inspiring. Well done.

Professor Gerald Roberts, 1st July 2025

This is a very detailed study which plugs a number of significant gaps in knowledge.  To their credit, the authors have themselves identified limitations of their work.    One of the most crucial seismic hazard parameters is the maximum magnitude. The authors conclude as follows:

Our results indicate that the maximum credible earthquake in the Rieti Basin is in the order of magnitude Mw 6.5, which is consistent with the general setting of the Central Apennines. Given the resolution of chronological constraints obtainable with radiocarbon dating techniques in paleoseismic trenches, we cannot disentangle the occurrence of a single earthquake as compared to multiple earthquakes occurring over a short time interval (like the 2016 seismic sequence). Additionally, paleoseismic data inherently focus on surface-rupturing earthquakes, thus aliasing smaller seismic events, which however could have caused significant damage.

Given the ambiguity over the release of seismic energy in one large event, or a sequence of lesser events, it would be helpful if the authors could provide more substantive discussion over the maximum magnitude in the Rieti Basin. In particular, could the maximum magnitude be as high as 6.7 or 6.8?

We wish to thank the reviewer Gordon Woo for the positive feedback and for taking the time to provide improvements to our text. Here we provide a point-by-point answer to all the comments raised by the reviewer. Original comments are shown in plain text, while answers are in italics.

The revised manuscript will be uploaded in the journal system.

We thank the referee for the useful comment. We did not discuss in detail the maximum earthquake magnitude estimate. We will add the text below to the discussions. We derive constraints on Maximum Magnitude in the Rieti Basin from historical seismicity and earthquake surface rupture length. Epicentral intensity data on historical earthquakes, such as those that occurred on 76 BCE and 1298 CE, show Io = X MCS (Mercalli Cancani Sieberg scale) in the Italian seismic catalogue. Using Magnitude Vs. Intensity correlations this gives M6.40 for the 76 BCE earthquake (Guidoboni et al. 2018, 2019), and M6.26 for the 1298 earthquake (Brunamonte et al., 1993; Rovida et al., 2022). Data on similar ancient historical events are obviously not very detailed. However, they provide a reasonable minimum value.

As for estimates based on earthquake surface rupture length, based on our extensive coverage of paleoseismic trench sites we assume that rupture of the full 21 km length of the Rieti Basin eastern master fault is a credible hypothesis. Using literature empirical relations (e.g., Wells and Coppersmith, 1994; Pavlides and Caputo, 2004) this gives a value of ca. Mw 6.5. Since standard deviations of published empirical relations is ca. 0.3, we cannot rule out a Mw max of 6.8. However, comparison with well-studied recent earthquakes in the Central Apennines indicates that Mw 6.5 is a reasonable estimate. We believe that uncertainty in available data does not allow a more detailed numerical analysis, which is beyond the scope of the manuscript.

This manuscript presents a thorough paleoseismological investigation of the Rieti Basin in the Central Apennines. Through an extensive trenching involving 17 excavation sites along active fault segments, the authors successfully identified 15 paleoearthquakes during the last ca. 20 kyr. The scope and resolution of the work are impressive, demonstrating a significant investment in fieldwork, stratigraphic analysis, and chronological interpretation. While the study is of high quality, addressing a few specific issues could further enhance its clarity and impact.

We wish to thank the reviewer for the positive feedback and for taking the time to provide improvements to our text. Here we provide a point-by-point answer to all the comments raised by the reviewer. Original comments are shown in plain text, while answers are in italic.

The revised manuscript will be uploaded in the journal system.

1. Figure 2 contains only three subfigures, but the caption refers to a subfigure “d)”, which appears to be an error and should be corrected. In addition, the caption mentions a star symbol, but it is not visible in Figure 2a; the authors should ensure that all referenced symbols are clearly displayed. Lastly, in Figure 2b, the GPR17 and GPR18 survey lines are difficult to distinguish.

Thanks for spotting the errors. We will amend the caption as follows:

“Figure 2: Study Area along the Northern Border fault: a) simplified geological map; b) and c) panels display detailed views on the two studied Sectors with the traces of the capable faults, of the geophysical investigations and the footprints of the excavated paleoseismological trenches; red dot in Fig. 2a is the Piedicolle trench site from Michetti et al. (1995).”

Finally, we will improve the graphics of GPR17 and GPR18 survey lines to enhance                           readability.

1. On page 13, line 220, Figure 3d does not include the ERT_14 section. Therefore, the sentence “Figure 3d shows the ERT_14 section (see location in Fig. 2b);” should be corrected to “Figure 3 shows the ERT_14 section (see location in Fig. 2b);”.

Thanks for spotting this error. We will change it accordingly.

1. Please consider indicating the fault in Figure 3d (ERT_15 result) to help clarify the relationship between resistivity anomalies and fault structures. Additionally, the main text does not provide a detailed interpretation of this figure. The absolute error associated with the Wenner–Schlumberger array is smaller than that of the dipole–dipole array (Fig. 13d). Could this be interpreted as an indication that the Wenner–Schlumberger result is more reliable, or might it instead suggest that the inversion result is overfitting the observed data?

Yes, we will add the fault interpretation on Figure 3d. As for the errors in the two arrays: In comparative studies, Wenner-Schlumberger arrays have shown better stability and lower RMS errors during inversion, particularly in environments with moderate to low resistivity contrasts. Dipole-Dipole performs better in directly imaging possible fault planes, although it is noisier. Nonetheless, both the inversions show good statistical fitting. Additionally, the software we used for ERT inversion (REs2dInv) offers a robust inversion method that minimizes the absolute rather than squared differences. Its application is less sensitive to outliers and helps prevent overfitting to noisy data.

1. On page 16, “CAM_ERT_02” should be corrected to “CAM_ERT_17” to ensure consistency with the context.

Agreed and modified accordingly.

1. The fault zone in ERT17 (Figure 4) shows a weaker resistivity contrast than in ERT14 (Figure 3). It would be helpful if the authors could clarify whether this difference arises from subsurface lithological variations, differences in survey configuration, or other influencing factors, as this would offer useful guidance for ERT applications in fault zone imaging.

Thanks for pointing this out. The two ERT profiles are reaching different lithologies. In ERT 14, bedrock is outcropping just next to the ERT line and the very high resistivity values correlated with a fractured carbonate bedrock. In Section ERT17, instead, the Villafranchian succession is imaged (i.e., pedogenized fine sands and silts). Possibly, bedrock is imaged only in the lowermost part of the section, but due to the very high uncertainty characterizing that sector of the tomography at depth, we avoided presenting direct interpretations.