Dear reviewer,

Thank you for your valuable suggestions regarding the manuscript. I have implemented your recommended revisions, which are explained below and highlighted in the attached document.

Page 1, lines 14-15: sentence and number revised.

Page 2, lines 38-63: revised and citation added.

Page 4, lines 98-99-101-108: revised.

Page 5 Lines 115-116-117-119-129-131-132: table and numbers revised.

Page 6, lines 134-135-137-138: sentences revised.

Page 6, line 161: I used MAXQDA qualitative data analysis software to manage the data in this study. I honestly didn't know how to use natural language processing tools in my analysis, but I'm eager to learn and incorporate them into my future academic work.

Page 7, lines 174-175: sentence revised and explanation added.

Page 8: Figure 2 revised regarding time (am/pm) inconsistencies.

Page 9, lines 209-221: sentence and number revised.

Page 11, lines 251-253-256: sentence and citations revised.

Page 12, lines 263-265-267-268-271-175-280: suggested revisions implemented.

Page 13, lines 190-192-294-309-311-313-321: suggested revisions implemented.

Page 14, lines 323-325-341-347-349: suggested revisions implemented.

Page 15, line 361: revised regarding time (am/pm) inconsistencies.

Page 15, line 378: I downloaded the Copernicus-style files from the address and managed references with Mendeley automatically. However, I tried to revise some of the references manually. https://publications.copernicus.org/for_authors/manuscript_preparation.html

Page 16, line 389: revised.

Page 18, line 464: revised.

Page 20, line 518: revised.

Thank you so much for your suggestions on improving my academic approach. If you have any further recommendations, I will revise them.

Regards.

Dear reviewer, I have revised Figure 2 using natural language processing tools. I will replace it before submitting the latest version.

Dear reviewer,

Thank you for your valuable suggestions regarding the manuscript. I have implemented your recommended revisions, which are explained below and highlighted in the attached document.

General comments

1. Better present the raw data used in the study, with more precise details on how they were obtained, together with an in-depth descriptive analysis. In addition, justify the decision to consider only the official AFAD communication, which gives an oriented view of crisis management by the authorities.

Response: The document has been revised to emphasize how the data were obtained and the rationale for selecting AFAD as a sample for this study. It also includes information about the response process, which is made publicly available solely by AFAD. This is because all other response institutions convey their information to AFAD, which issues public statements during disasters.

1. Formulation of scientific questions (section 3.2): Either replace the term “Turkish government” with “AFAD”, or consider in the study other data sources that allow other government sources to be taken into account as a complement to AFAD.

Response: It was revised to AFAD, as you suggested.

1. The discussion chapter (section 5) draws only slightly on the results of the study (section 4). Thus, section 5 is essentially a discussion of generalities already well known in crisis management, with no critical analysis of AFAD's communication elements. On the contrary, some of the "results" highlighted are of little interest. For example, it is stated that AFAD issued press releases on average every 3 hours, and tweets every 50 minutes (p. 12, l.261-262). After this very brief mention of the results of the data analysis conducted in section 4, there follows a twenty-line discussion that looks more like a literature review, and in which there is not once a reference to the specific case of the 2023 earthquakes...

Response: The paragraph has been revised; some sentences were removed, and a reference to the earthquake has been added.

The paragraph on the contribution of social networks to crisis management (p14-15, l339-358)  is another example of the poor construction of section 5, since the analysis of tweets posted by AFAD does not really support this paragraph, particularly as regards the monitoring of social networks by the authorities for situational awareness purposes. This is highly problematic, and highlights both the lack of significant results and the lack of exploitation of these results. This section 5 needs to be rewritten in its entirety, after strengthening section 4, and attempting to draw specific lessons from the 2023 earthquakes (i.e. not just illustrating results from other previous studies).

Response: The paragraph has been revised; the “situational awareness” part of the sentence has been removed. Section 5 has been revised.

Conclusion (section 6): In its current wording, section 6 does not at all answer the research questions formulated in section 3.2. It is essential that this be the case. I think it would be important, for example, to cite the study by Platt and Drinkwatter  (https://doi.org/10.1016/j.ijdrr.2016.03.010), which in 2016 highlighted AFAD's inability to  anticipate the long term during crisis management, and to answer the question of whether  this criticism has been confirmed or invalidated in 2023?

Response: Section 6 has been reorganized, including the research questions. Additionally, more resources have been cited in the document.

1. References:

o Standardize the way references are cited.

Response: It was revised in the document.

o When the authors of a study are used as the subject of a sentence, add the year in parentheses and delete the duplicate reference at the end of the sentence (e.g.  sentence in p.12, l.265-267: "Jones et al. (2017) argued ..."

Response: It was revised in the document.

Specific comments

- P.1, l.9: What does "purposive sampling" mean? Do not introduce this technical term in the

abstract, or explain it briefly. This term should also be made clearer in section 3.

Response: The information was detailed and highlighted in the abstract and Section 3.

- P.3, l.85: choose a more convincing reference to illustrate the point than the one from

Mendoza et al.

Response: It was removed from the document.

- P.3, l86-89: The sentence about the value and flexibility of USAR teams is strange, because it reduces the operational management of rescue operations to the sole issue of "search and rescue".

Response: The sentence was revised and highlighted in the document.

- P.4, l.94: The title of section 2 is catchy, but not suitable for a scientific article. Change it.

Response:  The title was revised and highlighted

- P.4, l.95-96: Change the reference to a scientific source rather than that of the Turkish government.

Response: It was changed.

- P.4, l.101: missing reference to EM-DAT in bibliography

Response: It was revised.

- P.4, l.104: In the legend to figure 1, replace the word "earthquake" with "seismic"

Response: It was replaced.

- P.4, l.105: replace the term "fault lines" with "fault systems

Response: It was replaced.

- P.4, l.108: the term "tremor" is very specific to certain types of seismic event. Replace by "seismic sequence »

Response: It was replaced.

- P.4, l.110-111: The last sentence isn't very clear and needs rewording, as it's not clear which earthquake event was ultimately the most destructive.

Response: The sentence was revised.

- P.5, l.116-117: indicate that the underwater landslide caused a small tsunami

Response: It was indicated as suggested.

- P.5, l.123: “dismantling structures that pose a hazard” -> incomprehensible wording to be reformulated

Response: It was reformulated.

- P.6, l.138-140: “Plain reinforcement […] ground movement” -> incomprehensible wording to be reformulated

Response: It was reformulated.

- P.6, l.152 : ref. to Patton, 2015 is lacking in the bibliography

Response: It was revised.

- P.7, l.177-183: The table must be clearly referenced with a number and a legend.

Response: It was revised.

- P.8, figure 2: This figure takes up a lot of space but doesn't import much information. A time histogram would probably be more informative.

Response: I created this figure using MAXQDA software, as it allowed me to upload all relevant documents and transcriptions. Due to the irregular time intervals, the time histogram requires two separate charts, making it challenging to explain the entire process within a single histogram. I sincerely apologize for any confusion this may cause.

- P.8, l.202 to p.9, l.209: I don't understand why this discussion about publication times and frequency is interesting??? It's a description that doesn't do much for the reader.

Response: To provide additional clarification, the following paragraph has been added to the results section. “The first ten instances of information sharing, including social media and press releases, were analyzed based on frequency and timing. After these ten statements, information sharing continued regularly. The first 72 hours after a disaster are critical for individual survival and preventing secondary victimization (Codreanu et al., 2017; Sakurai et al., 2014). Furthermore, following crises and disasters, timely and rapid dissemination of information is crucial for effective crisis communication (Chen et al., 2021; Gurman and Ellenberger, 2015; Murthy et al., 2019). Responding to a disaster promptly and sharing timely information is essential for saving lives and ensuring effective crisis communication. The inclusion of response time in this study is based on this reason.”

- P.9, figure 3: remove the 1st mention of the source, in bold text

Response: It was removed.

- P.10, figure 4: It's unfortunate that the way in which the "tag" categories shown in Figure 4 are not presented and discussed in detail in the article, as this is the author's real work.

Response: The figure 4 has been reorganized, and recovery codes have been removed because they were the least frequently occurring codes. Additionally, the codes that occurred most frequently, such as search and rescue, Coast Guard and Navy, tent, and social media, have been discussed in the discussion section.

What's more, this classification is highly debatable, and therefore needs to be justified: it's strange, for example, to see the "shelter & meal" category appear at the same level as the "response" category?

Response: The following paragraph has been included for Figure 4 to present further justification. “Following the disaster, all statements prominently featured codes related to coordination, response, shelter, and recovery. As a result, these codes were evaluated collectively without any modifications to the documents. This approach explains why shelter and meals, along with response efforts, are displayed together in the exact figure. Consequently, these categories were also integrated into the crisis management assessment.”

- P.11, figure 5: A word cloud doesn't add much to a scientific article, because it's only illustrative. A more in-depth analysis is needed (e.g. different timelines counting the number of mentions of each code over time).

Response: The explanation of the figure has been revised as follows. “The data presented in Figure 5 clearly illustrates the predominant total codes referenced in the earthquake's aftermath. Covering a 25-day period from February 6, 2023, when the quake occurred, to March 3, 2023, a thorough analysis of 37 press releases and 1,347 social media statements was conducted using a rigorous qualitative analysis program. The significance of this figure lies in its ability to provide a comprehensive overview of the operational processes during the 25 days following the disaster without any external interference. The subsequent analysis revealed the most frequently used codes in the official statements during this time, which included search and rescue, coast guard, military, shelter, tent, evacuation, mobile kitchen, and the number of personnel involved in the rescue efforts.”

Also, code clouds envision the most frequently used codes as word clouds. They are handy for exploring and illustrating the use of codes within a process. In this article, the code cloud displays the intensified codes that explain the operational procedures following the major earthquake.

- P.12, l285-289: The discussion on the capacity of INSARAG teams needs to be revisited. In particular, the estimated need for 1.5 million USAR personnel means nothing at all, since each team investigates many buildings during its mission (HUSAR teams can even manage two buildings at the same time).

Response: The sentence has been revised.

- P.13, l.290-306: This discussion doesn't add up to much, and the reader doesn't really understand what it's for. …

Response: The sentences have been revised and some of the statements have been removed from the text.

- P.14, l.323-331: Interesting discussion, but absolutely not supported by the data ... which raises a serious problem.

Response: The sentences have been revised and some of the statements have been removed from the text.

- P.15, section « Conclusion »: The conclusion does not do enough to highlight the author's work, and is far too complacent towards the Turkish authorities. For example, line 369 states that the commitment of the volunteers was remarkable: this is true, but in an article dealing with the coordination of crisis management, the difficulty of coordinating such a surge of solidarity should be emphasized.

Response: The conclusion section has been revised.

Thank you again.

Dear reviewer,

Thank you for your valuable suggestions regarding the manuscript. I have implemented your recommended revisions, which are explained below and highlighted in the attached document.

General comments

1. Better present the raw data used in the study, with more precise details on how they were obtained, together with an in-depth descriptive analysis. In addition, justify the decision to consider only the official AFAD communication, which gives an oriented view of crisis management by the authorities.

Response: The document has been revised to emphasize how the data were obtained and the rationale for selecting AFAD as a sample for this study. It also includes information about the response process, which is made publicly available solely by AFAD. This is because all other response institutions convey their information to AFAD, which issues public statements during disasters.

1. Formulation of scientific questions (section 3.2): Either replace the term “Turkish government” with “AFAD”, or consider in the study other data sources that allow other government sources to be taken into account as a complement to AFAD.

Response: It was revised to AFAD, as you suggested.

1. The discussion chapter (section 5) draws only slightly on the results of the study (section 4). Thus, section 5 is essentially a discussion of generalities already well known in crisis management, with no critical analysis of AFAD's communication elements. On the contrary, some of the "results" highlighted are of little interest. For example, it is stated that AFAD issued press releases on average every 3 hours, and tweets every 50 minutes (p. 12, l.261-262). After this very brief mention of the results of the data analysis conducted in section 4, there follows a twenty-line discussion that looks more like a literature review, and in which there is not once a reference to the specific case of the 2023 earthquakes...

Response: The paragraph has been revised; some sentences were removed, and a reference to the earthquake has been added.

The paragraph on the contribution of social networks to crisis management (p14-15, l339-358)  is another example of the poor construction of section 5, since the analysis of tweets posted by AFAD does not really support this paragraph, particularly as regards the monitoring of social networks by the authorities for situational awareness purposes. This is highly problematic, and highlights both the lack of significant results and the lack of exploitation of these results. This section 5 needs to be rewritten in its entirety, after strengthening section 4, and attempting to draw specific lessons from the 2023 earthquakes (i.e. not just illustrating results from other previous studies).

Response: The paragraph has been revised; the “situational awareness” part of the sentence has been removed. Section 5 has been revised.

Conclusion (section 6): In its current wording, section 6 does not at all answer the research questions formulated in section 3.2. It is essential that this be the case. I think it would be important, for example, to cite the study by Platt and Drinkwatter  (https://doi.org/10.1016/j.ijdrr.2016.03.010), which in 2016 highlighted AFAD's inability to  anticipate the long term during crisis management, and to answer the question of whether  this criticism has been confirmed or invalidated in 2023?

Response: Section 6 has been reorganized, including the research questions. Additionally, more resources have been cited in the document.

1. References:

o Standardize the way references are cited.

Response: It was revised in the document.

o When the authors of a study are used as the subject of a sentence, add the year in parentheses and delete the duplicate reference at the end of the sentence (e.g.  sentence in p.12, l.265-267: "Jones et al. (2017) argued ..."

Response: It was revised in the document.

Specific comments

- P.1, l.9: What does "purposive sampling" mean? Do not introduce this technical term in the

abstract, or explain it briefly. This term should also be made clearer in section 3.

Response: The information was detailed and highlighted in the abstract and Section 3.

- P.3, l.85: choose a more convincing reference to illustrate the point than the one from

Mendoza et al.

Response: It was removed from the document.

- P.3, l86-89: The sentence about the value and flexibility of USAR teams is strange, because it reduces the operational management of rescue operations to the sole issue of "search and rescue".

Response: The sentence was revised and highlighted in the document.

- P.4, l.94: The title of section 2 is catchy, but not suitable for a scientific article. Change it.

Response:  The title was revised and highlighted

- P.4, l.95-96: Change the reference to a scientific source rather than that of the Turkish government.

Response: It was changed.

- P.4, l.101: missing reference to EM-DAT in bibliography

Response: It was revised.

- P.4, l.104: In the legend to figure 1, replace the word "earthquake" with "seismic"

Response: It was replaced.

- P.4, l.105: replace the term "fault lines" with "fault systems

Response: It was replaced.

- P.4, l.108: the term "tremor" is very specific to certain types of seismic event. Replace by "seismic sequence »

Response: It was replaced.

- P.4, l.110-111: The last sentence isn't very clear and needs rewording, as it's not clear which earthquake event was ultimately the most destructive.

Response: The sentence was revised.

- P.5, l.116-117: indicate that the underwater landslide caused a small tsunami

Response: It was indicated as suggested.

- P.5, l.123: “dismantling structures that pose a hazard” -> incomprehensible wording to be reformulated

Response: It was reformulated.

- P.6, l.138-140: “Plain reinforcement […] ground movement” -> incomprehensible wording to be reformulated

Response: It was reformulated.

- P.6, l.152 : ref. to Patton, 2015 is lacking in the bibliography

Response: It was revised.

- P.7, l.177-183: The table must be clearly referenced with a number and a legend.

Response: It was revised.

- P.8, figure 2: This figure takes up a lot of space but doesn't import much information. A time histogram would probably be more informative.

Response: I created this figure using MAXQDA software, as it allowed me to upload all relevant documents and transcriptions. Due to the irregular time intervals, the time histogram requires two separate charts, making it challenging to explain the entire process within a single histogram. I sincerely apologize for any confusion this may cause.

- P.8, l.202 to p.9, l.209: I don't understand why this discussion about publication times and frequency is interesting??? It's a description that doesn't do much for the reader.

Response: To provide additional clarification, the following paragraph has been added to the results section. “The first ten instances of information sharing, including social media and press releases, were analyzed based on frequency and timing. After these ten statements, information sharing continued regularly. The first 72 hours after a disaster are critical for individual survival and preventing secondary victimization (Codreanu et al., 2017; Sakurai et al., 2014). Furthermore, following crises and disasters, timely and rapid dissemination of information is crucial for effective crisis communication (Chen et al., 2021; Gurman and Ellenberger, 2015; Murthy et al., 2019). Responding to a disaster promptly and sharing timely information is essential for saving lives and ensuring effective crisis communication. The inclusion of response time in this study is based on this reason.”

- P.9, figure 3: remove the 1st mention of the source, in bold text

Response: It was removed.

- P.10, figure 4: It's unfortunate that the way in which the "tag" categories shown in Figure 4 are not presented and discussed in detail in the article, as this is the author's real work.

Response: The figure 4 has been reorganized, and recovery codes have been removed because they were the least frequently occurring codes. Additionally, the codes that occurred most frequently, such as search and rescue, Coast Guard and Navy, tent, and social media, have been discussed in the discussion section.

What's more, this classification is highly debatable, and therefore needs to be justified: it's strange, for example, to see the "shelter & meal" category appear at the same level as the "response" category?

Response: The following paragraph has been included for Figure 4 to present further justification. “Following the disaster, all statements prominently featured codes related to coordination, response, shelter, and recovery. As a result, these codes were evaluated collectively without any modifications to the documents. This approach explains why shelter and meals, along with response efforts, are displayed together in the exact figure. Consequently, these categories were also integrated into the crisis management assessment.”

- P.11, figure 5: A word cloud doesn't add much to a scientific article, because it's only illustrative. A more in-depth analysis is needed (e.g. different timelines counting the number of mentions of each code over time).

Response: The explanation of the figure has been revised as follows. “The data presented in Figure 5 clearly illustrates the predominant total codes referenced in the earthquake's aftermath. Covering a 25-day period from February 6, 2023, when the quake occurred, to March 3, 2023, a thorough analysis of 37 press releases and 1,347 social media statements was conducted using a rigorous qualitative analysis program. The significance of this figure lies in its ability to provide a comprehensive overview of the operational processes during the 25 days following the disaster without any external interference. The subsequent analysis revealed the most frequently used codes in the official statements during this time, which included search and rescue, coast guard, military, shelter, tent, evacuation, mobile kitchen, and the number of personnel involved in the rescue efforts.”

Also, code clouds envision the most frequently used codes as word clouds. They are handy for exploring and illustrating the use of codes within a process. In this article, the code cloud displays the intensified codes that explain the operational procedures following the major earthquake.

- P.12, l285-289: The discussion on the capacity of INSARAG teams needs to be revisited. In particular, the estimated need for 1.5 million USAR personnel means nothing at all, since each team investigates many buildings during its mission (HUSAR teams can even manage two buildings at the same time).

Response: The sentence has been revised.

- P.13, l.290-306: This discussion doesn't add up to much, and the reader doesn't really understand what it's for. …

Response: The sentences have been revised and some of the statements have been removed from the text.

- P.14, l.323-331: Interesting discussion, but absolutely not supported by the data ... which raises a serious problem.

Response: The sentences have been revised and some of the statements have been removed from the text.

- P.15, section « Conclusion »: The conclusion does not do enough to highlight the author's work, and is far too complacent towards the Turkish authorities. For example, line 369 states that the commitment of the volunteers was remarkable: this is true, but in an article dealing with the coordination of crisis management, the difficulty of coordinating such a surge of solidarity should be emphasized.

Response: The conclusion section has been revised.

Thank you again.