Review Reports - Late Quaternary Segment Faulting Behavior of Yilan-Yitong Fault and Its Potential Seismic Hazards, NE China, by Using Multisource Remote Sensing Data

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The submitted paper applies satellite remote sensing and UAV data to conduct a systematic study of the geometric length and the most recent seismic event of the YLYTF fault segment in the northeastern part of the TLF fault. It further compares these results with the latest Holocene activity of the TLF North China section. The topic is relevant to the journal. The key contribution of the study is the confirmation that the maximum rupture length of the YLYTF active fault zone exceeds previous estimates. This finding has important implications for reassessing seismic risk in Northeast China, where densely populated areas are distributed along these fault zones. The manuscript provides a valuable perspective on the background of intense earthquake activity in a region traditionally considered to have weak tectonic activity. Nevertheless, several issues should be addressed before publication:

1.The topic of submitted article focuses on the most recent paleoseismic event identified from four newly excavated trenches, but it does not adequately discuss evidence for earlier events visible in the trench stratigraphy. Similar issues exist for TCQH, TCXBT, and other paleoseismic records. The manuscript should provide a more detailed discussion of these additional events. If recurrence intervals remain uncertain, the key findings should at least be mentioned in the Discussion section to allow reviewers to evaluate the accuracy of stratigraphic interpretations better.

2. The co-authors makes use of 30 AMS-14C age determinations. However, the stratigraphic sequence shows rapid age increases within 1–2 m of the near-surface, and significant discrepancies exist between different dating methods, as reported in previous studies (e.g., Yu et al., 2018). The authors are encouraged to strengthen their comparison of multiple dating approaches in the Discussion. This would help clarify the reliability of age constraints for paleoseismic events. Reference should also be made to recent work on regional comparative dating of paleosols in Northeast China, which may provide valuable methodological insights.

3. The paper emphasizes the existence of a unified 160 km rupture segment in Northeast China, suggesting that it may produce an event comparable in scale to the 1668 earthquake. Given that the recurrence interval of Mw ≥ 8 earthquakes at plate boundaries is only ~500 years, and that far-field Coulomb stress continues to load this region (Shao et al., 2018), the authors should further clarify the significance of such a rupture under present conditions of low slip rates and long recurrence intervals. This clarification would help contextualize the necessity and urgency of assessing the large-earthquake potential of the Fangzheng–Tangyuan fault segment.

4. To remote sensing data, the UAV-based geomorphic analysis near the trenches is insufficiently presented. High-resolution geomorphic evidence supporting the reported fault displacements is also lacking. Adding one or two supplementary UAV or satellite images would greatly strengthen the geomorphological basis of the paper’s conclusions.

5.The manuscript would benefit from a more comprehensive discussion of Holocene TLF activity in adjacent regions of Russia if relevant studies are available.

Recommendation: The paper presents essential findings and is suitable for the journal. However, revisions are required to strengthen the paleoseismic interpretation, dating comparisons, and geomorphic evidence. I recommend moderate revision before further consideration.

Author Response

Response: Thank you very much for your general positive comments. We will follow your comments and advice below to improve our quality.

Response: Thank you very much for the valuable comments from reviewer #1. In addition to the late Holocene earthquake event discussed in this submitted article, the trenches we excavated also revealed evidence of several earlier events dating back to the late Pleistocene. Currently, the most challenging aspect of determining the recurrence interval of paleoseismic events along the TLF fault zone is obtaining reliable chronological samples.

As mentioned in the version submitted for review, we plan to present a separate and detailed discussion on the sequence and recurrence intervals of paleoseismic events based on the trench data. Given the substantial progress made in the remote sensing interpretation of linear scarps and fault segmentation along the northeastern and Shandong sections of the TLF, this paper focuses on estimating the minimum seismic potential of the northeast segment and comparing it with that of the Shandong segment. This analysis integrates remote sensing results with constraints from the most recent trenching data.

During the revision process, we carefully considered the reviewer’s suggestions. However, given the Journal of Remote Sensing's disciplinary focus, we believe that this paper is best suited to emphasize moment magnitude–based seismicity and hazard assessment for the Holocene segment, from a remote sensing perspective supplemented by selected age constraints. Therefore, we have not expanded the discussion on the recurrence interval in this version.

We sincerely appreciate the reviewer’s insightful suggestions. Following the completion of this study, we plan to publish a dedicated paper presenting our detailed findings on multiple consecutive rupture events identified in a single trench investigation (2018–2021), based on the research of Prof. Wei Min, Dr. Zhongyuan Yu, and others, for further reference and discussion within the academic community.

The co-authors makes use of 30 AMS-14C age determinations. However, the stratigraphic sequence shows rapid age increases within 1–2 m of the near-surface, and significant discrepancies exist between different dating methods, as reported in previous studies (e.g., Yu et al., 2018). The authors are encouraged to strengthen their comparison of multiple dating approaches in the Discussion. This would help clarify the reliability of age constraints for paleoseismic events. Reference should also be made to recent work on regional comparative dating of paleosols in Northeast China, which may provide valuable methodological insights.

Response: Thank you very much for the reviewer’s insightful suggestions.

Although the AMS-¹⁴C samples collected from the current cultivated layer at the trench surface yield near-modern ages, the stratigraphy rapidly transitions to late Pleistocene deposits at depths of 1–2 meters, which makes the identification of individual earthquake events challenging. As noted in the Discussion section, the chronological constraints from only two trenches near Fangzheng County—originally discovered by Professor Wei Min—are insufficient to precisely determine the timing of the latest event. Therefore, assigning the year 1700 as the date of this event remains highly uncertain.

In this study, we instead relied on remote sensing interpretation and newly excavated trenches from a previously unexamined section to provide improved temporal constraints. The results suggest that the latest event occurred earlier than 2,000 years ago, consistent with archaeological evidence from the Han Dynasty period in the region.

Regarding whether the existing chronological samples are sufficient to constrain the recurrence intervals of multiple paleoseismic events, as explained in our response to Question 1, we intend to present a comprehensive discussion of these results in a separate publication.

The paper emphasizes the existence of a unified 160 km rupture segment in Northeast China, suggesting that it may produce an event comparable in scale to the 1668 earthquake. Given that the recurrence interval of Mw ≥ 8 earthquakes at plate boundaries is only ~500 years, and that far-field Coulomb stress continues to load this region (Shao et al., 2018), the authors should further clarify the significance of such a rupture under present conditions of low slip rates and long recurrence intervals. This clarification would help contextualize the necessity and urgency of assessing the large-earthquake potential of the Fangzheng–Tangyuan fault segment.

Response: Thank you very much for the reviewer’s insightful comments!

The 160 km surface rupture length confirmed in the Discussion section of this paper is supported by two lines of evidence: (1) the presence of linear fault scarps extending along the terraces of the Songhuajiang River valley, which correspond well with the interpretation of petroleum geophysical profiles; and (2) both previous studies and the latest trenching results presented in this study indicate that a large-magnitude earthquake occurred during the late Holocene. Constrained by 30 AMS-¹⁴C dating samples, the evidence suggests that this event did not occur since 1,700 a BP, but more likely sometime after that period.

Given the limited number of dated trench exposures currently available in Northeast China, it remains challenging to determine the paleoseismic recurrence interval for the northeastern segment of the TLF fault zone. Based on the general pattern of “low slip rate and long recurrence interval” identified by Li et al. (2019) in the Shandong segment, we infer that the northeastern segment may exhibit a similar behavior. However, owing to its proximity to the plate boundary, the Coulomb stress changes associated with coseismic and interseismic loading from large subduction earthquakes could potentially trigger the next major event on the YYF earlier than expected (Shao et al., 2016). This remains an inference based on stress accumulation processes and requires further verification.

As this paper primarily focuses on confirming the geological and seismological significance of the long surface rupture zone, a detailed discussion on the recurrence interval of paleoseismic events will be presented separately in a forthcoming publication, in conjunction with our previous trenching investigations.

Thanks again for your advice.

To remote sensing data, the UAV-based geomorphic analysis near the trenches is insufficiently presented. High-resolution geomorphic evidence supporting the reported fault displacements is also lacking. Adding one or two supplementary UAV or satellite images would greatly strengthen the geomorphological basis of the paper’s conclusions.

Response: Thanks for your advice, we have updated several figures with adding subfigures to cover this point.

Fig.4

Figure 4 (a) GF-1 satellite imagery acquired after snowfall at high latitudes clearly reveals a linear scarp along the traces of YYF near Xibeitun village (indicated by the yellow arrow), red rectangle showing the location of trenches in this study. See location at Figure 2b. The linear scarp extends along different terraces on the left bank of the Songhuajiang River, with the scarp trending northwest and striking Northeast. The linear scarp is also clearly visible on the floodplain of a Songhuajiang River tributary, indicating its Holocene activity. (b) and (c) Field photographs show the continuity of the scarp with several meters in vertical height. (d) and (e) showing the linear scarp via UAV-DEM, the yellow arrows show the locations of trenches in this study.

Fig.7

Figure 7 UAV-DEM showing the clear linear fault scarp connecting the previous study around Xinxin Reservoir (a), Trench od TCQH crossing the scarp (b), original photo mosaic (c, flipped) and interpretation results (d, flipped) of a trench (No. TCQH) section excavated along the fault scarp at Xinxing Village, Qinghe Township (46.175351°N, 129.246225°E). The trench of TCQH located northeast of the Trench of TC10 excavated by [25], serves as a key trench for establishing the consistency of late Holocene activity of the YYF around Fangzheng County with that of Yilan and Tangyuan Counties. Four AMS-¹⁴C samples near the surface sedimentary layers of the trench log confirm the latest activity event in the late Holocene.

5.The manuscript would benefit from a more comprehensive discussion of Holocene TLF activity in adjacent regions of Russia if relevant studies are available.

Response: Thank you very much for the reviewer#1’s constructive comments.

The both corresponding author of this paper visited Khabarovsk in 2016 and held academic exchanges with scholars from the Far Eastern Federal University in Russia. To date, aside from several seismic safety assessments conducted for oil and gas pipelines, our literature search has not identified any published evidence confirming Holocene activity of the TLF within Russian territory.

Nevertheless, remote sensing interpretation indicates that the TLF extends continuously from the Jiamusi–Hegang area in China into Russia, maintaining a well-defined linear geomorphic expression. This suggests that the fault zone likely remains structurally continuous across the border. In the future, further investigations on this segment could be advanced through Sino–Russian scientific cooperation in earthquake research.

As this paper focuses specifically on the segment of the TLF within China, no content beyond the national boundary is included in the current version.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The submitted paper presents new evidence of Holocene faulting along the Fangzheng–Tangyuan section of the Tancheng–Lujiang Fault Zone (TLF), based on trenching and remote sensing. The identification of a ~160 km-long late Holocene scarp and its comparison with the 1668 Tancheng earthquake rupture is an important contribution, with significant implications for seismic hazard in eastern China.

The manuscript also raises the possibility of far-field Coulomb stress influences from Pacific Plate subduction, which enhances its broader tectonic relevance. However, the current version requires major revision. The current RS related figures are insufficient to convincingly demonstrate the continuity of the scarp, and additional high-resolution imagery should be included. Trench documentation also needs improvement through DEM maps and integrated imagery. The chronology is currently underdeveloped; greater emphasis should be placed on integrating OSL and AMS-¹⁴C results, while excavation timelines and dating samples should be more clearly presented in a supplementary table. Furthermore, the discussion of historical magnitudes, rupture segmentation, and far-field stress effects should be strengthened with more systematic engagement with recent seismological and tectonic literature.

Regarding the remote sensing interpretation, although the co-authors point out the existence of a continuous surface scarp in the segment of Fangzheng-Tangyuan North, and that the spatial location of the newly formed fault zone within the Fangzheng Rift (graben) is consistent with that of the fault zone within the basin interpreted by the petroleum geophysical profile, for the sake of the completeness of the remote sensing interpretation, it is recommended that one or two new figures to clearly demonstrate the continuity of the ~160 km surface rupture zone. This could be presented as a series of high-resolution satellite imagery strips or through a combination of local zooms using UAV-DEM.

The current version would benefit from more detailed characterization of the local deformation scarps at the newly excavated trenches. High-resolution DEM maps, combined with drone imagery and satellite data, should be used to illustrate the linear topography near the XBT and QHT trenches. Incorporating such maps alongside trench profiles would improve the demonstration of consistency between topographic features and recent faulting.

While the discussion section of the paper mentions the difficulty in obtaining dating material from the extensive trench sections excavated on the TLF, resulting in significant uncertainty in the confirmed interval of the latest earthquake event, and given the paucity of historical documentation in the study area over the past 2,000 years, further discussion is recommended on how to integrate OSL data with the existing data. The AMS-¹⁴C dating method increases the ability to constrain estimates of multiple earthquake events revealed within a single trench in the TLF since the Late Pleistocene.

4.The excavation dates of the newly added trenches in the paper span multiple time periods from 2017 to 2020, which can be inferred from the sample delivery times listed in the table. It is recommended that the necessary explanation of the working time of each trench be included in the revision. Furthermore, it is recommended that the trench-dated samples cited in previous research results be organized into a supplementary table. This not only addresses the issue of event constraints due to dating data in Comment 3 above but also provides important support for the paper's discussion of the consistency of the latest events confirmed within multiple trenches.

Regarding the relationship between magnitude assessment and rupture scale for the AD 1668 Tancheng EQ, the current version points out the significant uncertainty in the magnitude (Ms) of historical earthquakes calculated based on the maximum intensity at the epicenter given the damage. While this is not the focus of the paper, it does have implications for the seismogenic capacity of the entire TLF (Mw). Therefore, it is recommended to supplement the latest seismological literature on the TLF in North China and to strengthen the systematic citation of literature on small earthquake rupture gaps in this paper.

Author Response

The submitted paper presents new evidence of Holocene faulting along the Fangzheng–Tangyuan section of the Tancheng–Lujiang Fault Zone (TLF), based on trenching and remote sensing. The identification of a ~160 km-long late Holocene scarp and its comparison with the 1668 Tancheng earthquake rupture is an important contribution, with significant implications for seismic hazard in eastern China. The manuscript also raises the possibility of far-field Coulomb stress influences from Pacific Plate subduction, which enhances its broader tectonic relevance. However, the current version requires major revision. The current RS related figures are insufficient to convincingly demonstrate the continuity of the scarp, and additional high-resolution imagery should be included. Trench documentation also needs improvement through DEM maps and integrated imagery. The chronology is currently underdeveloped; greater emphasis should be placed on integrating OSL and AMS-14C results, while excavation timelines and dating samples should be more clearly presented in a supplementary table. Furthermore, the discussion of historical magnitudes, rupture segmentation, and far-field stress effects should be strengthened with more systematic engagement with recent seismological and tectonic literature.

Response: Thank you very much for the constructive general comments from Reviewer #2. We give the detailed response below.

Regarding the remote sensing interpretation, although the co-authors point out the existence of a continuous surface scarp in the segment of Fangzheng-Tangyuan North, and that the spatial location of the newly formed fault zone within the Fangzheng Rift (graben) is consistent with that of the fault zone within the basin interpreted by the petroleum geophysical profile, for the sake of the completeness of the remote sensing interpretation, it is recommended that one or two new figures to clearly demonstrate the continuity of the ~160 km surface rupture zone. This could be presented as a series of high-resolution satellite imagery strips or through a combination of local zooms using UAV-DEM.

Response: Thank you very much for your advice!

We had updated two figures (Fig.4 and Fig.7) to show the clearly fault scarp, and also the relationship between the trench sites and the scarps.

Fig.4

Fig.7

Thank you very much for the reviewer’s detailed comments. As shown in Figure 7a, the southwest side of Xinxing Reservoir corresponds to the endpoint of the approximately 70 km linear fault section identified by Min et al. (2013), while the northeast side of the reservoir marks the location of the TCQH trench excavated in this study. The trench site intersects the fault scarp, and the fault plane revealed in the trench profile aligns well with the surface scarp position.

This correspondence clearly demonstrates that, although Xinxing Reservoir is a man-made feature and the fault scarp morphology within the inundated area is no longer distinct, the linear geomorphic features on both sides of the reservoir remain consistent. This evidence effectively addresses the reviewer’s concern regarding fault continuity in this area.

For the linear characteristics of the area north of the TCQH site, please refer to Figures 4, 5, and 6, in which Figure 4 incorporates additional UAV-derived DEM data near the fault zone to enhance interpretation accuracy.

The current version would benefit from more detailed characterization of the local deformation scarps at the newly excavated trenches. High-resolution DEM maps, combined with drone imagery and satellite data, should be used to illustrate the linear topography near the XBT and QHT trenches. Incorporating such maps alongside trench profiles would improve the demonstration of consistency between topographic features and recent faulting.

Response: Thank you very much for your advice! As shown above, both Figs. 4 and 7 provide detailed views of the UAV-DEM.

While the discussion section of the paper mentions the difficulty in obtaining dating material from the extensive trench sections excavated on the TLF, resulting in significant uncertainty in the confirmed interval of the latest earthquake event, and given the paucity of historical documentation in the study area over the past 2,000 years, further discussion is recommended on how to integrate OSL data with the existing data. The AMS-14C dating method increases the ability to constrain estimates of multiple earthquake events revealed within a single trench in the TLF since the Late Pleistocene.

Response: Thank you very much for your advice!

Thank you very much for the reviewer’s valuable comments. In response to Comments 1 and 2, we would like to clarify that this paper primarily focuses on interpreting multi-source satellite remote sensing data along the fault zone to identify the distribution of continuous linear scarps in the Fangzheng–Tangyuan section. By constraining the chronology of the latest event revealed in multiple trenches, our analysis confirms the continuity of late Holocene fault activity.

Although previous researchers have used different types of dating samples from individual trenches, this study emphasizes constraining the timing of the most recent event rather than conducting a detailed uncertainty analysis of various dating methods for paleoseismic interpretation. To maintain the focus on the paper’s core objectives, we have cited relevant previous studies and further refined the 30 AMS-¹⁴C chronological results presented herein (Fig. 13).

We sincerely appreciate the reviewer’s insightful suggestion. In our future work, we plan to incorporate a more detailed discussion of methodological uncertainties and to further explore the integration of multi-technical approaches for studying weakly active tectonic regions at a regional scale.

Figure 13 The most recent credible earthquake in the Fangzheng-Tangyuan section of YYF is estimated to have occurred between 1500 and 3000 years ago (red dashed rectangle), with a median age of approximately 2200 years, which is 500 years later than the initial trenching results [20-26]. Although the Holocene black soil in Northeast China is relatively thick, there is a lack of chronological samples that can be used for trenching to accurately date earthquake events. A comprehensive comparison of multiple trenches is needed for an exhaustive definition.

The excavation dates of the newly added trenches in the paper span multiple time periods from 2017 to 2020, which can be inferred from the sample delivery times listed in the table. It is recommended that the necessary explanation of the working time of each trench be included in the revision. Furthermore, it is recommended that the trench-dated samples cited in previous research results be organized into a supplementary table. This not only addresses the issue of event constraints due to dating data in Comment 3 above but also provides important support for the paper's discussion of the consistency of the latest events confirmed within multiple trenches.

Response: Thank you very much for the reviewer’s constructive comments. In fact, we have also compiled and tabulated the relevant chronological data from previously published studies. However, since the primary focus of this paper is not a comprehensive analysis of trench paleoseismic events, the detailed age data for the northeastern segment of the TLF have already been thoroughly documented in the study by Yu et al. (2018, JAES).

Accordingly, in this paper we present only the summarized age results in Table 1, which serve as reference data for subsequent analyses and for future research by other investigators.

Regarding the relationship between magnitude assessment and rupture scale for the AD 1668 Tancheng EQ, the current version points out the significant uncertainty in the magnitude (Ms) of historical earthquakes calculated based on the maximum intensity at the epicenter given the damage. While this is not the focus of the paper, it does have implications for the seismogenic capacity of the entire TLF (Mw). Therefore, it is recommended to supplement the latest seismological literature on the TLF in North China and to strengthen the systematic citation of literature on small earthquake rupture gaps in this paper.

Response: Thank you very much for the reviewer’s insightful comments.

The reviewer raises an important question regarding the calibration of historical and paleoseismic magnitudes. Since the advent of modern seismographs in the early 1900s, the global seismic network was initially sparse, and the use of surface-wave magnitude (Ms) led to saturation effects at higher magnitudes. As a result, many instrumentally recorded earthquakes from the early to mid-20th century were overestimated, particularly strong intraplate events such as the 1920 Haiyuan earthquake and the 1931 Fuyun earthquake in China (Ou et al., 2020, JGR; Liu-Zeng et al., 2023).

Historical earthquake magnitudes, which were derived empirically from the maximum observed intensities rather than instrumental data, are further affected by issues such as uneven network coverage and amplification effects at observation sites. Consequently, the empirical formulas used for magnitude estimation tend to overstate earthquake sizes, leading to systematically higher values in historical earthquake catalogs both in China and globally.

Since the 1980s, the moment magnitude scale (Mw) has been widely adopted to better reflect the seismic moment and energy release of an earthquake, particularly improving magnitude estimation for intraplate events. For instance, the 1668 Tancheng earthquake, previously estimated at approximately M8.5, has been recalibrated to Mw 7.5. Although such events remain capable of causing extensive damage, scientific rigor requires that earthquake magnitudes and parameters be interpreted objectively, without exaggeration.

We sincerely appreciate the reviewer’s thoughtful suggestion. This issue has already been discussed in the Discussion section, and relevant parts of the manuscript have been further refined in this revision to improve clarity and accuracy.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

Dear Editor and authors,

in the following we present our combined formal review of the article entitled “Using Multisource Remote Sensing data to Confirm Late Quaternary Segment Faulting Behavior of Yilan-Yitong Fault and its Potential Seismic Hazards, NE China”, by Wei et al.

The manuscript addresses the detection of further unknown sectors of important faults in the eastern China. The authors use a multidisciplinary approach based on the analysis of optical images integrated with trenching and radiometric dating to provide stronger constraints of the possible seismogenic activity of the Yilan-Yitong Fault. The authors were able to strength the possible age of the latest seismic event along the fault and improve the seismic hazard knowledge of the area.

The manuscript is well written and organized, tables and figures well display the concepts, results and discussions. The applied methodology and the used data is appropriate for the scope of the author’s analysis.

I have only minor suggestions, whose annotations are provided in the attached PDF.

Based on this, I recommend this manuscript for publication in the Remote Sensing Journal pending minor revisions.

Best Regards,

Comments for author File: Comments.pdf

Author Response

n the following we present our combined formal review of the article entitled “Using Multisource Remote Sensing data to Confirm Late Quaternary Segment Faulting Behavior of Yilan-Yitong Fault and its Potential Seismic Hazards, NE China”, by Wei et al.

I have only minor suggestions, whose annotations are provided in the attached PDF.

Based on this, I recommend this manuscript for publication in the Remote Sensing Journal pending minor revisions.

Response: Thank you very much for your positive comments and valuable advice in detail! We followed your detailed comments in the PDF file and provided responses to each point below in detail.

Other minor comments:

Line 34: “displays pronounced segmental activity” should be replaced with “is characterized by segmentation”.

Response: Updated, Thanks for your kind advice!

Line50-51: Here the authors could add specific references for further destructive earthquakes occurred in slow-slip area worldwide. An interesting example is the Morocco earthquakes as discussed by Carboni et al 2025 ( https://doi.org/10.1016/j.jsg.2025.105394 ). A review paper about earthquake environment characteristics is also offer by Molnar 2020 (https://doi.org/10.1029/ 2019JB019335 )

Response: We added the key references in the list.

Line61-69: This is a figure caption, therefore it should only contain references to the figure. All the descriptions about the historical seismicity must go in the text, referring to Figure 1.

Response: We thank the advice from Reviewer#3. We think that keeping enough caption to help readers understand the key information and the data source, so we do not move the content to the text.

Line 83-84: “this statement needs a reference”

Response: updated.

The TLF has experienced episodes of strike-slip, normal-slip, and compressional faulting [1,36-38], although its present slip rate is relatively low [2,24]. The distribution of large earthquakes along the fault zone is highly heterogeneous. The North China section has produced the Anqiu earthquake (M=7.0, 70 BC) and the Tancheng earthquake (M=8.5, 1668), while the Bohai Bay area, where the fault intersects the ZBSZ, has hosted multiple strong offshore earthquakes. By contrast, the Northeast China section has experienced only a limited number of moderate events (M < 6) in the past 120 years, leading some Chinese scholars to assume the absence of active faults in this region initially [38].

Line 105:”in such a framework of hidden faults, it might be possible to define this as a possible minimum, since further sectors of this fault might be buried but still active?”

Response: updated.

This study examines whether the Fangzheng–Yilan Holocene rupture zone extends northeastward. Through a combination of remote sensing interpretation and trench investigations, we determine the minimum rupture length of the YYF (for some sections, maybe buried but still active in the near future), establish the timing of its most recent paleo-earthquake event, and assess its potential risk based on estimation by moment magnitude calculation [39-40]. The results provide new insights into the seismic hazard of the YYF and contribute to more robust seismic risk assessments for Northeast China.

Line 125: ”I would suggest to make the fault lines thicker to be more visible”

Response: updated.

Figure 2 (a) Simplified distribution map of the TLF (the part located in China). The relatively continuous Holocene active fault segments (marked as A-C) found in the northeastern section of the TLF are situated in northeast Jilin Province and Heilongjiang Province. However, the longer stretch from Shenyang City to Changchun (marked as D) shows no clear evidence of recent activity. The North China section of the TLF also contains localized earthquake-deficient gaps (marked as E) [33]. Note that the continuous rupture segment A studied in this paper is comparable in scale to the coseismic surface rupture zone of the 1668 earthquake in the Shandong section of the TLF [38]. (b) The distribution of linear scarps and the spatial locations of the main excavated trenches along the extension of the YYF from Fangzheng County to Tangyuan County. The black rectangles represent excavations by previous researchers [21-27].

Line 140-141: “this was already stated more than once. I would suggest to remove it”

Response: Done.

During the Early Cretaceous, tectonic processes of TLF propagated northeastward into Northeast Asia, leading to the development of the Yilan–Yitong Fault (YYF) [1,37,47]. Geophysical exploration profiles indicate that the YYF extends as a lithospheric-scale structure, cutting through both the crust and upper mantle, whereas the DMF shows a less pronounced deep continuation. Consequently, the YYF is considered to play a dominant role [18,29].

Line 179-183:”this sentence better fits the acknowledgments, with the names of these famous geologists. I would remove it from here.This sentence somehow might discredit the authors, suggesting that they needed external help. I believe this is not the case, thus I would suggest to move this sentence in the acknowledgments.

Response: Done.

We excavated or cleared five trenches across locations where linear scarps were most prominent (Figure 2b). Chronological constraints on the most recent surface-rupturing events were obtained from three of these trenches [5,49]. Excavations were conducted perpendicular to the fault trace, utilizing a combination of mechanical digging, manual cleaning, and leveling. Stratigraphic relationships and fault planes exposed within the trenches were documented through detailed field photography, sketching, and on-site interpretation. Dating samples were collected from trenches to constrain the age of the stratigraphic sequence and to obtain a more precisely defined age range for earthquake events. Thirty AMS-¹⁴C were processed and tested in Beta Analytic Inc. in the USA (Table 1). Calibration of radiocarbon ages was conducted using Oxcal v4.3.2 with the IntCal 09 atmospheric model.

Line 184: “I would place this table in the supplementary materials, as here it is not a fundamental data.”

Response: Keep it for the fundamental data to support the Figs.9-13.

Line 209: “I suggest to move the panel c on top, since it also shows the fault trace. This will make the link between the different data easier”

Response: No change. The current fig.3 just shows the relationship between the surface scarp and the location of each trench. So keep it.

Line 218-223: “here the authors are describing the remote sensing data. I would move this part in an other chapter”

Response: Done.

The southwestern termination of the linear scarp is located at the confluence of the Songhuajiang River and its tributaries on the right bank, where it appears as a surface scarp approximately 0.5 m in height [25-26]. A chain of lakes is distributed along this interpreted trace of scarps, which dips to the northwest; Fangzheng Lake in Fangzheng County is a representative example (Figure 3b). After crossing the Songhuajiang River northeastward, the interpreted fault trace is expressed as a ~1 m-high scarp on the first terrace, dipping to the southeast (Figure 2b). Multiple trenches excavated in this segment between 2010 and 2014 (Figure 2b, black rectangle) [20-22,25-26] confirmed that the surface scarp corresponds to fault breakpoints exposed in the trench logs. From Xinxing Reservoir to the northwest of Yilan County, the scarp again dips northwest (Figure 4a). Four trenches were excavated in this segment in this paper; three of them provided systematic chronological constraints on the most recent paleoseismic events (Figure 4b).

Line 256: “this is the same title as the previous chapter. The authors should change the title name to reflect their description of the specific fault segments. Maybe: "Trenches interpretation" ?

Response: Done.

4.2 Trenches interpretation

4.2.1 Qinghe Trench (TCQH)

Line 266: “I believe this figure can be omitted, since it does not give any important information. No clear scarps are visible in none of the panels”.

Response: Done.

Thank you very much for the reviewer’s comments. We believe that this Fig holds significant value in confirming the 160 km extension of the surface rupture zone. Taken together, Figures 2 to 6 clearly demonstrate the importance of remote sensing interpretation in identifying and validating continuous linear fault structures. Therefore, we have chosen to retain this map in the revised manuscript.

Line 288: “the authors could leave this photo clear and empty, since they have already add the interpreted one”

Response: Done.

Figure 7 UAV-DEM showing the clear linear fault scarp connecting the previous study around Xinxin Reservoir (a), Trench od TCQH crossing the scarp (b), original photo mosaic (c, flipped) and interpretation results (d, flipped) of a trench (No. TCQH) section excavated along the fault scarp at Xinxing Village, Qinghe Township (46.175351°N, 129.246225°E). The trench of TCQH located northeast of the Trench of TC10 excavated by Shu et al. (2014), serves as a key trench for establishing the consistency of late Holocene activity of the YYF around Fangzheng County with that of Yilan and Tangyuan Counties. Four AMS-¹⁴C samples near the surface sedimentary layers of the trench log confirm the latest activity event in the late Holocene.

Line 318: “since the authors conducted UAV photogrammetry, it would be better to add the VOMs per each trench. The photo boundaries disturb the view and reduce the clarity of the figure”.

Response: Done. We had added the UAV-DEM in Fig.4.

Line 377: “Since the authors show two different portions of the trench I would suggest to add at the top panel the entire trench with 2 insets. The 2 panels in the middle showing the insets clear and further 2 panels at the bottom showing the interpretation.”

Response: Thanks for your nice advice, we think keep the location of the site and the detail information around the trace of the fault near the surface.

Line 413: “this figure should go after the chapter 4.2.4”

Response: Done! Thanks for your suggestion!

Line 457-458: “this sentence is not clear? what do the authors mean with consistent strike but changeable tendency? Along-strike sinuous geometry of the fault trace/scarp?”

Response: Done! Thanks for your suggestion!

We confirmed that the Fangzheng–Tonghe segment of YYF from the Xinxing Reservoir to the northern part of Tangyuan County (Figure 2) is laterally continuous. Evidence from five trenches (Figures 5–10) in this study further substantiates its Holocene activity, with topographic scarps directly corresponding to the exposed upper fault breakpoints. These results demonstrate that the three previously recognized independent active faults within the Fangzheng–Tangyuan section of the YYF are in fact part of a single integrated rupture segment (Figure 2), consistent with structures revealed by petroleum geophysical surveys in the graben [36].

Line 489,492,509: “this paper is now relatively outdated, and Leonard 2014 should be considered. Leonard, M., 2010. Self-Consistent Earthquake Fault-Scaling Relations: Update and Extension to Stable Continental Strike-Slip Faults. Bulletin of the Seismological Society of America, Vol. 104, No. 6, pp. 2953–2965, December 2014, doi: 10.1785/0120140087

”””

Response: Done! Thanks for your suggestion!

Regarding magnitude estimates, earlier earthquake catalogs assigned the 1668 Tancheng earthquake a magnitude of M=8.5, based on empirical relationships between maximum intensity and magnitude [52] by using the equation as follows,

M=0.58I₀ + 1.5 (1);

where I₀ means the maximum intensity of the epicenter, which is defined by Gu et al.(1983), however, this method does not account for rupture length and may have led to overestimation [5,13,53-54]. Since the mid-20th century, earthquake magnitudes have been recorded in terms of moment magnitude, M_w, with [39] providing empirical scaling relationships between rupture length and M_w for various faulting styles. Leonard (2014) proposed an updated fault-scaling relationship [40]. If the TLF is considered a strike-slip to compressional structure, the appropriate regressions are:

M_w = 5.16 + 1.12 × log(SRL) for strike-slip faults (2);

M_w = 5.00 + 1.22 × log(SRL) for thrust faults (3);

M_w= 5.08 + 1.16 × log(SRL) for all fault types (4);

M_w= 5.27 + 1.0 × log(L) for interplate strike slip faults (5);

where SRL (in km) is the Surface Rupture Length [39], L (in km) is the length of coseismic surface rupture [40].

Using a rupture length of 160 km for the 1668 event yields M_w=7.6–7.7, significantly smaller than the previously assigned M=8.5. Even if the maximum rupture length of 220 km is adopted, the calculated magnitude is M_w≈7.8, only marginally larger. This is consistent with the magnitude of the 1920 Haiyuan earthquake [5,13]. These results underscore the need to reassess the source parameters of historical large earthquakes, particularly magnitude, in terms of rupture scale and seismic moment release. For the ~155 km length of the Fangzheng-Tangyuan segment of the YYF, the estimated moment magnitude is M_w≈7.6, directly comparable to the magnitude of the most significant historical event along the TLF. This value is significantly larger than the previous magnitude of M=7.5, based on the last recorded length of 70 km [55].

Line 549: “Who?”

Response: Done! Thanks for your suggestion!

In this study, based on three trenches excavated between Xinxing Reservoir and Yilan County, we yielded a relatively high number of age results, constrained the age of the latest earthquake event to be between 990±30 BP (from Trench TCAL) and 6170±30 BP (from Trench TCQH). Yu et al. (2018) constrained the E1 event near Tangyuan to between 1500±600 BP (OSL sample age) and 2680±30 BP (AMS-¹⁴C sample age) (Table 1) [23-24]. Combining the latest results of E1 event constraints from different researchers, we believe that the most recent event (E1) of the largest rupture segment on the YYF should be constrained to a time range of between 1500 and 3270 BP. Recent regional archaeological findings from the Fangzheng-Tangyuan rupture section of the YYF confirm that the destruction of numerous sites dating from the Han Dynasty to the Wei, Jin, and Southern and Northern Dynasties along the Songhua River in Fangzheng-Yilan-Tangyuan-Jiamusi, located along the YYF, does not involve a large-scale, concentrated, strong earthquake as a causal explanation. Nor has any soil liquefaction been found along the river's low terraces, which could be associated with a decisive earthquake rupture [58-60]. This suggests that site destruction in China since the Han Dynasty is unrelated to the adjacent YYF destruction. Therefore, no recent reports of YYF-induced destruction have been found in archaeological excavations of numerous sites in the Songhuajiang River and Mudanjiang River basins since the Han Dynasty, which supports the conclusion that the lower limit of the latest event, 1730 ± 40 BP, is not appropriate. Its lower limit of age may be in the middle and late Holocene, which was not recorded in historical documents of the time. It is reasonable that it was not recorded in archaeological sites after the Han Dynasty.

Author Response File: Author Response.pdf