The manuscript presented by Felix et al. evaluates the tsunami hazard associated with a potential rupture of the Flores backarc thrust along the Lombok Strait. The manuscript is well written, and properly organized, and figures are legible and of good quality. While modelling in this work uses higher resolution bathymetric and topographic data than previous studies, the following aspects need to be clarified/improved before considering this work for publication:

• The fault model is mostly interpreted on the basis of 2D seismic sections, earthquake location, and seafloor morphology (which is not clearly visible from the current figures). Yet, the region used to rupture is > 100 km wide, so one would appreciate few words regarding how potential along-strike structural variations could affect modelling results.
• A close up of the bathymetry from Fig.1a would be needed along the Lombok Strait and North Lombok, to properly see that the inferred ramp in Fig. 1a is supported by morphological variations of the seafloor.
• The potential rupture area spans along the entire Lombok Strait, yet, the 2018 sequence did not rupture through the region of the Strait, its westernmost sequence localized North Lombok (during the 5^th of August) according to Lythgoe et al. (2021). So I wonder how realistic is the proposed scenarios in which the entire Lombok Strait ruptures at the same time?
• The following comment concerns lines 259 to 267. The conversion to Mw from Slip assumes a standard rigidity of 30GPa (Table 1), thus assuming that rigidity is constant along the entire rupture area. In the last few years, it has become clear that rigidity of rocks overlying the fault in the upper-plate decreases trenchwards up to values of < 5 GPa in megathrust regions worldwide. Most importantly, it has been demonstrated that this variation determines shallower larger slip, longer duration and depletion of high-frequencies in the shallow thrust, conditioning tsunami wave height (Sallarès & Ranero, 2019 1038/s41586-019-1784-0; Prada et al., 2021; 10.1029/2021JB022328). Realistic rigidity variations in turn, allowed to explain the rupture of particular megathrust events (Sallarès et al., 2021 10.1126/sciadv.abg8659). Based on this, the authors should assume realistic rigidity variations with depth to provide more accurate values of Mw, assuming a constant rigidity is likely to result in incorrect estimates of Mw and slip along the rupture area (which leads me to my final comment).
• It is not clear to me what is the rationale behind the choice of slip values (1, 3, and 5 m). If the authors take these values based on previous estimates they should explain it. Alternatively, if the authors estimated the different slip values from Mw of previous events, they should include rigidity values used to perform the conversion. If the authors assumed a constant rigidity (as they did to convert slip into Mw), it is quite likely that they are underestimating the amount of slip that the shallow thrust may generate, and thus, the amount of uplift, and tsunami wave height. Additionally, overestimation of rigidity may also result in overestimation of tsunami arrival times, given that the rupture is likely to propagate much slower in the updip region than in the downdip.

The manuscript studies potential tsunami hazards in Indonesia for the east coast of Mataram (Lombok) and west coast of Denpasar (Bali). Authors apply a simple deterministic tsunami modelling to assess the tsunami hazards for those two study areas. Even tough simple, the authors utilised the most up-to-date tectonic knowledge from this region. Therefore, I see that this manuscript would provide new knowledge of tsunami hazards for this region.

The manuscript also provides short but complete summary on the regional tectonic setting, seismicity, and historical events from the past. Moreover, the authors provide a much more than adequate detail on how they build or set-up the tsunami model until they analyse the results and draw conclusion. I see that this manuscript potentially also acts as a guideline for other researchers.

I general, I recommend this manuscript to be published at NHESS after some clarification as follow:

Further clarification:

• [Section 2.1 Faul model setup] Authors fixed the dip angle of its fault plane model to 25°. based on the study by Lythgoe et al. (2021). Whereas it is written “26±8°” in Section 1.2 Seimicity of the Flores Thrust and there is some degree of uncertainty from the study by Lythgoe et al. (2021). Could you please qualitatively analyse what will be the impact for the tsunami model?
• [Section 2.4 Tsunami modelling using COMCOT] 
  • Second paragraph, it needs many reads to understand on how the authors set up the nested grid. Please clarify it. Adding the resolution on Fig 6 would help.
  • It seems that the largest domain model (L1) does not cover all the initial sea surface deformation used. Further the L5 layer is very close to the SW corner of the boundary model leave only a few km distance from the edge. Even tough the model works fine, it is uncommon to set the targeted area (L5) like in this manuscript. The wave might be affected by the boundary domain model. Moreover, peninsula of southern west part of Lombok (about 20-30 km south of Mataram) is not fully covered by the largest domain model. It might affect on how the tsunami wave propagate from the source to the target area. I recommend to enlarge the most-outer domain model (L1) so that can fully cover the initial sea surface deformation model and to show the result outside the targeted area. Some readers might interest to see the result, for example, for all Bali and Lombok coastline.
• [Section 3.3 Inundation in Mataram] A digital surface model is used from modelling the tsunami. Could you please qualitatively analyse what will be the impact if a digital terrain model is used? 

Suggestions:

• I recommend to assign a ‘name’ for each scenarios so that it would be easier to read the manuscript. For example, Fault Model A with 1 m slip = A-1; Model A with 3 m slip = A-3; etc. Then use these in the text.
• Line 395, please refer “headlands at 8.38°s” to a figure.
• Table 1. Please provide full parameters of these scenarios so that can be reproduced by other researchers. Parameters required such as: fault plane coordinate (centroid or a coordinate of one corner, depth), fault mechanism (strike, dip, rake).
• Figure 1. Please provide one (or two) reference(s) for the white squares.
• Figure 9 caption, please remind reader that these are results from L4.
• Figure 11. Please add a vertical line to indicate the estimated time arrival.
• Figure 11 caption, “Sea surface deformation generated by ...”, do you mean sea water elevation?
• Figure 12. Please use the same format as other figures on how you show the coordinate.

General comments

This manuscript presents a series of tsunami scenarios in Lombok and Bali, Indonesia. The tsunamis are modeled as resulting from prospective earthquakes generated by the Flores back-arc thrust, a south-dipping thrust in the upper plate of the Java subduction zone. Tsunamis generated by other than megathrust earthquakes pose serious threats and are worth being investigated. Felix et al. make a good case of the Flores thrust as it may represent a potential threat to the Indonesian Islands as concerns should have been raised after the recent Lombok earthquake of 2018. Their work is thus of great interest and timely. However, I'm afraid that at the moment, the work suffers from a few weaknesses that need to be addressed before publication.

First of all, I am afraid that the term "hazard" is a bit abused on several occasions (e.g., title, introduction line 47) so that the presented work does not meet the reader's expectations. A hazard assessment should incorporate the likelihood for future events and their consequences to happen and usually consider a large number of possible events. Instead, what has been done in this work, is the simple exploration of six tsunami scenarios.

Most importantly, however, I have concerns about the design of the earthquake rupture scenarios. Of the six rupture scenarios, only in one case (Model A/b) the dimensions (length, width, slip and magnitude) are rather consistent with one another, following fault scaling relations. Deviations from fault scaling relations are expected but should be explored systematically to estimate the variability of the resulting tsunami impacts and their usefulness for understanding the hazard. The presented earthquake scenarios would hardly become a reference for hazard analyses or a benchmark for future in-depth studies in their present form. Also, the rupture scenarios are very simplistic (planar fault and uniform slip), and this approach's limitations are not at all discussed.

Recommendations

I strongly recommend redesigning the earthquake rupture scenarios in the following way. 1) Select a range of earthquake magnitudes to explore. 2) Derive fault rupture dimensions (length, width, slip) from appropriate scaling relationships (Leonard, 2010, 2014; Thingbaijam et al., 2017). 3) Depending on your resources and objective, try different realizations around those values to explore the variability of the rupture parameters or discuss the possible implications of the variability. The occurrence of ruptures of the same size in different positions can also be explored. To explore the case of enhanced slip in the upper part of the ramp (Model B), which is worth considering, subdivide the rupture in discrete patches and apply variable slip values by conserving the mean. 4) Perform the tsunami simulations.

The use of planar fault ruptures with homogeneous slip should be then discussed in light of the potential outcomes or more realistic rupture scenarios involving a three-dimensional fault representation and heterogeneous slip distribution (see Serra et al., 2021). Is the fault model geologically well-constrained (see statement at line 205) to design more complex rupture scenarios? A justification is needed in this respect.

Below I add several specific comments and suggestions (sorted by line number) to improve the text up to the beginning of Chapter 3. The analysis and discussion about the tsunami simulations should be reconsidered in light of the new results.

L4 versus L41 and all other occurrences: backarc or back-arc? I prefer the hyphenated spelling. In any case, please make a choice and stick to it.

L12-36: It would be better to rewrite the abstract without citations.

L40-45: It is unclear whether the focus is on studies on back-thrust faulting earthquakes or hazard studies incorporating seismic sources other than subduction megathrusts. Depending on the collected data, studies on specific events became available as the events occurred. It is true instead that many hazard analyses focus mainly on megathrusts, but hazard studies involving crustal earthquakes are not so rare. However, these are not the cases in the cited works. I suggest reviewing this part of the introduction by seeking inspiration from recent review papers on tsunami hazards that address this circumstance quite clearly (Behrens et al., 2021; Grezio et al., 2017).

L88-89: How were the 29 earthquakes attributed to the Flores Thrust? Please specify if it is just because they occurred in the vicinity of the thrust or if there was a more detailed analysis.

L94: Replace "This fault system has also produced uplift on its hanging wall." with "The activity of this fault system is also testified by uplift recorded on its hanging wall."

L144-145: What do you mean by "observational window"? If it refers to the historical records, the observational window is short almost anywhere (Geist & Parsons, 2006; Grezio et al., 2017). That is why we rely on paleoseismological studies and other inferences on the long-term behavior of faults. Please elaborate on this statement.

L172: Why "unrealistically large"? Please give your justification.

L179-185: I am afraid the Horspool et al. (2014) paper has been misunderstood. The model's description for the Flores thrust is the "unit sources" that are then linearly combined to form earthquake sources of all magnitudes in a magnitude-frequency distribution up to the maximum magnitudes reported in Table 1. The use of unit sources is part of a technique widely used in seismic hazard studies to save computational time. Also, what do you mean by "return period on the fault"? A return period is a quantity selected on a hazard curve for a site, it is not a property of the fault. Please reconsider all this paragraph and make sure to have understood what Horspool and coauthors did.

L266-267: In this statement, the cited relations do not apply. Wells and Coppersmith (1994) consider fault displacement at the surface, and only indirectly can one extrapolate the coseismic slip at depth. Hanks (2002), which is Hanks and Bakun (2002) in reality, and Hanks and Bakun (2008) focus on strike-slip earthquake ruptures, not thrusts. Biasi and Weldon's (2006) relation is about surface rupture length, not area. More recent and more appropriate scaling relations exist (Leonard, 2010, 2014; Thingbaijam et al., 2017), which would help reconsider this statement.

L364-365: The worst-case rupture scenario does not uniquely yield the worst-case tsunami scenario at a given location (Salaree et al., 2021). Various techniques exist (Lorito et al., 2015; Volpe et al., 2019) to reduce the computational burden of inundation modeling. Please reconsider your approach or at least discuss its limitations and potential pitfalls.

L385-387: Was any filter (Kajiura, 1963) applied to transfer the sea-bottom dislocation to the water surface? Please explain.

References (if not already cited in the manuscript)

Behrens, J., Løvholt, F., Jalayer, F., Lorito, S., Salgado-Gálvez, M. A., Sørensen, M., et al. (2021). Probabilistic Tsunami Hazard and Risk Analysis: A Review of Research Gaps. Frontiers in Earth Science, 9, 628772. https://doi.org/10.3389/feart.2021.628772

Geist, E. L., & Parsons, T. (2006). Probabilistic Analysis of Tsunami Hazards*. Natural Hazards, 37(3), 277–314. https://doi.org/10.1007/s11069-005-4646-z

Grezio, A., Babeyko, A., Baptista, M. A., Behrens, J., Costa, A., Davies, G., et al. (2017). Probabilistic Tsunami Hazard Analysis: Multiple Sources and Global Applications. Reviews of Geophysics, 55(4), 1158–1198. https://doi.org/10.1002/2017RG000579

Kajiura, K. (1963). The leading wave of a tsunami. Bull. Earthq. Res. Inst., 41, 535–571.

Leonard, M. (2010). Earthquake Fault Scaling: Self-Consistent Relating of Rupture Length, Width, Average Displacement, and Moment Release. Bulletin of the Seismological Society of America, 100(5A), 1971–1988. https://doi.org/10.1785/0120090189

Leonard, M. (2014). Self-Consistent Earthquake Fault-Scaling Relations: Update and Extension to Stable Continental Strike-Slip Faults. Bulletin of the Seismological Society of America, 104(6), 2953–2965. https://doi.org/10.1785/0120140087

Lorito, S., Selva, J., Basili, R., Romano, F., Tiberti, M. M., & Piatanesi, A. (2015). Probabilistic hazard for seismically induced tsunamis: accuracy and feasibility of inundation maps. Geophysical Journal International, 200(1), 574–588. https://doi.org/10.1093/gji/ggu408

Salaree, A., Huang, Y., Ramos, M. D., & Stein, S. (2021). Relative Tsunami Hazard From Segments of Cascadia Subduction Zone For M  w  7.5–9.2 Earthquakes. Geophysical Research Letters, 48(16). https://doi.org/10.1029/2021GL094174

Serra, C. S., Martínez‐Loriente, S., Gràcia, E., Urgeles, R., Gómez de la Peña, L., Maesano, F. E., et al. (2021). Sensitivity of Tsunami Scenarios to Complex Fault Geometry and Heterogeneous Slip Distribution: Case‐Studies for SW Iberia and NW Morocco. Journal of Geophysical Research: Solid Earth, 126(10), e2021JB022127. https://doi.org/10.1029/2021JB022127

Thingbaijam, K. K. S., Martin Mai, P., & Goda, K. (2017). New Empirical Earthquake Source‐Scaling Laws. Bulletin of the Seismological Society of America, 107(5), 2225–2246. https://doi.org/10.1785/0120170017

Volpe, M., Lorito, S., Selva, J., Tonini, R., Romano, F., & Brizuela, B. (2019). From regional to local SPTHA: efficient computation of probabilistic tsunami inundation maps addressing near-field sources. Natural Hazards and Earth System Sciences, 19(3), 455–469. https://doi.org/10.5194/nhess-19-455-2019