Integration of Terrestrial Laser Scanner (TLS) and Ground Penetrating Radar (GPR) to Characterize the Three-Dimensional (3D) Geometry of the Maoyaba Segment of the Litang Fault, Southeastern Tibetan Plateau

Round 1

Reviewer 1 Report

This is good application example for detecting the deformation of active fault by using latest technology and methods. Because of the extreme conditions of the plateau, it is extremely difficult to carry out fine detection of active faults. It is rare to establish 3D and 3D fault structure models. This paper carried out a high-resolution detection of the Maoyaba segment of the Litang Fault in the eastern Tibet Plateau, which can better explain the surface rupture characteristics and fault geometry structure. It is a good manuscript on the application of new technologies. I suggest that the paper can be published in this journal after minor revision.

The following are the review comments on the content of each part:

1.      Abstract:

The results obtained in the paper should be better summarized in the abstract. The current abstract section only summarizes the methods and techniques. It is necessary to summarize the advantages of the GPR compared with other methods, or to express the scientific project that the GPR can solve.

2.      Figures:

In Fig. 1a, the text ‘The Ordos Plate’ is placed in wrong position. And the expression of ‘The West of China’, which I suggest to delete, or revise to  ‘the Alashan Plate’..

In Fig. 2, the red fault lines extend too straight and complete, which cannot represent the geometric characteristics of the fault. In addition, there should be more rivers in the figure, especially those that pass through the Maoyaba fault, flowing into the basin from the surrounding area. And the way the faults move should to be more clearly labeled.

In Fig. 3a, ‘the Fault scarp’ is located in wrong place which should be check again, and I suggest that the orientation of the photo should be marked in Fig. 3b.

There should be a scale in Fig. 4.

In Fig. 6, I suggest that the vertical and horizontal coordinates of the detection profile should be as consistent as possible, which will better represent relevant information, such as the dip of the fault.

In Fig.9, the 250 MHz and 500 MHz GPR interpreted images can get images with different depth ranges, you can put them on the same depth axis to show both profiles instead of stretching or flattening one map, which may lost many informations.

In addition, I suggest that a 3D scale should be added to the Fig. 11 and Fig.12.

3.      Manuscript:

TLS, GPR and other words are expressed completely when they appear for the first time, followed by abbreviations.

In line 46, GPS, which cannot be used directly to measure terrain, is not a topographical surveying instrument similar to the total station. I suppose you mean ‘the total station and RTK’.

It is recommended to explain and summarize the results after data processing, which is more conducive to explaining that this method may obtain clearer results. Similarly,  the fault kinematics and characteristics can be explained and understood through the three-dimensional model.

The first conclusion should not be the main introduction of this paper, which should be summarized from your detection results. In addition, other parts of the conclusion, especially 4 and 5, are mainly to illustrate the applicability of the method. It is recommended to summarize the advantages and better results obtained in the paper with other methods, especially the understanding of the fault structure.

Besides, this manuscript needs careful editing and particular attention to English grammar, spelling, and sentence structure.

 

Author Response

Thank you very much for your letter and the advices about our manuscript entitled “Integration of TLS and GPR to characterize the three-dimensional geometry of the Maoyaba segment of the Litang fault, southeastern Tibetan Plateau” (remotesensing-2022192) submitted to Remote sensing. These comments and advices are very valuable to improve our manuscript. After carefully studying the comments, we have made the revisions by the reviewers' comments.

 

Point 1: Abstract: The results obtained in the paper should be better summarized in the abstract. The current abstract section only summarizes the methods and techniques. It is necessary to summarize the advantages of the GPR compared with other methods, or to express the scientific project that the GPR can solve.

 Response 1:  In oder to summarize the advantages of the GPR method compared with other methods , The content “Three faults were identified on the main fault zone and the fault F₁ and F₃ were the boundary fault, while the fault F₂ was the secondary fault.”and “For reducing the environmental disruption and economic losses, the GPR was the most optimal method for detecting the subsurface structures of active faults in the Litang fault with a non-destructive and cost-effective fashion.”has been added in the abstract.

 

Point 2: In Fig. 1a, the text ‘The Ordos Plate’ is placed in wrong position. And the expression of ‘The West of China’, which I suggest to delete, or revise to  ‘the Alashan Plate’..

Response 2: The text ‘The Ordos Plate’ has been placed in correct position and the expression of ‘The West of China’ has been deleted in Fig.1a.

Point 3: In Fig. 2, the red fault lines extend too straight and complete, which cannot represent the geometric characteristics of the fault. In addition, there should be more rivers in the figure, especially those that pass through the Maoyaba fault, flowing into the basin from the surrounding area. And the way the faults move should to be more clearly labeled.

Response 3: The red fault lines has been revised in Fig.2 and they are in well matching with the geometric characteristics of the fault. The fault lines has been separated in the Maoyaba basin. More rivers has been added in the Maoyaba basin. The way of the normal fualts move has been clearly labeled in the Fig.2.

Point 4: In Fig. 3a, ‘the Fault scarp’ is located in wrong place which should be check again, and I suggest that the orientation of the photo should be marked in Fig. 3b.

Response 4: ‘The Fault scarp’ has been located in correct position in Fig.3a, and the orientation of the photo has been marked in Fig.3b.

Point 5: There should be a scale in Fig. 4.

Response 5: The scales have been added in Fig.4.

Point 6: In Fig. 6, I suggest that the vertical and horizontal coordinates of the detection profile should be as consistent as possible, which will better represent relevant information, such as the dip of the fault.

Response 6: The processed workflow of GPR data are shown in Fig.6 using the same GPR data. Fia.6a and Fia.6b have the same the vertical without time-zero correction. To show real two-way travel time of electromagnetic waves in the underground media, the zero timing of 5.4 ns was performerd to removel the direct waves in the air. T    hat is why that the vertical coordinates in Fig.6c-6g is different to the Fia.6a and Fia.6b. The topographic correction was conducted to compensate the two-way travel time in vertical direction of the GPR data by the high-resolution topographic data. Finally, the relevant information of the faults on the GPR data in this paper were imaged and anlysed by the GPR profiles after applying the topographic correction.

Point 7: In Fig.9, the 250 MHz and 500 MHz GPR interpreted images can get images with different depth ranges, you can put them on the same depth axis to show both profiles instead of stretching or flattening one map, which may lost many informations.

Response 7: The 250 MHz and 500 MHz GPR interpreted images (Fig.9) has been revised with the same depth ranges instead of stretching or flattening one map. In addition, the topograpohic data of the GPR survey lines has been diaplayed in Fig.9a to provide the evident for the current representation of the surface rupture.

Point 8: In addition, I suggest that a 3D scale should be added to the Fig. 11 and Fig.12.

Response 8: The 3D scale has been added to the Fig. 11 and Fig.12.

Point 9: TLS, GPR and other words are expressed completely when they appear for the first time, followed by abbreviations.

Response 9: The manuscript title ‘Integration of TLS and GPR to characterize the three-dimensional geometry of the Maoyaba segment of the Litang fault, southeastern Tibetan Plateau’ has been revised ‘Integration of terrestrial laser scanner (TLS) and ground penetrat-ing radar (GPR) to characterize the three-dimensional (3D) geome-try of the Maoyaba segment of the Litang fault, southeastern Tibetan Plateau’.

Point 10: In line 46, GPS, which cannot be used directly to measure terrain, is not a topographical surveying instrument similar to the total station. I suppose you mean ‘the total station and RTK’.

Response 10: The sentences ‘The classical methods, such as the total station and GPS, were generally used to obtain high-resolution topographic data with a time-consuming and expensive way. ’ has been revised ‘The classical methods, such as the total station and RTK, were generally used to obtain high-resolution topographic data with a time-consuming and expensive way’.

Point 11: It is recommended to explain and summarize the results after data processing, which is more conducive to explaining that this method may obtain clearer results. Similarly, the fault kinematics and characteristics can be explained and understood through the three-dimensional model.

Response 11: The content is described in the “4.2.2 3D GPR data” part in the revised manuscript. The added content is ‘From the 3D GPR data, the wedge-shaped deformation zone between the fault F₁ and F₂ was clearly observed by the pronounced variations in the waveform and amplitude of the radar reflections. What’s more, the characteristics of the fault F₁ and F₂ were also directly confirmed on the 3D GPR data, such as the location of the fault plane, the strikes and dipping. The fault F1 was regarded as the main fault with a SE dipping of the nearly 90°, while the fault F₂ was the secondary fault in the fault zone with a NE dipping’.

The content is described in the “4.3 Data visualization of point clouds and GPR data” part in the revised manuscript. The added content is ‘As a result, the integrated data of TLS and GPR rendered more the realistic surface and subsurface geometry of active faults, which not only allows discerning the fault traces at the surface (red dotted line in Figure 12a ) and its strikes, but also obtain the range of the deformation zone and the location of fault dislocation. In additon, this hybrid data could offer the opportunity that the subsurface structures of the fault can be better understanding with its corresponding superficial data’.

Point 12: The first conclusion should not be the main introduction of this paper, which should be summarized from your detection results. In addition, other parts of the conclusion, especially 4 and 5, are mainly to illustrate the applicability of the method. It is recommended to summarize the advantages and better results obtained in the paper with other methods, especially the understanding of the fault structure.

Response 12: The first conculusion ‘(1) From the Google Earth image and the panoramic image of the study site, the fault offset terrace T₁ and T₂ are visible along the northern margin fault of the Maoyaba basin. Furthermore, the surface evidences of the main fault are clearly identified by the surface expression and geomorphologic features, and they all indicate that the Maoyaba fault has been highly active in the late Quaternary with the normal component. ‘ has been moved to the “2. Geological setting” part in the revised manuscript.

The contents of the conculsion has been modified in the revised manuscript, and the revised content is ‘In this study, the integration of TLS and GPR was used to image the 3D surface and subsurface geometry of the Maoyaba fault and also to reveal its motion mode. The con-clusions associated to the study were described as the followings:

(1) The fault offset T₁ and T₂ landscape with nearly the W-E trending and other ge-omorphic evidences of faulting were apparently revealed by the TLS-derived data. A relative lower elevation area, paralleling to the fault, was well revealed on the TLS results and it implied that the hanging wall had been locally affected by later sedimentation.

(2) On the 250 MHz and 500 MHz GPR profiles along the survey lines 2 to 4, a wedge-shaped zone of the electromagnetic wave was observed and it was considered as the main fault zone with a small graben structure, which the maximum width on the surface could up to ~40 m.

(3) The characteristics of the fault F₁ and F₂ were also directly confirmed on the 3D GPR data, such as the location of the fault plane, the strikes and dipping. The fault F₁ was regarded as the main fault with a SE dipping of the nearly 90°, while the fault F₂ was the secondary fault in the fault zone with a NE dipping.

(4) The 3D surface and subsurface geometry of the fault were established by the in-tegrated data of TLS and GPR. This hybrid data rendered more the realistic surface and subsurface geometry of active faults, which not only allows discerning the fault traces at the surface and its strikes, but also obtain the range of the deformation zone and the location of fault dislocation.

(5) The study results demonstrate that integration of the TLS and GPR is suitable for delineating the 3D surface and subsurface geometry of the fault on the Maoyaba fault. In the future research, the integration of TLS and GPR will be widely used for active faults investigation and seismic hazard assessment in the different geological environment, especially in the Qinghai-Tibet Plateau area. ‘

Point 13: Besides, this manuscript needs careful editing and particular attention to English grammar, spelling, and sentence structure.

Response 13: We revised the manuscript point-by-point as showed in the reivsed manuscript (red words), such as English grammar, spelling, and sentence structure.

 

Author Response File: Author Response.pdf

Reviewer 2 Report

This manuscript integrates TLS with GPR to delineate the three-dimensional geometry of the Litang fault. However, it is stated that such a technique integration has been widely applied to various fields. The technical or scientific innovation is unclear in this manuscript. A simple application of these two techniques to the Litang fault is of less innovation. The results seem not deepen the understanding of the fault.

Besides, the results of both the TLS and GPR looks bad. The density of the point cloud is uneven, and the surface features cannot be clearly represented. The GPR profile seems serious suffer from the system instability. The topography data used for the GPR data processing were acquired from GPS, not TLS.

In conclude, I cannot recommend this manuscript for publication in Remote Sensing.

Author Response

We feel great thanks for your professional review work on our article. These comments and advices are very valuable to improve our manuscript. We tried our best to improve the manuscript and made some changes in the manuscript. The main corrections in the paper and the responds to your comments are as flowing:

Point 1: This manuscript integrates TLS with GPR to delineate the three-dimensional geometry of the Litang fault. However, it is stated that such a technique integration has been widely applied to various fields. The technical or scientific innovation is unclear in this manuscript. A simple application of these two techniques to the Litang fault is of less innovation. The results seem not deepen the understanding of the fault.

Response 1: Althought the integration of TLS and GPR has been gradually applied in active faults investigation all over the world in recent years, there were little the integration of TLS and GPR surveys had been performed on the Litang fault, especially in the Qinghai-Tibet Plateau area.  Owing to the severe natural environment in the Tibetan Plateau, it is extremely difficult to carry out fine detection of active faults by the traditional methods without the environmental disruption. This paper carried out to establish the detailed 3D surface and subsurface geometry of the Maoyaba segment of the Litang fault in the eastern Tibet Plateau based on the TLS and GPR, which can better explain the surface rupture characteristics ,fault geometry structure and the motion mode of the Maoyaba fault. So this paper was not a simple application of TLS and GPR method in the Litang fault. The manuscript submission to Remote sensing has two objectives:

(1) Because of the severe natural environment in the southeast Tibetan Plateau, the traditional methods were time-consuming and high-costing to obtain the topographic data and shallow geometry of the Maoyaba fault without the environmental disruption. In addtion, the Litang area locates in Qinghai-Tibet Plateau area and has a severe natural environment with water-saturated clay in near-surface layer, where the electrical conductivity constitutes a challenge for radar wave propagation. One objective of this paper is demonstrat the integration of TLS and GPR for delineating the 3D surface and subsurface geometry of the fault scarp on the Maoyaba segent of the Litang fault.

(2) revealing the motion mode of the Maoyaba fault on the basis of the correlating geomorphologic features and stratigraphic data. Previous studies have different views on the motion mode of the Maoyaba fault as the followings: 1) the sinistral strike-slip fault with the thrust components; 2) the normal fault with a vertical rate of 0.6 ± 0.1 mm/yr. What’s more, the Sichuan-Tibet railway will traverse the Maoyaba fault, so it is essential to study the motion mode of the Maoyaba fault in terms of the high-resolution topographic data and subsurface geometry.

In addtion, the motion mode of the Maoyaba fault has been revealed in the section 5.2. The content is “The remarkable geomorphic features of the late Quaternary in the Litang fault are numerous tectonic basins with NW, nearly EW and NNW striking, such as the Litang basin, Maoyaba basin and Jiawa basin. As the largest basin along the Litang fault, the Maoyaba basin is approximate diamond shape with nearly E-W striking, and it domi-nated by the Maoyaba fault. Zhou et.al (2005) suggested that the blocks between the Ba-tang fault (a right-lateral strike-slip fault) and the Litang fault (a sinistral-lateral strike slip fault) slipped to south by the principal compressive stress with nearly EW directing, causing a nearly EW trending normal fault in the blocks, such as the Maoyaba fault . Based on the geomorphic features and morphology and tectonic of the basin, Ma et al (2014) illustrated that the Maoyaba basin is the rifted-basin, which controlled by the normal faults with nearly E-W striking, and the basin was formed from the nearly N-S trending extension and tension stress. The GPS data and focal mechanism analysis also confirmed that there was obvious tensional movement in the Litang-Batang area. We concluded that the Maoyaba basin is the rifted-basin, which controlled by the normal faults with nearly E-W striking. It is not a compressional fault basin as some re-searchers previously thought .

The Maoyaba fault, located in the northwestern section of the Litang fault, is the main boundary fault of the Maoyaba basin and it dominates the development of the Maoyaba basin. Satellite image and field investigations demonstrate that active faults are hander to follow on the south flank of the Maoyaba basin, and it indicates that the master fault is represented by the active fault along the northern edge of the Maoyaba basin. Numerous triangular facets are particularly impressive on the northern edge of the basin, which hints the normal component of the Maoyaba fault. In addition, the well-developed fault scarps cutting late Quaternary geomorphic features are well visible on the alluvial fans at the northern edge of the Maoyaba basin, and the steams or gullies flowing through the fault seem a plume spread at the base of the scarp. This typical geomorphology show a well consistency with the normal motion of the Maoyaba fault, as well as the graben, horst and the well-developed alluvio-glacial fans attest to the normal motion of the fault and its recent normal faulting activity. Owing to the severe natural environment on the Maoyaba fault, the traditional methods were time-consuming and high-costing to record the topographic data and shallow geometry of active faults. The integration of TLS and GPR was chosen to detect and reveal the 3D surface and subsurface geometry of the fault on the Maoyaba fault. Our study results show that the Maoyaba fault is characterized by normal fault activity in terms of tectonic geomorphology and shallow structure, so it should be a typical active normal fault rather than a sinistral strike-slip fault with the thrust components.”

The contents of the conculsion has been modified in the revised manuscript, and the revised content is ‘In this study, the integration of TLS and GPR was used to image the 3D surface and subsurface geometry of the Maoyaba fault and also to reveal its motion mode. The con-clusions associated to the study were described as the followings:

(1) The fault offset T₁ and T₂ landscape with nearly the W-E trending and other ge-omorphic evidences of faulting were apparently revealed by the TLS-derived data. A relative lower elevation area, paralleling to the fault, was well revealed on the TLS results and it implied that the hanging wall had been locally affected by later sedimentation.

(2) On the 250 MHz and 500 MHz GPR profiles along the survey lines 2 to 4, a wedge-shaped zone of the electromagnetic wave was observed and it was considered as the main fault zone with a small graben structure, which the maximum width on the surface could up to ~40 m.

(3) The characteristics of the fault F₁ and F₂ were also directly confirmed on the 3D GPR data, such as the location of the fault plane, the strikes and dipping. The fault F₁ was regarded as the main fault with a SE dipping of the nearly 90°, while the fault F₂ was the secondary fault in the fault zone with a NE dipping.

(4) The 3D surface and subsurface geometry of the fault were established by the in-tegrated data of TLS and GPR. This hybrid data rendered more the realistic surface and subsurface geometry of active faults, which not only allows discerning the fault traces at the surface and its strikes, but also obtain the range of the deformation zone and the location of fault dislocation.

(5) The study results demonstrate that integration of the TLS and GPR is suitable for delineating the 3D surface and subsurface geometry of the fault on the Maoyaba fault. In the future research, the integration of TLS and GPR will be widely used for active faults investigation and seismic hazard assessment in the different geological environment, especially in the Qinghai-Tibet Plateau area. ‘

Point 2: Besides, the results of both the TLS and GPR looks bad. The density of the point cloud is uneven, and the surface features cannot be clearly represented. The GPR profile seems serious suffer from the system instability. The topography data used for the GPR data processing were acquired from GPS, not TLS.

Response 2: Due to there were many vegetation ( ~5-10 cm height) and rocks at the surface of the fualt scarps in the study site, the raw 3D color point clouds of the fault seems to be uneven. Before empolying the data integration of TLS data and GPR data, the raw color clouds have been processed by filtering procedure (the manual operation and the maximum local slope filter) to eliminate the noise points using the Geomagic Studio software, such as such as vegetation, rocks, vehicles and people. Then, the point clouds resampling algorithm was conducted to reduce the amount of the point clouds datasets and retain the useful information, the data processing procedure illustrated in “3.1 TLS data acquisition and processsing “ section. In addtion, the DEM, the overlain map of the surface slope map and the topographic contours map and 3D surface model of the fault all show a good results for the morphologic features of the fault, which are all estiblished by the point clouds, and it also indicates that the point clouds are not bad.

Although the study site has a severe natural environment with water-saturated clay in near-surface layer, the 250 MHz and 500 MHz GPR have a effective detection for the shallow subserface of active faults. To show the GPR results, the 250 MHz and 500 MHz GPR interpreted images (Fig.9) has been revised with a gray scale representation. In addition, the topograpohic data of the GPR survey lines has been diaplayed in Fig.9a to provide the evident for the current representation of the surface rupture. The 250 MHz and 500 MHz GPR images (Fig.10) has been re-processed and re-interpreted. Fig.11a has been changed to help the reader to identify the fault. From the Fig.11a, the 3D model of the active fault can be visible by the orthobeam projection, and the fault F₁ and F₂ trace can be clearly observed at the surface. We could also determined the range of the deformation zone and the location of fault dislocation.

 In this study, the integrated system of GPR and differential GPS (DGPS) was used to obtain the GPR data with geographical information (Figure 5) to determine the precise location of the subsurface structures. When the GPR antenna was pulled on the surface, the pulse signals of the survey wheel were applied to trigger the GPR control unit and the GPS signal receiver, respectively. Meanwhile, the GPR data and the geographical coordinates were simultaneously achieved by the GPR and the GPS signal receiver. Compared to the TLS, the REK-GPS has the advantage of collectiong the topography data in data collection and the accuracy of the GPR data.

 

Author Response File: Author Response.pdf

Reviewer 3 Report

The study about the integration of TLS and GPR methods has been developed to image the 3D surface and subsurface geometry of the Maoyaba fault system. These were established by the integration of TLS and GPR data sets, obtaining an hybrid data set.

Authors are invited to improve the description about the processing method employed to integrate the different data sets.

Author Response

Many thanks for the insightful comments and suggestions about our manuscript entitled “Integration of TLS and GPR to characterize the three-dimensional geometry of the Maoyaba segment of the Litang fault, southeastern Tibetan Plateau” (remotesensing-2022192) submitted to Remote sensing. We tried our best to improve our manuscript in the revised manuscript.

Point 1: Authors are invited to improve the description about the processing method employed to integrate the different data sets.

Response 1: The content is described in the “3.3. Data integration method of TLS and GPR” part in the revised manuscript. The revised content is ‘The TLS collects the dense point clouds with Cartesian coordinates (x, y, z) and texture maps to show the high-resolution topographic data of active faults. The point clouds mainly involves x, y, z coordinates of the scanning scene, the point attributes (such as the intensity value of the reflected laser beam) and the RGB images captured by the camera. GPR has been used to delineate the subsurface materials or buried-object on a gray or color 2D radargram, which is composed of the continuous single-channel reflection waves with the same trace interval. The subsurface materials or buried-object on the GPR image was identified by the variation in the pattern and relative amplitude of the electromagnetic wave.

In this study, the data integration method of TLS and GPR was present to compre-hensive interpretation and analysis of the fault. The data integration workflow was de-scribed in Figure 7. First, the raw point clouds data and the GPR data were both processed in the light of the data processing procedure illustrated in Section 3.1 and 3.2. With the help of the spherical reflector targets, the point clouds of the scanning stations were combined into a single set of point cloud datasets using the Faro Scene software. The filtering and re-sampled procedures were also used to reduce the amount of the point cloud datasets and retain the useful information by the Geomagic Studio software. To show the realistic geomorphologic features of the fault, the color point clouds were generated by the processed point clouds and the panoramic images. This color point clouds was transformed as the xyz feature in the geodetic coordinate system by means of the geodetic coordinates of the GCPs. Second, the GPR data and the geographical co-ordinates were simultaneously achieved by the pulse signals of the survey-wheel on the GPR equipment, and the geographical information of GPR data was determined by the post-processed differential correction using a based station data or a reference station data. What’s more, the time synchronization algorithm was proposed for data integration of GPR and DGPS, and each trace on the GPR profile had a well spatial correlation with the geographical coordinates of the GPS receiver. In view of the mathematical model of electromagnetic wave, the time-depth conversion was implemented to each trace on the GPR profiles. Then the GPR data was converted into the point clouds datasets with xyz feature in the geodetic coordinate system, which was the same as the TLS point clouds. At last, once the TLS data and GPR data were georeferenced in the same geodetic coordinate system, the subsurface data (GPR) should be integrated with the surface data (TLS) for comprehensive interpretation and analysis of the fault. ‘

Author Response File: Author Response.pdf

Reviewer 4 Report

The work is interesting and the approach based on the integration of GPR and laser scan can give useful information about the surficial development of a fault.

Maoyaba fault is analyzed. My general comment is that more than integration the authors make a comparison of the data. Indeed, the TLS data are used for supporting the GPR data interpretation. For this reason, the last sentence of the paper (raws 535-537) should be deleted or smoothed.

The bibliography is good but should be improved by adding works employing GPR with conventional geophysical methods for faults detection (seismic, geoelectric). For example:

 Nappi, Rosa, et al. "Joint Interpretation of Geophysical Results and Geological Observations for Detecting Buried Active Faults: The Case of the “Il Lago” Plain (Pettoranello del Molise, Italy)." Remote Sensing 13.8 (2021): 1555.

Finizola, Anthony, et al. "Adventive hydrothermal circulation on Stromboli volcano (Aeolian Islands, Italy) revealed by geophysical and geochemical approaches: implications for general fluid flow models on volcanoes." Journal of Volcanology and Geothermal Research 196.1-2 (2010): 111-119.

The paper is well organized, and the introduction is good. The strong limit of the research is related to the presentation of GPR results. Radargrams presented in fig. 9 are unclear. I suggest to use a gray scale representation. Furthermore, where the red line indicates the surface rupture (deformation area), it is not identifiable any evident reflections at least in the current representation. The same consideration can be made with reference to fig.10. More interesting the results plotted in figure 11. But, the 3D visualization does not help the reader to identify the fault except for the frontal part of the model. Can you change the point of view? In Fig.12c I can see some "flying" TLS data. Is it normal or is it simply noise?

Finally, fig 12b, where it is identified the surface rupture of the laser scanner data are missed. Verify, please. 

   

 

Author Response

Thank you very much for your letter and the advices about our manuscript entitled “Integration of TLS and GPR to characterize the three-dimensional geometry of the Maoyaba segment of the Litang fault, southeastern Tibetan Plateau” (remotesensing-2022192) submitted to Remote sensing. These comments and advices are very valuable to improve our manuscript. After carefully studying the comments, we have made the revisions by the reviewers' comments.

Point 1: Maoyaba fault is analyzed. My general comment is that more than integration the authors make a comparison of the data. Indeed, the TLS data are used for supporting the GPR data interpretation. For this reason, the last sentence of the paper (raws 535-537) should be deleted or smoothed.

Response 1: The last sentence of the paper (raws 535-537) has been deleted. The contents of the conculsion has been modified in the revised manuscript, and the revised content is ‘In this study, the integration of TLS and GPR was used to image the 3D surface and subsurface geometry of the Maoyaba fault and also to reveal its motion mode. The con-clusions associated to the study were described as the followings:

(1) The fault offset T₁ and T₂ landscape with nearly the W-E trending and other ge-omorphic evidences of faulting were apparently revealed by the TLS-derived data. A relative lower elevation area, paralleling to the fault, was well revealed on the TLS results and it implied that the hanging wall had been locally affected by later sedimentation.

(2) On the 250 MHz and 500 MHz GPR profiles along the survey lines 2 to 4, a wedge-shaped zone of the electromagnetic wave was observed and it was considered as the main fault zone with a small graben structure, which the maximum width on the surface could up to ~40 m.

(3) The characteristics of the fault F₁ and F₂ were also directly confirmed on the 3D GPR data, such as the location of the fault plane, the strikes and dipping. The fault F₁ was regarded as the main fault with a SE dipping of the nearly 90°, while the fault F₂ was the secondary fault in the fault zone with a NE dipping.

(4) The 3D surface and subsurface geometry of the fault were established by the in-tegrated data of TLS and GPR. This hybrid data rendered more the realistic surface and subsurface geometry of active faults, which not only allows discerning the fault traces at the surface and its strikes, but also obtain the range of the deformation zone and the location of fault dislocation.

(5) The study results demonstrate that integration of the TLS and GPR is suitable for delineating the 3D surface and subsurface geometry of the fault on the Maoyaba fault. In the future research, the integration of TLS and GPR will be widely used for active faults investigation and seismic hazard assessment in the different geological environment, especially in the Qinghai-Tibet Plateau area. ‘

Point 2: The bibliography is good but should be improved by adding works employing GPR with conventional geophysical methods for faults detection (seismic, geoelectric). For example:

Nappi, Rosa, et al. "Joint Interpretation of Geophysical Results and Geological Observations for Detecting Buried Active Faults: The Case of the “Il Lago” Plain (Pettoranello del Molise, Italy)." Remote Sensing 13.8 (2021): 1555.

Finizola, Anthony, et al. "Adventive hydrothermal circulation on Stromboli volcano (Aeolian Islands, Italy) revealed by geophysical and geochemical approaches: implications for general fluid flow models on volcanoes." Journal of Volcanology and Geothermal Research 196.1-2 (2010): 111-119.

Response 2: The bibliography with employing GPR with conventional geophysical methods for faults detection (seismic, geoelectric) has been added in revised manuscript.

[29] Nappi, Rosa, et al. "Joint Interpretation of Geophysical Results and Geological Observations for Detecting Buried Active Faults: The Case of the “Il Lago” Plain (Pettoranello del Molise, Italy)." Remote Sensing 13.8 (2021): 1555.

[30] Finizola, Anthony, et al. "Adventive hydrothermal circulation on Stromboli volcano (Aeolian Islands, Italy) revealed by geophysical and geochemical approaches: implications for general fluid flow models on volcanoes." Journal of Volcanology and Geothermal Research 196.1-2 (2010): 111-119.

Point 3: The paper is well organized, and the introduction is good. The strong limit of the research is related to the presentation of GPR results. Radargrams presented in fig. 9 are unclear. I suggest to use a gray scale representation. Furthermore, where the red line indicates the surface rupture (deformation area), it is not identifiable any evident reflections at least in the current representation. The same consideration can be made with reference to fig.10.

Response 3: The 250 MHz and 500 MHz GPR interpreted images (Fig.9) has been revised with a gray scale representation. In addition, the topograpohic data of the GPR survey lines has been diaplayed in Fig.9a to provide the evident for the current representation of the surface rupture. The 250 MHz and 500 MHz GPR images (Fig.10) has been re-processed and re-interpreted.

Point 4: More interesting the results plotted in figure 11. But, the 3D visualization does not help the reader to identify the fault except for the frontal part of the model. Can you change the point of view?

Response 4: Fig.11a has been changed to help the reader to identify the fault.From the Fig.11a, the 3D model of the active fault can be visible by the orthobeam projection, and the fault F₁ and F₂ trace can be clearly observed at the surface. We could also determined the range of the deformation zone and the location of fault dislocation.

Point 5: In Fig.12c I can see some "flying" TLS data. Is it normal or is it simply noise?

Response 5: Due to the limitation of the veiw, there were many "flying" point clouds in the Fig.12c, and it actually the point clouds at the surface not noise points.

Point 6: Finally, fig 12b, where it is identified the surface rupture of the laser scanner data are missed. Verify, please.

Response 6: Due to there were many vegetation located on the surface rupture as shown in the figure below, the suface rupture were not collected and displayed by the TLS data. In addition, the noise points of the vegetation were removed by the filtering procedures in Geomagic Studio software, the blank areas on the TLS data indicate the location of the surface rupture with the vegetation.

 

Author Response File: Author Response.pdf