the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 10 Dec 2024
Insight into the tectonostratigraphy of the historic Kefalonia island (Greece): a reflection seismic survey
Abstract. Kefalonia island, in front of the Greek west coast, is placed in a peculiar tectonic setting characterized by a transition from an oceanic subduction contact to a continental collision. This tectonic setting results in strong tectonic activities and seismicity in the area making the island a testbed for geological, geophysical, and archeological studies. To improve the subsurface knowledge and shed light in the top 100s of meters, we acquired three seismic profiles in the isthmus connecting the main part of the island to the Paliki peninsula, in the Thinia valley, where the presence of a possible channel has been disputed to make Paliki the Homer’s Ithaca (home of Odysseus). A total of approximately 3.5 km of seismic data was acquired using 5 m receiver and shot spacing and a 25 kg accelerated weight-drop as the main source. The sharp topographic changes and morphological features of the valley made the survey challenging, limiting the spread, precluding uniform shot points, and resulting in strongly crooked profiles. The acquired data, however, show visible reflections with variable quality down to 0.5 s and occasionally to 1 s. First-break traveltime tomography and 3D reflection traveltime modelling were performed to complement the seismic reflection processing work together with lithological columns from three boreholes present along the profiles. Results show a low-velocity zone with no reflectivity from the surface to approximately 100 m depth probably related to the presence of loose material, under which two main east-dipping reflections are imaged. With the help of surface geology and tectonic history of the valley, we interpret these features as the same lithological boundary displaced by three highly east-dipping thrust/reverse faults probably part of the Hellenide thrusts. These findings further constrain the local recent tectonic history and thus, the long-debated presence of an historic water channel in the valley.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(4149 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(4149 KB) - Metadata XML
- BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2024-3783', Edward Marc Cushing, 17 Jan 2025
Dear authors, dear editor,
I have carefully reviewed the proposed preprint: “Insight into the tectonostratigraphy of the historic Kefalonia island (Greece): a reflection seismic survey.” My comments and observations are based on regional knowledge and the relevant literature. As seismic processing is not my area of expertise, my feedback focuses primarily on the geological interpretation proposed in this preprint. I do not assess the validity of the hypotheses that motivated this work, being aware of its potentially contentious nature.
Nonetheless, I note that the paper has the merit of providing valuable geophysical and structural information on the Thinia area and, as such, deserves to be published. This study complements the extensive and remarkable work presented in Kirsten Hunter's dissertation, which I strongly recommend the authors consider carefully, particularly regarding the use of her illustrations, as they would greatly enhance the clarity of the paper.
I therefore encourage the authors to improve and refine the figures, maps, and cross-sections at appropriate scales, enabling readers to better locate the investigations conducted and to better understand the structural interpretations they propose.General Comments
The authors present a geophysical study based on seismic reflection and seismic tomography methods. After briefly describing the geological, seismic, and structural environment of the island of Cephalonia, the authors focus on a specific area within the Thinia Valley. This area has been the subject of numerous studies since the 2000s aimed at exploring a theory linked to the myth of Homer’s Odyssey, namely the location of Odysseus’ palace.
The main body of the text provides a fairly detailed description of the geophysical methods used to image the studied structures. Using seismic reflection and tomography techniques, the study reveals deep structures up to 500 meters beneath the surface. The results highlight intense tectonic activity characterized by thrust faults and complex deformations. Although the data is limited by the variable quality of the surveys, it indicates the presence of unconsolidated sedimentary material, likely originating from landslides, and east-dipping structures. The valley is described as a syncline, emphasizing the need for further research to refine geological models and understand tectonic interactions.
The scope of the article aligns well with the theme of the Special issue: « Seismic imaging from the lithosphere to the near surface » and represents its most developed aspect. Not being a specialist in seismic data processing, I do not address this section in detail, but I highlight a few points that might need verification.
The imaging obtained from the geophysical measurements appears to be of relatively good quality, considering the complex tectonic context of the area. It reveals duplicated thrust structures, which is consistent with the structural framework of Cephalonia’s thrust system (see Sorel; Underhill, Cushing). The additional value of the study lies in the interpretation of the upper layer velocity (Vp), which corroborates the presence of slipped formations as suggested in previous studies.
The authors remain cautious in interpreting this seismic imaging with respect to the "archaeological" hypotheses.
Some figures are too small or lack sufficient detail. The paper would benefit from improved figures (maps, legends, etc.), as detailed below.Individual Scientific Questions/Issues ("Specific Comments")
Title
The title is somewhat ambiguous. I would suggest focusing on the specific study area, as the current title may lead readers to expect a general study of the island. Additionally, I recommend emphasizing the contribution of geophysical methods to the structural understanding of this area. The term “Historical” is ambiguous. The “archaeological” investigations aim to confirm a hypothesis derived from the interpretation of a text referencing a myth (Odysseus’), which is not strictly historical.
• L24: Do you have any information about the age of the presumed landslide? It seems that Hunter’s study did not confirm any activity younger than 1.8 Ma. The dating from coring appears to be highly questionable.
• L40: You should cite the following references: Stiros et al. (1994), Sorel (1988, 1976).
• L41: You mention "dynamic" motions (translational from GPS) and rotational movements. Please clarify further.
• L55-57: The relationship between the studied fault structure and the presumed landslide is unclear. Better understanding the tectonic history alone is not sufficient. The key question seems to be: what is the structural (and potentially causal) relationship between the fault structure (folds and faults) and the overburdened landslide? You should also explain how this study contributes beyond the comprehensive work of Kirsten Hunter (2013).
• L60: Indicate that these investigations build upon and complement the comprehensive studies performed in Hunter’s thesis, which is detailed extensively. You may list all methods and datasets used in Section 3.
• L64: The "Aeanos Thrust" is a structural feature identified by Underhill (1989), but other names and traces exist in the literature (e.g., Kontogourata-Agon Fault). Refer also to Lekkas et al. (2016), Lagios et al. (2012), Stiros (1994), and Sorel (1976). Note that Sorel (1976), had already identified this structure as a reverse fault as many faults on Kefalonia Island.
• L66: “The area is approximately 773 km²”. This value is “accurate” and not approximative.
• L70 (Figure 1): While the text includes a topographic description, an altimetric map is clearly missing. Consider using a DEM or a topographic map from Hunter’s thesis (e.g., Figure 4.1).
• L74: In the Poros town area and eastern coast.
• L83: Figure 1b appears to be missing or is not easily located.
• L89: These faults are interpreted as ancient normal faults within the Preapulian-Paxos zone, which underwent tectonic inversion during the lower Pliocene and lower Pleistocene (Sorel, 1976, 1988).
• L97 (Figure 1): As previously noted, there is no hypsometric map. The red inset is very small and difficult to read.
• L100: Explain in the text why this specific zone was chosen. It is one of the locations studied by Hunter (paleolake of Katokhori) and the site of other investigations, such as drill holes, gravimetry, ERT, seismic refraction, and airborne magnetism. This study builds on and complements previous investigations.
• L114 (Figure 2): This figure does not provide information on the profile locations. There is no way to locate these profiles. Include a detailed, large-scale map with orientation, scale, contour lines, and coordinates, similar to Figure 2.9 in Hunter’s thesis.
• L167: Why not consider velocities in the range of 2400 to 3300 m/s, as previously mentioned?
• Figures 3 and 4: It is difficult to clearly identify R1 and R2.
• L187-188: These velocities differ again from those mentioned earlier. Please explain this discrepancy or modify.
• L280 (Figure 8): A comparative geological cross-section and a detailed, large-scale geological map are missing. The figure is too small. Consider using the geological map from Figure 3.29 in Hunter’s thesis and the cross-section from Figure 5.15, which includes two of the three boreholes (C5C, C4a). Indicate the location of the landslide referenced in Line 311 on both the cross-section and geological map.
• L318: Based on Hunter’s thesis conclusion, the hypothesis of a covered channel is ruled out by her data analysis. The only remaining possibility seems to be that a large landslide, encompassing the entire slope, could have carried and possibly destroyed the supposed channel. Refer to Section 462 of Hunter’s thesis for further details.Technical corrections
References:
• L 39: (Gaki-papanastassiou et al., 2010 or 2011 ?) see ref listCitation: https://doi.org/10.5194/egusphere-2024-3783-RC1 -
AC1: 'Reply on RC1', Samuel Zappalá, 18 Feb 2025
Dear reviewer 1,
Thank you for your valuable comments that helped us improving the quality of this paper. Most of the suggested comments have been implemented and when not possible, a detailed explanation is given.
Figures have been improved as suggested to help the readers to better locate the investigations conducted and to reach a broader audience.
Replies to individual comments:
Title R: Thanks for the suggestions, we changed the title to be less ambiguous, more concise and more focused on the geophysical aspect as follow “Reflection seismic imaging across the Thinia valley (Greece)”
L24R: Unfortunately, we do not have any information about the age. As you say, the dating in literature of the possible landslides is highly questionable and reflection seismology do not add any information regarding ages.L39R: We refer to (Gaki-papanastassiou et al., 2010). In this phrase, a list of references of past geological and geophysical works done in the island highlight their variety. This reference is reported as an example of geomorphological study in Kefalonia island.
L40R: Thanks for the suggested references. We added Stiros et al. (1994) reference. We could not find the suggested Sorel studies but we can see that they are included in later works that are already referenced in our paper, such as Hunter (2013), Karymbalis et al. (2013) and Gaki-Papanastassiou et al. (2010 and 2011)
L41R: The text has been updated as follow to add more details: “Analyses of the island’s displacements from the observed seismicity have shown complex horizontal motion that suggests two separate crustal blocks in the island, the main island with a displacement towards south-west and Paliki peninsula with a displacement towards north (Ganas et al., 2015). In addition, an unclear history of rotational movements characterized the island with probably both clockwise and anticlockwise rotational phases (Duermeijer et al., 2000; Sbaa et al., 2017).”.
L55-57R: Thanks for the suggestion, we modified the text to clarify how this study contributes beyond previous studies in the same area, including Hunter (2013), and we better defined the key question as you suggested. We included a text comparison also in the discussion section to highlight what the presented results add to Hunter work and in what they agree or differ.
L60R: The text has been updated and the relation between this study and previous studies is better indicated. Previous geophysical methods applied along the Thinia valley are now summarized in the introduction.
L64R: Thanks for the comment, other names for this structure found in literature are added to the paper to increase the audience understanding.
L66R: You are right. Thanks for pointing it out. Text updated accordingly.
L70R (Figure 1): Figure 2 was reporting the topography of the area but we agree that was not complete. Now it has been updated with a map view of the same area complete of all needed information.L74R: Thanks for the clarification. Text updated.
L83R: You are right. The figure was modified but the text was not updated. Now the text is updated according to the figures. Thanks for spotting it.
L89R: Thank for the clarification. The text has been updated to include this information as follow: “shows eastwards ancient dipping normal faults on the west of the valley which underwent tectonic inversion”.
L97R (Figure 1): elevation information added in the updated figure 2. The profiles in the red inset have been replaced with an asterisk for the whole survey while their precise locations are now reported in figure 2 with a larger scale.
L100R: Text has been updated in this chapter and in the introduction to include this work relation with previous studies.
L114R (Figure 2): Figure updated as suggested from the reviewer to include all missing information and be of easier readability for the audience.
L167R: Different velocities and parameters are always tested to ensure the best results from each processing step and in this case, this was the model resulting in the best results. In general, NMO velocities are usually different from real velocities and valid only for the NMO correction. Therefore, even if they can be used as starting reference for other velocities models (e.g. the one used for time to depth conversion) they will most likely be different. Text has been updated to include more details on the time to depth model selection.
Figures 3 and 4R: Extra arrows are added in figures 3 and 4 to help better identify the reflections.
L187-188R: See reply L167R.
L280R (Figure 8): Thanks for the comment, the geological map shown in Figure 8b has similar scale as the one suggested from Hunter (2013) and shows mostly the same features, but considers also other works as the IGME 1985 geological map. The small local features shown in Hunter’s map and not reported in Figure 8b are below our acquisition resolution. We agree that the map can be shown larger so we added a zoom in inset in Figure 8b showing the interpreted area to better visualize the details.This paper is not meant to be a review of previous works, we would not like to do a direct comparison with previous geological cross-sections since many studies (i.e., Underhill, 1989; IGME, 1985, 1996; Gaki-Papanastassiou et al. 2011) were done in the area often with different results and all of them reliable only for shallower depth than this paper. In the discussion section a paragraph is added to compare the obtained results with Hunter (2013) results and highlighting corresponding features of the geology in the area (thrusts and stratigraphy) but also the limit of comparing these different works.
We are not able to define the exact location of the probable landslide material since landslides are very localized and below our resolution and seismic velocities are an average of all materials they cross. We suggest that landslide material is probably present in the low velocity chaotic near surface layer (transparent grey in Figure 8a) since its average seismic characteristics (velocity and reflectivity) are similar to the ones typical of landslides material and different from the ones expected from only quaternary sediments. Text has been updated in the interpretation and conclusion sections to make it clearer.
L318R: Thanks for the clarification, the text has been updated to remove the connection with a covered channel as follow: “and overburden landslide materials are detected.”.
-
AC1: 'Reply on RC1', Samuel Zappalá, 18 Feb 2025
-
RC2: 'Comment on egusphere-2024-3783', Anonymous Referee #2, 20 Jan 2025
Dear Authors,
Thank you for submitting your manuscript to Solid Earth. The manuscript presents a study that utilizes seismic reflection and tomography methods to investigate the subsurface structure of Kefalonia Island.
I have evaluated this manuscript for its geophysical content. My expertise does not extend to the geological and tectonic interpretations specific to Kefalonia Island; therefore, this review excludes a detailed critique of the geological findings and tectonic implications.
While the study offers valuable insights into seismic near-surface characterization, several aspects need refinement to improve clarity and accuracy.
- The title needs rephrasing for clarity and precision.
- Lines 33, 50, 55, 58, 68, 83: You referred to Figures 1a and 1b. However, Figure 1 does not include subfigures (1a, 1b). Update Figure 1 accordingly.
- Lines 62–64: You provided your findings at the end of the introduction. The introduction should set up the study, not summarize its findings. Instead, clearly define the manuscript content. The authors should ensure that the introduction flows smoothly into the rest of the manuscript.
- Line 71: Provide full wording for the abbreviation “a.s.l.” on its first use.
- Lines 159–160: Figures 4c and 4d are incorrectly referenced; they should be moved to the correct location in this sentence to avoid reader confusion.
- Figure 6: I recommend including a 3D visualization of velocity profiles in an integrated model to validate the velocity at intersection points.
- The coherence and continuity of reflections across profiles are insufficiently validated. Including a 3D visualization of intersecting profiles (after migration) would enhance the understanding of spatial relationships. Additionally, any discrepancies between reflections at the intersections, even if minor, should be critically analyzed and discussed.
- Lines 165–166: Clarify how the migration velocity field was derived. Simply referencing only the velocity range estimated for the NMO corrections is insufficient.
- In Table 2, specify where refraction statics were applied on Profiles 1, 2, and 3.
- Profile 1 shows promising results, yet the absence of migration for other profiles undermines the reliability of structural interpretations. Why didn’t you apply migration to Profiles 2, 3, and CS? Your current reasoning is not satisfactory. These profiles exhibit dipping reflections, and post-stack migration is essential for accurate positioning and interpretation. Superimposing a near-surface velocity-depth model on stack sections is inappropriate. It is expected that these profiles be migrated, converted from the time domain to the depth domain, and tied to the borehole data (Figure 7) to ensure a more accurate and robust interpretation.
- Line 221: “the best quality” is not appropriate and scientific. Rephrase it and make this sentence comparative. For example, replace it with “higher resolution compared to other profiles.”
- Line 233: This sentence is vague and requires rephrasing. What characteristics do you mean?
- Line 257: You plotted the boreholes with respect to elevation but referred to them in terms of depth (“where the available boreholes reach a depth of 90 to 100 m”), which is inconsistent and potentially confusing. Please clarify and ensure consistency.
- Line 305: “the best result” is not appropriate and scientific. Replace it with a comparative phrase. See 11.
- Lines 307–309: In the conclusion, you stated that the 3D traveltime modeling of reflection R1, both in the shot gather and the unmigrated stack section, suggests a N-23° strike and a 44° dip angle towards the east, consistent with the local geology. However, you did not present the modeling results for the unmigrated stack section in the manuscript. Conclusions must be based on the results shown; therefore, this statement needs to be revised or supported with the missing data.
- You claimed that R1 is observed with an estimated orientation of N23°E/44°E. However, Profile 3 in Figure 7d does not show a north-dipping reflection at its intersection with Profile 1. As mentioned before, adding a 3D visualization of intersecting profiles (after migration) would provide a clearer understanding of this relationship.
Recommendation:
Overall, this manuscript has the potential to make a valuable contribution to seismic near-surface characterization. However, it requires major revisions to address the issues outlined above. I recommend resubmission after revisions.
Citation: https://doi.org/10.5194/egusphere-2024-3783-RC2 -
AC2: 'Reply on RC2', Samuel Zappalá, 18 Feb 2025
Dear reviewer 2,
Thank you for your valuable comments that helped us improving the quality of this paper. Most of the suggested comments have been implemented and when not possible, a detailed explanation is given.
Following you can find detailed replies to your comments.
- R: The title has been updated to be less ambiguous, more concise and more focused on the geophysical aspect as follow “Reflection seismic imaging across the Thinia valley (Greece)”
- Lines 33, 50, 55, 58, 68, 83R: You are right. The figure was modified but the text was not updated. Now the text is updated according to the figures. Thanks for spotting it.
- Lines 62–64R: Thanks for the comment. We removed the phrase summarizing the findings in the introduction and we better defined the manuscript content.
- Line 71R: Text updated.
- Lines 159–160R: Thanks for the comment. Text has been updated to properly match the figures.
- Figure 6R: The velocity profiles intersect only on a single point because the velocity model from profile 3 has no values at the intersection with profile 1 (see attached file). Since no 3D analysis or interpretation are carried out in this study and since a 2D display of the profiles enable to visualize more details respect than a 3D visualization, we think that a 3D visualization of the velocity models will not add any useful information to the paper while hiding important features. Still, if the reviewer insists we can add a figure with the 3D visualization of the velocity models.
- R: The coherence of reflections across profiles is not analyzed in the paper since migration of all profiles was not possible (see reply to comment 10) and as you suggest it is meaningless to compare unmigrated profiles. We do not want to plot a 3D visualization of the seismic profiles to do not give the wrong impression that the interpretation is done on all profiles. Indeed, the interpretation is done only on profile 1 and only a very cautious suggestion of eventual corresponding features is done on the other profiles that is expected to be a base for future studies and nothing more.
- Lines 165–166R: Text has been updated to give more details as follow: “first using the estimated NMO velocities, and then refining them with the evaluation of smiles and frowns in the resulting migrated sections”.
- In Table 2R: We are not sure we understand the question; refraction statics are applied at step 11 of our processing flow as expressed in table 2. If something else is not clear please let us know.
- R: We completely agree with the benefits of migrating all profiles and we largely tried before submitting the paper. Unfortunately, the low quality of the data in profile 2, 3 and CS and the absence of an accurate deep velocity model resulted on the migration of noise instead than of the reflections when migration is applied to the data. The quality of the acquired data is this for the reasons explained in the data acquisition chapter and we cannot change it. Reason why in our interpretation we only considered the results from profile 1 and we clearly stated that the quality from the other profiles is not enough for any interpretation, limiting ourselves to show the obtained results. Data from boreholes are shallow and discontinue and do not clearly intersect any main seismic reflector and therefore they cannot be used to tie the seismic sections. Text has been updated to make it clearer.
- Line 221R: Thanks for the suggestion, the text has been updated as follow: “The unmigrated stacked section of profile 1 (Figure 7b) shows a higher S/N ratio and reflections continuity with respect to the other profiles.”.
- Line 233R: The text has been rephrased for better clarity as follow: “The deepest of these reflections shows amplitude values, frequency content and shape of the signal similar to the ones from R2 reflection of profile 1 (Figure 7b) and considering its location (between 0 and -150 m a.s.l.), it could correspond to the same horizon. The reflection located at a depth of 150 to 50 m a.s.l. matches amplitude values, frequency content and shape of the signal from reflection R1 of profile 1 suggesting their correspondence.”.
- Line 257R: The text has been updated to improve its consistency as follow: “Comparison with the boreholes lithologies (Figure 6) is more meaningful for profile 2 (Figure 6b) where the available boreholes are 90 to 100 m deep and reach an elevation of approximately 60 and 70 m a.s.l. (Figure 6b).” and “. Borehole C5a along profile 1 (Figure 6a) is only 30 m deep and reach an elevation of approximately 150 m a.s.l. (Figure 6d) showing low velocities corresponding to the logged marl.”.
- Line 305R: Thanks for the suggestion, the text has been updated for a more scientific language as follow: “profile 1 showed higher S/N ratio and reflection continuity respect to the other acquired profiles”
- Lines 307–309R: You are right and thanks for noting it. A phrase regarding the modeling in stack domain is added to the text (section 4.3) and the interpretation text is updated accordingly. The resulting modeled horizon is shown in figure 8.
- R: As we said in the text, the north part of profile 3 is not reliable because of the source issue (visible also from the missing traveltime tomography) and therefore a comparison with profile 1 will be meaningless, without considering the migration issue detailed in comment 10. Furthermore, the N23°E/44°E means a slight dipping towards south and not towards north, with the strike almost parallel to profile 3 that should result on an almost horizontal apparent dip.
-
EC1: 'Comment on egusphere-2024-3783', David Snyder, 22 Jan 2025
Reviews by two referees indicate that the manuscript could be acceptable for this special volume with some significant revision. The seismologist requests clarification and more details of the processing using the entire data set available to the authors. The interpreter suggests less speculation about archeological significance but greater discussion about geomorphology/landslides. Both request more minor clarifications in numerous locations. All of these comments seem reasonable to me and would greatly improve the manuscript and make it a more substantial contribution.
Citation: https://doi.org/10.5194/egusphere-2024-3783-EC1 -
AC3: 'Reply on EC1', Samuel Zappalá, 18 Feb 2025
Dear editor,
thanks for the opportunity to improve our manuscript implementing the comments and suggestions provided by the reviewers. We really appreciate the time and effort that the reviewers placed on this manuscript and we addressed their comments in the revised paper. Detailed responses to each comment are provided in the discussion section.
Citation: https://doi.org/10.5194/egusphere-2024-3783-AC3
-
AC3: 'Reply on EC1', Samuel Zappalá, 18 Feb 2025
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2024-3783', Edward Marc Cushing, 17 Jan 2025
Dear authors, dear editor,
I have carefully reviewed the proposed preprint: “Insight into the tectonostratigraphy of the historic Kefalonia island (Greece): a reflection seismic survey.” My comments and observations are based on regional knowledge and the relevant literature. As seismic processing is not my area of expertise, my feedback focuses primarily on the geological interpretation proposed in this preprint. I do not assess the validity of the hypotheses that motivated this work, being aware of its potentially contentious nature.
Nonetheless, I note that the paper has the merit of providing valuable geophysical and structural information on the Thinia area and, as such, deserves to be published. This study complements the extensive and remarkable work presented in Kirsten Hunter's dissertation, which I strongly recommend the authors consider carefully, particularly regarding the use of her illustrations, as they would greatly enhance the clarity of the paper.
I therefore encourage the authors to improve and refine the figures, maps, and cross-sections at appropriate scales, enabling readers to better locate the investigations conducted and to better understand the structural interpretations they propose.General Comments
The authors present a geophysical study based on seismic reflection and seismic tomography methods. After briefly describing the geological, seismic, and structural environment of the island of Cephalonia, the authors focus on a specific area within the Thinia Valley. This area has been the subject of numerous studies since the 2000s aimed at exploring a theory linked to the myth of Homer’s Odyssey, namely the location of Odysseus’ palace.
The main body of the text provides a fairly detailed description of the geophysical methods used to image the studied structures. Using seismic reflection and tomography techniques, the study reveals deep structures up to 500 meters beneath the surface. The results highlight intense tectonic activity characterized by thrust faults and complex deformations. Although the data is limited by the variable quality of the surveys, it indicates the presence of unconsolidated sedimentary material, likely originating from landslides, and east-dipping structures. The valley is described as a syncline, emphasizing the need for further research to refine geological models and understand tectonic interactions.
The scope of the article aligns well with the theme of the Special issue: « Seismic imaging from the lithosphere to the near surface » and represents its most developed aspect. Not being a specialist in seismic data processing, I do not address this section in detail, but I highlight a few points that might need verification.
The imaging obtained from the geophysical measurements appears to be of relatively good quality, considering the complex tectonic context of the area. It reveals duplicated thrust structures, which is consistent with the structural framework of Cephalonia’s thrust system (see Sorel; Underhill, Cushing). The additional value of the study lies in the interpretation of the upper layer velocity (Vp), which corroborates the presence of slipped formations as suggested in previous studies.
The authors remain cautious in interpreting this seismic imaging with respect to the "archaeological" hypotheses.
Some figures are too small or lack sufficient detail. The paper would benefit from improved figures (maps, legends, etc.), as detailed below.Individual Scientific Questions/Issues ("Specific Comments")
Title
The title is somewhat ambiguous. I would suggest focusing on the specific study area, as the current title may lead readers to expect a general study of the island. Additionally, I recommend emphasizing the contribution of geophysical methods to the structural understanding of this area. The term “Historical” is ambiguous. The “archaeological” investigations aim to confirm a hypothesis derived from the interpretation of a text referencing a myth (Odysseus’), which is not strictly historical.
• L24: Do you have any information about the age of the presumed landslide? It seems that Hunter’s study did not confirm any activity younger than 1.8 Ma. The dating from coring appears to be highly questionable.
• L40: You should cite the following references: Stiros et al. (1994), Sorel (1988, 1976).
• L41: You mention "dynamic" motions (translational from GPS) and rotational movements. Please clarify further.
• L55-57: The relationship between the studied fault structure and the presumed landslide is unclear. Better understanding the tectonic history alone is not sufficient. The key question seems to be: what is the structural (and potentially causal) relationship between the fault structure (folds and faults) and the overburdened landslide? You should also explain how this study contributes beyond the comprehensive work of Kirsten Hunter (2013).
• L60: Indicate that these investigations build upon and complement the comprehensive studies performed in Hunter’s thesis, which is detailed extensively. You may list all methods and datasets used in Section 3.
• L64: The "Aeanos Thrust" is a structural feature identified by Underhill (1989), but other names and traces exist in the literature (e.g., Kontogourata-Agon Fault). Refer also to Lekkas et al. (2016), Lagios et al. (2012), Stiros (1994), and Sorel (1976). Note that Sorel (1976), had already identified this structure as a reverse fault as many faults on Kefalonia Island.
• L66: “The area is approximately 773 km²”. This value is “accurate” and not approximative.
• L70 (Figure 1): While the text includes a topographic description, an altimetric map is clearly missing. Consider using a DEM or a topographic map from Hunter’s thesis (e.g., Figure 4.1).
• L74: In the Poros town area and eastern coast.
• L83: Figure 1b appears to be missing or is not easily located.
• L89: These faults are interpreted as ancient normal faults within the Preapulian-Paxos zone, which underwent tectonic inversion during the lower Pliocene and lower Pleistocene (Sorel, 1976, 1988).
• L97 (Figure 1): As previously noted, there is no hypsometric map. The red inset is very small and difficult to read.
• L100: Explain in the text why this specific zone was chosen. It is one of the locations studied by Hunter (paleolake of Katokhori) and the site of other investigations, such as drill holes, gravimetry, ERT, seismic refraction, and airborne magnetism. This study builds on and complements previous investigations.
• L114 (Figure 2): This figure does not provide information on the profile locations. There is no way to locate these profiles. Include a detailed, large-scale map with orientation, scale, contour lines, and coordinates, similar to Figure 2.9 in Hunter’s thesis.
• L167: Why not consider velocities in the range of 2400 to 3300 m/s, as previously mentioned?
• Figures 3 and 4: It is difficult to clearly identify R1 and R2.
• L187-188: These velocities differ again from those mentioned earlier. Please explain this discrepancy or modify.
• L280 (Figure 8): A comparative geological cross-section and a detailed, large-scale geological map are missing. The figure is too small. Consider using the geological map from Figure 3.29 in Hunter’s thesis and the cross-section from Figure 5.15, which includes two of the three boreholes (C5C, C4a). Indicate the location of the landslide referenced in Line 311 on both the cross-section and geological map.
• L318: Based on Hunter’s thesis conclusion, the hypothesis of a covered channel is ruled out by her data analysis. The only remaining possibility seems to be that a large landslide, encompassing the entire slope, could have carried and possibly destroyed the supposed channel. Refer to Section 462 of Hunter’s thesis for further details.Technical corrections
References:
• L 39: (Gaki-papanastassiou et al., 2010 or 2011 ?) see ref listCitation: https://doi.org/10.5194/egusphere-2024-3783-RC1 -
AC1: 'Reply on RC1', Samuel Zappalá, 18 Feb 2025
Dear reviewer 1,
Thank you for your valuable comments that helped us improving the quality of this paper. Most of the suggested comments have been implemented and when not possible, a detailed explanation is given.
Figures have been improved as suggested to help the readers to better locate the investigations conducted and to reach a broader audience.
Replies to individual comments:
Title R: Thanks for the suggestions, we changed the title to be less ambiguous, more concise and more focused on the geophysical aspect as follow “Reflection seismic imaging across the Thinia valley (Greece)”
L24R: Unfortunately, we do not have any information about the age. As you say, the dating in literature of the possible landslides is highly questionable and reflection seismology do not add any information regarding ages.L39R: We refer to (Gaki-papanastassiou et al., 2010). In this phrase, a list of references of past geological and geophysical works done in the island highlight their variety. This reference is reported as an example of geomorphological study in Kefalonia island.
L40R: Thanks for the suggested references. We added Stiros et al. (1994) reference. We could not find the suggested Sorel studies but we can see that they are included in later works that are already referenced in our paper, such as Hunter (2013), Karymbalis et al. (2013) and Gaki-Papanastassiou et al. (2010 and 2011)
L41R: The text has been updated as follow to add more details: “Analyses of the island’s displacements from the observed seismicity have shown complex horizontal motion that suggests two separate crustal blocks in the island, the main island with a displacement towards south-west and Paliki peninsula with a displacement towards north (Ganas et al., 2015). In addition, an unclear history of rotational movements characterized the island with probably both clockwise and anticlockwise rotational phases (Duermeijer et al., 2000; Sbaa et al., 2017).”.
L55-57R: Thanks for the suggestion, we modified the text to clarify how this study contributes beyond previous studies in the same area, including Hunter (2013), and we better defined the key question as you suggested. We included a text comparison also in the discussion section to highlight what the presented results add to Hunter work and in what they agree or differ.
L60R: The text has been updated and the relation between this study and previous studies is better indicated. Previous geophysical methods applied along the Thinia valley are now summarized in the introduction.
L64R: Thanks for the comment, other names for this structure found in literature are added to the paper to increase the audience understanding.
L66R: You are right. Thanks for pointing it out. Text updated accordingly.
L70R (Figure 1): Figure 2 was reporting the topography of the area but we agree that was not complete. Now it has been updated with a map view of the same area complete of all needed information.L74R: Thanks for the clarification. Text updated.
L83R: You are right. The figure was modified but the text was not updated. Now the text is updated according to the figures. Thanks for spotting it.
L89R: Thank for the clarification. The text has been updated to include this information as follow: “shows eastwards ancient dipping normal faults on the west of the valley which underwent tectonic inversion”.
L97R (Figure 1): elevation information added in the updated figure 2. The profiles in the red inset have been replaced with an asterisk for the whole survey while their precise locations are now reported in figure 2 with a larger scale.
L100R: Text has been updated in this chapter and in the introduction to include this work relation with previous studies.
L114R (Figure 2): Figure updated as suggested from the reviewer to include all missing information and be of easier readability for the audience.
L167R: Different velocities and parameters are always tested to ensure the best results from each processing step and in this case, this was the model resulting in the best results. In general, NMO velocities are usually different from real velocities and valid only for the NMO correction. Therefore, even if they can be used as starting reference for other velocities models (e.g. the one used for time to depth conversion) they will most likely be different. Text has been updated to include more details on the time to depth model selection.
Figures 3 and 4R: Extra arrows are added in figures 3 and 4 to help better identify the reflections.
L187-188R: See reply L167R.
L280R (Figure 8): Thanks for the comment, the geological map shown in Figure 8b has similar scale as the one suggested from Hunter (2013) and shows mostly the same features, but considers also other works as the IGME 1985 geological map. The small local features shown in Hunter’s map and not reported in Figure 8b are below our acquisition resolution. We agree that the map can be shown larger so we added a zoom in inset in Figure 8b showing the interpreted area to better visualize the details.This paper is not meant to be a review of previous works, we would not like to do a direct comparison with previous geological cross-sections since many studies (i.e., Underhill, 1989; IGME, 1985, 1996; Gaki-Papanastassiou et al. 2011) were done in the area often with different results and all of them reliable only for shallower depth than this paper. In the discussion section a paragraph is added to compare the obtained results with Hunter (2013) results and highlighting corresponding features of the geology in the area (thrusts and stratigraphy) but also the limit of comparing these different works.
We are not able to define the exact location of the probable landslide material since landslides are very localized and below our resolution and seismic velocities are an average of all materials they cross. We suggest that landslide material is probably present in the low velocity chaotic near surface layer (transparent grey in Figure 8a) since its average seismic characteristics (velocity and reflectivity) are similar to the ones typical of landslides material and different from the ones expected from only quaternary sediments. Text has been updated in the interpretation and conclusion sections to make it clearer.
L318R: Thanks for the clarification, the text has been updated to remove the connection with a covered channel as follow: “and overburden landslide materials are detected.”.
-
AC1: 'Reply on RC1', Samuel Zappalá, 18 Feb 2025
-
RC2: 'Comment on egusphere-2024-3783', Anonymous Referee #2, 20 Jan 2025
Dear Authors,
Thank you for submitting your manuscript to Solid Earth. The manuscript presents a study that utilizes seismic reflection and tomography methods to investigate the subsurface structure of Kefalonia Island.
I have evaluated this manuscript for its geophysical content. My expertise does not extend to the geological and tectonic interpretations specific to Kefalonia Island; therefore, this review excludes a detailed critique of the geological findings and tectonic implications.
While the study offers valuable insights into seismic near-surface characterization, several aspects need refinement to improve clarity and accuracy.
- The title needs rephrasing for clarity and precision.
- Lines 33, 50, 55, 58, 68, 83: You referred to Figures 1a and 1b. However, Figure 1 does not include subfigures (1a, 1b). Update Figure 1 accordingly.
- Lines 62–64: You provided your findings at the end of the introduction. The introduction should set up the study, not summarize its findings. Instead, clearly define the manuscript content. The authors should ensure that the introduction flows smoothly into the rest of the manuscript.
- Line 71: Provide full wording for the abbreviation “a.s.l.” on its first use.
- Lines 159–160: Figures 4c and 4d are incorrectly referenced; they should be moved to the correct location in this sentence to avoid reader confusion.
- Figure 6: I recommend including a 3D visualization of velocity profiles in an integrated model to validate the velocity at intersection points.
- The coherence and continuity of reflections across profiles are insufficiently validated. Including a 3D visualization of intersecting profiles (after migration) would enhance the understanding of spatial relationships. Additionally, any discrepancies between reflections at the intersections, even if minor, should be critically analyzed and discussed.
- Lines 165–166: Clarify how the migration velocity field was derived. Simply referencing only the velocity range estimated for the NMO corrections is insufficient.
- In Table 2, specify where refraction statics were applied on Profiles 1, 2, and 3.
- Profile 1 shows promising results, yet the absence of migration for other profiles undermines the reliability of structural interpretations. Why didn’t you apply migration to Profiles 2, 3, and CS? Your current reasoning is not satisfactory. These profiles exhibit dipping reflections, and post-stack migration is essential for accurate positioning and interpretation. Superimposing a near-surface velocity-depth model on stack sections is inappropriate. It is expected that these profiles be migrated, converted from the time domain to the depth domain, and tied to the borehole data (Figure 7) to ensure a more accurate and robust interpretation.
- Line 221: “the best quality” is not appropriate and scientific. Rephrase it and make this sentence comparative. For example, replace it with “higher resolution compared to other profiles.”
- Line 233: This sentence is vague and requires rephrasing. What characteristics do you mean?
- Line 257: You plotted the boreholes with respect to elevation but referred to them in terms of depth (“where the available boreholes reach a depth of 90 to 100 m”), which is inconsistent and potentially confusing. Please clarify and ensure consistency.
- Line 305: “the best result” is not appropriate and scientific. Replace it with a comparative phrase. See 11.
- Lines 307–309: In the conclusion, you stated that the 3D traveltime modeling of reflection R1, both in the shot gather and the unmigrated stack section, suggests a N-23° strike and a 44° dip angle towards the east, consistent with the local geology. However, you did not present the modeling results for the unmigrated stack section in the manuscript. Conclusions must be based on the results shown; therefore, this statement needs to be revised or supported with the missing data.
- You claimed that R1 is observed with an estimated orientation of N23°E/44°E. However, Profile 3 in Figure 7d does not show a north-dipping reflection at its intersection with Profile 1. As mentioned before, adding a 3D visualization of intersecting profiles (after migration) would provide a clearer understanding of this relationship.
Recommendation:
Overall, this manuscript has the potential to make a valuable contribution to seismic near-surface characterization. However, it requires major revisions to address the issues outlined above. I recommend resubmission after revisions.
Citation: https://doi.org/10.5194/egusphere-2024-3783-RC2 -
AC2: 'Reply on RC2', Samuel Zappalá, 18 Feb 2025
Dear reviewer 2,
Thank you for your valuable comments that helped us improving the quality of this paper. Most of the suggested comments have been implemented and when not possible, a detailed explanation is given.
Following you can find detailed replies to your comments.
- R: The title has been updated to be less ambiguous, more concise and more focused on the geophysical aspect as follow “Reflection seismic imaging across the Thinia valley (Greece)”
- Lines 33, 50, 55, 58, 68, 83R: You are right. The figure was modified but the text was not updated. Now the text is updated according to the figures. Thanks for spotting it.
- Lines 62–64R: Thanks for the comment. We removed the phrase summarizing the findings in the introduction and we better defined the manuscript content.
- Line 71R: Text updated.
- Lines 159–160R: Thanks for the comment. Text has been updated to properly match the figures.
- Figure 6R: The velocity profiles intersect only on a single point because the velocity model from profile 3 has no values at the intersection with profile 1 (see attached file). Since no 3D analysis or interpretation are carried out in this study and since a 2D display of the profiles enable to visualize more details respect than a 3D visualization, we think that a 3D visualization of the velocity models will not add any useful information to the paper while hiding important features. Still, if the reviewer insists we can add a figure with the 3D visualization of the velocity models.
- R: The coherence of reflections across profiles is not analyzed in the paper since migration of all profiles was not possible (see reply to comment 10) and as you suggest it is meaningless to compare unmigrated profiles. We do not want to plot a 3D visualization of the seismic profiles to do not give the wrong impression that the interpretation is done on all profiles. Indeed, the interpretation is done only on profile 1 and only a very cautious suggestion of eventual corresponding features is done on the other profiles that is expected to be a base for future studies and nothing more.
- Lines 165–166R: Text has been updated to give more details as follow: “first using the estimated NMO velocities, and then refining them with the evaluation of smiles and frowns in the resulting migrated sections”.
- In Table 2R: We are not sure we understand the question; refraction statics are applied at step 11 of our processing flow as expressed in table 2. If something else is not clear please let us know.
- R: We completely agree with the benefits of migrating all profiles and we largely tried before submitting the paper. Unfortunately, the low quality of the data in profile 2, 3 and CS and the absence of an accurate deep velocity model resulted on the migration of noise instead than of the reflections when migration is applied to the data. The quality of the acquired data is this for the reasons explained in the data acquisition chapter and we cannot change it. Reason why in our interpretation we only considered the results from profile 1 and we clearly stated that the quality from the other profiles is not enough for any interpretation, limiting ourselves to show the obtained results. Data from boreholes are shallow and discontinue and do not clearly intersect any main seismic reflector and therefore they cannot be used to tie the seismic sections. Text has been updated to make it clearer.
- Line 221R: Thanks for the suggestion, the text has been updated as follow: “The unmigrated stacked section of profile 1 (Figure 7b) shows a higher S/N ratio and reflections continuity with respect to the other profiles.”.
- Line 233R: The text has been rephrased for better clarity as follow: “The deepest of these reflections shows amplitude values, frequency content and shape of the signal similar to the ones from R2 reflection of profile 1 (Figure 7b) and considering its location (between 0 and -150 m a.s.l.), it could correspond to the same horizon. The reflection located at a depth of 150 to 50 m a.s.l. matches amplitude values, frequency content and shape of the signal from reflection R1 of profile 1 suggesting their correspondence.”.
- Line 257R: The text has been updated to improve its consistency as follow: “Comparison with the boreholes lithologies (Figure 6) is more meaningful for profile 2 (Figure 6b) where the available boreholes are 90 to 100 m deep and reach an elevation of approximately 60 and 70 m a.s.l. (Figure 6b).” and “. Borehole C5a along profile 1 (Figure 6a) is only 30 m deep and reach an elevation of approximately 150 m a.s.l. (Figure 6d) showing low velocities corresponding to the logged marl.”.
- Line 305R: Thanks for the suggestion, the text has been updated for a more scientific language as follow: “profile 1 showed higher S/N ratio and reflection continuity respect to the other acquired profiles”
- Lines 307–309R: You are right and thanks for noting it. A phrase regarding the modeling in stack domain is added to the text (section 4.3) and the interpretation text is updated accordingly. The resulting modeled horizon is shown in figure 8.
- R: As we said in the text, the north part of profile 3 is not reliable because of the source issue (visible also from the missing traveltime tomography) and therefore a comparison with profile 1 will be meaningless, without considering the migration issue detailed in comment 10. Furthermore, the N23°E/44°E means a slight dipping towards south and not towards north, with the strike almost parallel to profile 3 that should result on an almost horizontal apparent dip.
-
EC1: 'Comment on egusphere-2024-3783', David Snyder, 22 Jan 2025
Reviews by two referees indicate that the manuscript could be acceptable for this special volume with some significant revision. The seismologist requests clarification and more details of the processing using the entire data set available to the authors. The interpreter suggests less speculation about archeological significance but greater discussion about geomorphology/landslides. Both request more minor clarifications in numerous locations. All of these comments seem reasonable to me and would greatly improve the manuscript and make it a more substantial contribution.
Citation: https://doi.org/10.5194/egusphere-2024-3783-EC1 -
AC3: 'Reply on EC1', Samuel Zappalá, 18 Feb 2025
Dear editor,
thanks for the opportunity to improve our manuscript implementing the comments and suggestions provided by the reviewers. We really appreciate the time and effort that the reviewers placed on this manuscript and we addressed their comments in the revised paper. Detailed responses to each comment are provided in the discussion section.
Citation: https://doi.org/10.5194/egusphere-2024-3783-AC3
-
AC3: 'Reply on EC1', Samuel Zappalá, 18 Feb 2025
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 250 | 68 | 17 | 335 | 11 | 14 |
- HTML: 250
- PDF: 68
- XML: 17
- Total: 335
- BibTeX: 11
- EndNote: 14
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Alireza Malehmir
Haralambos Kranis
George Apostolopoulos
Myrto Papadopoulou
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(4149 KB) - Metadata XML