Dear Anonymous reviewer:

Many thanks for your thorough and dedicated review of our paper. We are sure that your observations contribute to a better explanation of the ideas behind this research contribution.

In the following we present our answers to your questions and observations. Every answer follows the specific question/observation, in bold italics, and a direct reference to the modifications made in the text.

Best regard,

Gonzalo Yanez

General comments



The paper entitled: On the role of Trans-Lithospheric Faults in the long-term seismotectonic segmentation of active margins: a case study in the Andes by the authors Gonzalo Yáñez, José Piquer and Orlando Rivera, seeks to establish the hypothesis that the large structures called Trans-Lithospheric Faults recognized in the active continental margin of Chile, could have an influence in the seismotectonic segmentation of large subduction earthquake ruptures, because these structures would be able to transport and contribute an important amount of fluids to the subduction zone, producing a creeping zone surrounded in a more coupled zone. To prove this, the authors establish spatial relationships with different observations and factors determined at the margin among them are: historical seismicity, distance between the trench and the continent, coupling models and Pearson correlation parameters. Although it is a novel hypothesis and the manuscript is clear and well written, there are certain aspects that are not clear to me both in the writing, the postulated and the Figures presented that in my opinion are necessary and I request to improve the article. These aspects are specified below.



Specific comments



In lines 106-110 of the manuscript, it is explained how Trans-Lithospheric Faults (TLF) have been defined through several observations. One of these aspects you point out is the seismicity associated with this type of structures, with which we could have an idea of the depth that these structures reach. However, I am very surprised that in Figure 1 (introductory) none of the TLFs have associated seismicity. This is why I ask that in Figure 1 they incorporate a panel B showing the cortical seismicity associated with this type of structures. In the manuscript they indicate that thanks to temporal networks it has been possible to detect seismicity, therefore, it seems to me relevant to incorporate in Figure 1 a panel B showing this seismicity. Showing this seismicity associated with these faults is something powerful that would undoubtedly help to improve the quality of the article.

We agree with the reviewer on the great value of having seismicity directly associated with TLF, but this is not the case, most likely due to their large recurrence time, in the time frame of thousand years . The focus of the paper is the seismicity in the subduction plane, in other papers, like Piquer et al., 2019, we discuss the few evidences of seismicity linked to TLF, but not enough evidences to populate a panel in Figure 1. Although indirect evidences of activity related with ETL is presented in Figure 3a, in the cumulative inter-seismic activity, in particular during seismic swarms, and the normal event of March 11, 2010 at the Pichilemu TLF (22) (linked to the Maule 8.8 Mw event of 2010).

On the other hand, a doubt: ¿are TLFs restricted in depth and spatially to the continental upper crust or can they also partly affect the oceanic crust? Please make this clear when introducing TLFs in the manuscript (lines 106-110).

We don’t know in detail the TLF behaviour with depth, from the geological and geophysical evidences that show the alignment of magmatic and hydrothermal activity we are confident that they involve the whole lithosphere. We have no evidence of a prolongation towards the oceanic crust below the Benioff plane, most likely is not the case due to the creep nature of the process postulated for the interaction of TLF and plate coupling. We added a sentence in the paragraph to clarify this point:    

“The geometry and depth extension of TLF is unknown, but based on their control of continental-scale magmatic and hydrothermal processes and their surface traces in the order of hundreds of kms, we consider that they involve, exclusively, the whole lithosphere”.    



2.- In Figure 2, it strikes me that the Iquique 2014, Tocopilla 2007 and Antofagasta 1995 earthquakes do not follow the hypothesis put forward in the article. In these earthquakes the zone of greater slip or roughness, is just located in the trace of the TLF recognized in this place and not so in the earthquakes of the south, where if the postulated by you in the article is fulfilled, ¿how can I explain this difference between the earthquakes of the north and the south with respect to your hypothesis? Please deepen this through a deeper discussion.

We partially agree with the reviewer observation, for the case of Iquique 2014 event, Iquique TLF (4) is cutting the slip zone however the offshore extension of this TLF is not well resolved (in the seaward extrapolation we use bathymetric morphology as the principal guide), if it continues straight from the landward side, most of the slip zone would be to the south of TLF 4. For the case of Tocopilla 2007 half of the slip zone is outside the slip zone definition. Finally, for the case of the Antofagasta 1995 event, we totally agree with the reviewer observation, the slip zone is indeed cut by two TLF (7: Agua Verde-Exploradora, and 8: Antofagasta-Chonchi). Thus, in these particular cases against the model prediction, we envision two possible explanations for this lack of consistency: (1) the fact that this is a low magnitude event (8 Mw) compared to the other cases, and or (2) not all TLF behave as barriers. We add a discussion of this particular lack of consistency in point 6 of the discussion section 3.1 as follows:

“The most conspicuous case against the rule is the slip zone of the Antofagasta 1995 that cut two TLF (7: Agua Verde-Exploradora, and 8: Antofagasta-Chonchi) and partially the Tocopilla 2007 event (Mejillones-Llullaillaco TLF 6). Two complementary explanations are proposed in this case: (1) both are small events (8Mw) compared to the other megathrust events, (2) not necessarily all TLF behave as barriers all the time. For the case pf Iquique 2014 event, the seaward extension of of Iquique TLF is not well constrained, and most likely run straight from landward segment, leaving the slip zone entirely to the south of TLF 4.  .”

Line 305: although the coupling models indicated are good, there are new models published especially in the segment between Antofagasta and Copiapo. I recommend perhaps updating the models of this article with the most recent models published and incorporating to the references of these articles: Yáñez-Cuadra et al., 2022 (Geophysical Research Letters) and González-Vidal et al., 2023 (Geophysical Research Letters).

Thanks for providing these new references. Looking at the new coupling models derived from GPS observations as shown in these two papers we noticed that results do not depart significantly with the model presented in Figure 4b, and for the large-scale purpose of our research is not adding more information, so we decided to keep the original GPS coupling. But we add a sentence in section 2.7, explaining that the new GPS models in the northern Chile region are consistent with the GPS model used in the paper: 

“For the segment between Antofagasta and Copiapo (24-28°S), two new GPS plate coupling models are available (Yáñez-Cuadra et al., (2022) and González-Vidal et al., (2023)), however, we noticed that these new results share similarities with the model presented in Figure 4b, and is therefore not necessarily  included in this case.”

In lines 453-458 it is explained that at 25° and 30°S there is a potential barrier zone due to the high correlation of the Pearson index. However, these zones also coincide with the Taltal ridge subduction at 25°S (León-Rios et al., 2024 G3) and the Challenger Fracture zone at 30°S (Poli et al., 2017 Geology; Maksymowicz, 2015 Tectonophysics). In that sense, further discussion of this correlation is lacking in the manuscript. Please discuss these points, as, while there is a spatial correlation between these barrier zones with TLFs, there is also correlation with other important bathymetric structures, which can either carry a significant amount of fluids or produce a considerable degree of fracturing, enhancing creeping seismogenic behaviour. Incorporate a deeper discussion considering other possibilities to the correlations you find, i.e., incorporate to the article that, although you find a correlation between TLFs and creeping barrier zones, this would not be the only possibility. When improving this discussion, please incorporate the references mentioned above.

We acknowledge the fact that other features associated with the oceanic Nazca plate, like aseismic ridges, and fracture zones can carry large volumes of fluids that can also enhance the fluid pressure at the Wadatti-Benioff zone acting in complementary fashion with the proposed mechanism. We include a new paragraph at this regard in the discussion section 3.2.:

“Our proposed conceptual model in which TLF’s promote the development of barrier domains along the subducting margin through the enhancement of fluid pressure complement other process at subduction zones that also enhances the budget of localized fluids at the plate contact, among them the collision of aseismic ridges and fracture zones, bending of the subducting plate (e.g. Ranero et al., 2008, Ranero et al., 2005, Martinez-Loriente et al., 2019; Arai et al., 2024). In the Nazca-South America plate interaction authors had highlighted this increase in fluids at passive ridges such as the Taltal ridge 33°S (Leon-Rios et al., 2014) and the Juan Fernandez ridge 33.5°S (Garrido et al., 2002), and fracture zones such as the Challenger Fracture zone 30°S (Poli et al., 2017; Maksymowicz, 2015). The volume of fluids in aseismic ridges is enhanced by oceanic water percolation along the thicker oceanic crust, while in fracture zones as a result of the high permeability that provides a mechanism to increase water storage prior to subduction.  These complementary mechanisms share a common origin at the subducting plate, and in the particular case of the Nazca plate they are oblique to the margin (roughly NE).  Thus, the main difference with the proposed model is their along strike migration with time, while in the proposed mechanism TLF belongs to the overriding plate.”

Specific comments for Figures



Figure 2:



In panel A, the symbology used of gray lines indicating magnitude is very confusing and not well understood. Although it may be useful for higher magnitude earthquakes, for magnitude 7 events the line is too thin and cannot be identified well in the Figure. On the other hand, the word magnitude is in Spanish and not in English.

Figure 2a corrected and improved in terms of the visibility of small events (making to black the magnitude legend, and putting lighter the topo/bathymetry background)



The caption of the Figure is incomplete and is not in tune with what is written in the manuscript. The segmentation says that it is marked by semitransparent yellow ribbons when in fact they are pink.

Caption corrected



In panel B, please point out to which earthquake (earthquake name) each slip patch corresponds. There may be readers who are not familiar with Chile's earthquakes, so indicating or pointing out each earthquake in the Figure (panel B) may be helpful to readers.

Included the names of the major events in panel B



I recommend improving or rewriting the caption of this Figure to be more precise in the information provided.

Caption redaction improved

Figure 3:



It is missing to indicate in the caption that the seismicity was extracted from the National Seismological Center.

Included

I think there is an error in indicating the 2015 earthquake as "Vallenar 2015" in the caption, is it not the Illapel earthquake of 2015? I have no recollection of a Vallenar earthquake in that year.

Modified



Incorporate the abbreviation DTC in panel B, it could be indicated on the color scale indicating distance.

Included

In general, I recommend rewriting or rephrasing all the captions of the Figures as well as the wording of these. As they are written they give very little information and are inaccurate. They could definitely be much better.

Most of the captions have been improved, with a more complete description of each figure panel.



Figure 7



Enlarge the letters of the symbology

Legend corrected



Technical corrections



Line 23: specify in a better way what type of observations are referred to, these can be seismotectonic, seismological, geodetic...etc.

To keep this sentence of the abstract succinct, we include the end members only:

 “We tested this hypothesis against key short- and long-term observations in the study area, seismological, geodetic, and geological, obtaining consistent results.”

Line 44: take out "including the development of asperities and barriers in the same spatial and time frame".

Removed

Lines 49 to 51: In this part it seems necessary to include Scholz's reference that indicates these different landslide states.

Added

Line 67: add reference Moreno et al., 2014 Nature Geoscience.

Added

Line 81: Hayes et al., 2018? Or just Hayes, 2018? In this publication it is not just Hayes, 2018, it is Hayes et al., 2018.

The reference is indeed Hayes 2018:

Hayes, G. (2018). Slab2 - A Comprehensive Subduction Zone Geometry Model [Data set]. U.S. Geological Survey. https://doi.org/10.5066/F7PV6JNV

Line 82: Yanez to Yañez et al., 1988.

Corrected

Line 152: Add reference Calle-Gardella et al., 2021 Journal of Seismology.

Added, thanks

Lines 196-199: this sentence is confusing, please rewrite or rephrase.

Rephrased and separate in two sentences, the new paragraph reads as follows:

 “For the present analysis, we define seven domains from north to south; the boundary between domains is defined by a region of roughly 100-200 kilometres that represents the uncertainty in the rupture length of the major events. We consider wider boundaries for the cases of lacking information, in particular in the northern area where the historic record is scarce.” 

Line 209: Vi to VI

Corrected



Line 219: Magnitude Mw 9.3 What reference determines this magnitude? Please incorporate reference or change the magnitude.

Corrected to 9.5 Mw

Line 237: remove double parenthesis in "Omori's Law".

Corrected

Dear Professor Booth Rea

We do appreciate your comments and observations regarding this work, and in particular your concerns on the evidences that support the existence of a large number of TLF cutting the Andean margin (36 in total according to Table 1). This discussion provides the opportunity for us to reinforce the relevance of TLF in the geological evolution of the Andes since at least Mesozoic times.

First, we would like to clarify, as described in table 1, that at least 36 publications support the occurrence of 36 TLF, and only 11 TLF are taken from the work of Yanez and Rivera (2019) (30%), with 3 of them coming from previous studies (see Table 1 of the paper). Thus, only 22% of the TLF were defined in Yanez and Rivera (2019), as a result of an intensive work of geological and mining exploration done by Codelco (Chilean coper mine company) and collaborators from the academia in northern Chile, interpreting the results for more than a decade,most of this work unpublished before the synthesis presented by Yanez and Rivera (2019) paper. The remaining 70% of TLF’s are based on studies carried out by several and independent researchers using different mechanisms (structural geology, potential methods, seismology, indirect evidences of highly permeable zones such as volcanic and ore deposits alignments, high Vp/Vs tomographic zones, among the most relevant ones), during more than 40 years of research.  Thus, TLF have been documented during a long period of time in the Andean margin by a diverse group of researchers using different approaches with extensive and growing evidences in the last decades.

Regarding geological evidence in particular, until a few years ago, the main evidence supporting the existence of this type of structures was indirect (geophysical data, alignments of volcanic centres, mineral deposits and/or intrusive bodies). Field-based data documenting these structures was scarce. Because of this, pioneer works proposing these long-lived, arc-transverse structures in the Andes (e.g., Salfity, 1985; Cembrano and Lara, 2009), classified them as lineaments or inferred faults, as highlighted by the reviewer. However, this situation has changed drastically in recent years, as several works published during the last decade (e.g., Lanza et al., 2013; Piquer et al., 2016, 2021; Giambiagi et al., 2017; Farrar et al., 2023, among others) have provided detailed structural maps of various Andean regions, based on hundreds or even thousands of fault-plane measurements, which demonstrate without any doubt the existence of this arc-transverse fault systems.

Second, we do agree with Prof. Booth Rea that faults are not straight lines, we represent TLF as relatively straight entities for simplicity, due to the lack of continuous evidences in structures running for several hundred of km along strike. These structures are involving damage zones of 1-10 km width. The lack of continuous evidences along each TLF is the result of the pervasive action of principal stress perpendicular to the trench (mostly EW). This subduction-related process has been active for more than 300 Ma in the Andean orogeny, controlling the spatial distribution of the geological units, magmatism and structures along the margin, with a first order NS alignment. Thus, geological process occurring perpendicular to this first order forcing factor have been mostly overlooked in the early studies in the Andes, basically due to sampling evidences. In this regard, geophysical techniques, in particular the magnetic technique, has been a key tool in this particular geographic setting. Earth magnetic field is NS, and thus represent a natural filter to enhance EW features, one representative case is the “Melipilla Anomaly” (Yanez et al., 1998). On the other hand, seismic evidences of deformation in the high Andes are oriented NS in good correspondence with the mapped thrust faults and folds associated with the long-term EW convergence. However, the largest event recorded in the last 70 years in Central Chile is the Las Melozas (1958) 6.8 Mw earthquake, with a strike slip focal mechanism oriented N37°E, and no surface evidences so far (Barrientos et al., 2004).   As highlighted before, detailed geological and structural mapping (Piquer et al., 2016;Giambiagi et al., 2017; Farrar et al., 2023), mostly in the vicinity of ore deposits is revealing new evidences of major structural domains controlling the occurrence of ore deposits in the main cordillera, that can be extrapolated to the west coinciding with seismic swarms in the seismogenic zone (Piuquencillo Faul System; Piquer et al., 2021). Also, the Pichilemu TLF observed normal faulting during the aftershock sequence of the Maule 8.8 Mw earthquake (2010) (Farias et al, 2011). Such a seismological evidences are scarce due to the short time window, with recurrence times of hundreds to thousands of years, thus poorly sampled.

Third, the origin of TLF is still a matter debate, most likely due to processes going at different time intervals along the Andean evolution, including rifting processes, subduction underplating, active and passive ridge collision, among other episodic processes overlapped to the “normal” subduction process. A detailed discussion on this fundamental process is beyond the scope of this contribution. However, we are certain that in a time window of 200-300 Ma, such episodic event most likely happened several times along the margin.

Fourth, we disagree with the affirmation that some TLF only coincides with the seismo-tectonic segmentation by chance, and mostly due to the over population of TLF. According to our long-lasting research, compilation, and documentation of TLF, they constitute an integral part of the margin, controlling magmatism and hydrothermal processes, as well as first order geological units. We postulate here that, some of them, well oriented, more evolved, and with active segments in the coastal region in present times, are also acting as barriers for the seismic coupling due to their high permeability nature, implying a geological/fluid pressure control on the seismo-tectonic segmentation of the margin.  Thus, some TLF meet this requirement and six relatively indirect and independent spatial proxies are consistent with this working hypothesis.   

Last but not least, we do acknowledge that the evidences presented in this work are fragmentary and of semi-quantitative origin, but as a whole they present a consistent pattern in agreement with the working hypothesis. We do also agree that the present proposal of plate coupling and tectonic segmentation requires more focused observations to support the conceptual model. Our aim is to open a discussion with new ideas for such a fundamental problem.                   

Dear Professor Booth Rea

We do appreciate your comments and observations regarding this work, and in particular your concerns on the evidences that support the existence of a large number of TLF cutting the Andean margin (36 in total according to Table 1). This discussion provides the opportunity for us to reinforce the relevance of TLF in the geological evolution of the Andes since at least Mesozoic times.

First, we would like to clarify, as described in table 1, that at least 36 publications support the occurrence of 36 TLF, and only 11 TLF are taken from the work of Yanez and Rivera (2019) (30%), with 3 of them coming from previous studies (see Table 1 of the paper). Thus, only 22% of the TLF were defined in Yanez and Rivera (2019), as a result of an intensive work of geological and mining exploration done by Codelco (Chilean coper mine company) and collaborators from the academia in northern Chile, interpreting the results for more than a decade,most of this work unpublished before the synthesis presented by Yanez and Rivera (2019) paper. The remaining 70% of TLF’s are based on studies carried out by several and independent researchers using different mechanisms (structural geology, potential methods, seismology, indirect evidences of highly permeable zones such as volcanic and ore deposits alignments, high Vp/Vs tomographic zones, among the most relevant ones), during more than 40 years of research.  Thus, TLF have been documented during a long period of time in the Andean margin by a diverse group of researchers using different approaches with extensive and growing evidences in the last decades.

Regarding geological evidence in particular, until a few years ago, the main evidence supporting the existence of this type of structures was indirect (geophysical data, alignments of volcanic centres, mineral deposits and/or intrusive bodies). Field-based data documenting these structures was scarce. Because of this, pioneer works proposing these long-lived, arc-transverse structures in the Andes (e.g., Salfity, 1985; Cembrano and Lara, 2009), classified them as lineaments or inferred faults, as highlighted by the reviewer. However, this situation has changed drastically in recent years, as several works published during the last decade (e.g., Lanza et al., 2013; Piquer et al., 2016, 2021; Giambiagi et al., 2017; Farrar et al., 2023, among others) have provided detailed structural maps of various Andean regions, based on hundreds or even thousands of fault-plane measurements, which demonstrate without any doubt the existence of this arc-transverse fault systems.

Second, we do agree with Prof. Booth Rea that faults are not straight lines, we represent TLF as relatively straight entities for simplicity, due to the lack of continuous evidences in structures running for several hundred of km along strike. These structures are involving damage zones of 1-10 km width. The lack of continuous evidences along each TLF is the result of the pervasive action of principal stress perpendicular to the trench (mostly EW). This subduction-related process has been active for more than 300 Ma in the Andean orogeny, controlling the spatial distribution of the geological units, magmatism and structures along the margin, with a first order NS alignment. Thus, geological process occurring perpendicular to this first order forcing factor have been mostly overlooked in the early studies in the Andes, basically due to sampling evidences. In this regard, geophysical techniques, in particular the magnetic technique, has been a key tool in this particular geographic setting. Earth magnetic field is NS, and thus represent a natural filter to enhance EW features, one representative case is the “Melipilla Anomaly” (Yanez et al., 1998). On the other hand, seismic evidences of deformation in the high Andes are oriented NS in good correspondence with the mapped thrust faults and folds associated with the long-term EW convergence. However, the largest event recorded in the last 70 years in Central Chile is the Las Melozas (1958) 6.8 Mw earthquake, with a strike slip focal mechanism oriented N37°E, and no surface evidences so far (Barrientos et al., 2004).   As highlighted before, detailed geological and structural mapping (Piquer et al., 2016;Giambiagi et al., 2017; Farrar et al., 2023), mostly in the vicinity of ore deposits is revealing new evidences of major structural domains controlling the occurrence of ore deposits in the main cordillera, that can be extrapolated to the west coinciding with seismic swarms in the seismogenic zone (Piuquencillo Faul System; Piquer et al., 2021). Also, the Pichilemu TLF observed normal faulting during the aftershock sequence of the Maule 8.8 Mw earthquake (2010) (Farias et al, 2011). Such a seismological evidences are scarce due to the short time window, with recurrence times of hundreds to thousands of years, thus poorly sampled.

Third, the origin of TLF is still a matter debate, most likely due to processes going at different time intervals along the Andean evolution, including rifting processes, subduction underplating, active and passive ridge collision, among other episodic processes overlapped to the “normal” subduction process. A detailed discussion on this fundamental process is beyond the scope of this contribution. However, we are certain that in a time window of 200-300 Ma, such episodic event most likely happened several times along the margin.

Fourth, we disagree with the affirmation that some TLF only coincides with the seismo-tectonic segmentation by chance, and mostly due to the over population of TLF. According to our long-lasting research, compilation, and documentation of TLF, they constitute an integral part of the margin, controlling magmatism and hydrothermal processes, as well as first order geological units. We postulate here that, some of them, well oriented, more evolved, and with active segments in the coastal region in present times, are also acting as barriers for the seismic coupling due to their high permeability nature, implying a geological/fluid pressure control on the seismo-tectonic segmentation of the margin.  Thus, some TLF meet this requirement and six relatively indirect and independent spatial proxies are consistent with this working hypothesis.   

Last but not least, we do acknowledge that the evidences presented in this work are fragmentary and of semi-quantitative origin, but as a whole they present a consistent pattern in agreement with the working hypothesis. We do also agree that the present proposal of plate coupling and tectonic segmentation requires more focused observations to support the conceptual model. Our aim is to open a discussion with new ideas for such a fundamental problem.