This study investigates the clustering properties of seismicity in southern and southeastern Spain from 1970 to 2023 using the Nearest-Neighbor (NN) algorithm. The study applies graph theory metrics (outdegree centralization, closeness centralization, and average leaf depth) to characterize earthquake clusters and proposes three tectonic zonation models. Results indicate that the western region exhibits complex swarm-like clusters, while the eastern region shows simpler burst-like sequences. Two of the proposed zonation models pass the Kolmogorov-Smirnov (KS) test, demonstrating statistically significant differences in clustering behavior between zones.

This study reveals spatial heterogeneity in seismicity, linking clustering behavior to regional tectonic settings. The proposed zonation models can support earthquake risk mitigation and engineering design in a seismically active region.

I recommend minor revisions prior to acceptance, focusing on:

1. The point"few earthquakes are associated with specific fault segments and occurrence time periods (when indicated) are affected by high uncertainties " in the abstract is not quite clearexplained in introduction. It is suggested to supplement relevant content.
2. In section 3.1, Mc and b value is presented using Table 1 and Figure 5. These two contain a lot of duplicate information, it is recommended to merge. And the graphs and tables in the article are many, it is suggested to adjust appropriately and retain the more critical ones.
3.  From Table 2, it is observed that the two datasets exhibit a difference of 0.2 in the Mc value; however, there is a significant disparity in the number of recorded earthquakes. Are there any additional differences between the two earthquake catalogs that could account for this variation?
4. In Section 4.1, the metrics of outdegree centralization, closeness centralization, and average leaf depth are utilized to assess the characteristics of clusters. Additionally, various types of seismicity are mentioned, including burst-like, swarm-like, chain-like, and umbrella-like phenomena. It is recommended to unify these terms, as some possess similar meanings, to enhance clarity and coherence. Furthermore, a more detailed explanation of the specific characteristics of each cluster type would provide a deeper understanding of the underlying seismic activities.
5. Figure 3 presents the magnitude-depth distribution of seismic events. It would be interesting to further investigate whether the differences in seismic activity characteristics between tectonic regions can be better understood through a combined analysis of focal depths and focal mechanisms. This could be a direction for future research.

Review of the manuscript Insights into tectonic zonation models from the clustering analysis of seismicity in South and South-Eastern Spain,  submitted to Solid Earth by Montiel-López et al.

In this manuscript, the authors analyze the clustering of the seismicity catalog during > 50 years (1970-2023) in the south and south-Eastern portions of Spain. Based on the properties of the clusters of seismicity that they identified, using a nearest-neighbor approach in the space-time and magnitude domain, they propose an updated model of seismic zonations for this region. The paper is well written and the results hold the potential to be of interest to understand earthquake dynamics in this interesting and populated region. However, there are a few aspects that are not well discussed during the paper and need further explanations, and revisions to strengthen the results and the interpretations inferred from them. I recommend this manuscript for publication upon revision of the aspects below.

1# On the declustering of the seismicity catalog: The authors discuss in the introduction the importance of declustering the catalog to study seismic hazard analysis. However, to my understanding, in the following, the paper mostly focuses on the analysis of the seismicity clusters, rather than on the background (declustered) catalog. Thus, it is not clear what is the main purpose of the manuscript (e.g. the declustered seismicity, or the analysis of clusters). Depending on which one, the nearest-neighbor strategy to be applied is different (i.e. b=0, see Zaliapin and Ben-Zion, 2021 for this topic). See for example the name of the section 3.2 “Role of Nu in the declustering”. I believe “Role of Nu in the identification of clusters” would better represent the goals of the paper.

2# On the homogeneity of magnitude types: Can the authors confirm that all employed magnitudes correspond to Moment Magnitudes M_W? Typically, regional catalogs employ rather local magnitudes, and scaling conversions need to be applied. This could affect these results and would be important to confirm it.

3# On the two datasets from Table 2: I did not understand what is the purpose of this separation of datasets, I believe that there is some mistake in the table, as the number of events included is dramatically different, but the parameters specified in Table are almost the same. Even if included in the Supp. Materials, a minimum of 1-2 lines are needed to justify the employment of these datasets and what they represent.

4# On the analysis of clusters: It is not specified whether for estimating the rescaled distances, authors are using hypocentral or epicentral distance. This is important to understand which fractal dimension should be used.

5# Metrics used: Returning to the issue of analysis of clusters vs declustering: the metrics defined in section 4.1 solely analyze the clustered part of the seismicity catalog. I strongly suggest to include some other metrics that also analyze the background seismicity of the region. Some suggestions include the ratio of clustered to background seismicity, or the proportion of single seismicity on these regions (see e.g. Martínez-Garzón et al., 2018; 2019 for some example of these metrics considering also the background seismicity).

6# On the distinction between burst and swarms: In fact, I don’t see this distinction very clear, other than in the Adra vs Granada sequences. To check the potential bimodality of the clusters, I recommend to plot average leaf depth as a function of cluster size, similar to what done in Figure 2d from Martínez-Garzón et al., 2019. This should give us a better perspective if indeed it is related to the larger magnitude size, or if there might be physical reasons promoting bursts (here called umbrellas) or swarms.

7# Finally, although this paper does not focus on understanding the physical processes, it would be good to note somewhere in the paper than the behaviour of bursts vs swarms has been mainly related to the heat flow as well as the content of fluids in the crust. I think it may be worth to connect and refer to papers characterizing these processes in the here analyzed region.

I hope the authors find these comments useful.

Regards,

Patricia Martínez-Garzón

Other comments:

Lines 35-and after: Please indicate these notable earthquakes in Fig. 1 or in Fig. 2.

Line 83-85: We need to know a bit more details on this work about the splitting of the catalog on these four periods, without having to consult Gonzalez (2017) for a minimum info. “It can be seen clearly from Figure 4”, in fact I cannot see this so clearly.

Line 102: I think the correct references should be Zaliapin and Ben-Zion (2013a; 2013b), where the bimodality of the distributions is presented for first time.

Line 107: Labels (-1,1 and 2) are not needed in the paper.

Line 109: I am not 100% sure but I think the correct term for “founder” in the Z&BZ nomenclature is parent.

Section 4.1: The term “umbrella-cluster” is repeatedly used throughout the text, but it is not a well defined concept. I suggest to remove it from the text, as I feel that it is the same than the burst-like cluster topology defined by Z&BZ (2013b).

Figure 1: please add if possible stations to this figure, color encoded with the time of start operation.

Figure 2: As M < 2 is below Mc during the entire time period, I suggest to remove them to try to clean a bit the figure.

Please also include the plot of the distribution of rescaled times and distances, preferentially as part of Figure 7.

Figure 9: Please add the name of main fault structures and cities to these maps.

References

Leptokaropoulos, K., Staszek, M., Lasocki, S., Martínez-Garzón, P., & Kwiatek, G. (2018). Evolution of seismicity in relation to fluid injection in the North-Western part of The Geysers geothermal field. Geophysical Journal International, 212(2), 1157–1166. https://doi.org/10.1093/gji/ggx481

Martínez‐Garzón, P., Zaliapin, I., Ben‐Zion, Y., Kwiatek, G., & Bohnhoff, M. (2018). Comparative Study of Earthquake Clustering in Relation to Hydraulic Activities at Geothermal Fields in California. Journal of Geophysical Research: Solid Earth, 0(0). https://doi.org/10.1029/2017JB014972

Martínez-Garzón, P., Ben-Zion, Y., Zaliapin, I., & Bohnhoff, M. (2019). Seismic clustering in the Sea of Marmara: Implications for monitoring earthquake processes. Tectonophysics, 768, 228176. https://doi.org/10.1016/j.tecto.2019.228176

Zaliapin, I., & Ben-Zion, Y. (2013a). Earthquake clusters in southern California I: Identification and stability. Journal of Geophysical Research: Solid Earth, 118(6), 2847–2864. https://doi.org/10.1002/jgrb.50179

Zaliapin, I., & Ben-Zion, Y. (2013b). Earthquake clusters in southern California II: Classification and relation to physical properties of the crust. Journal of Geophysical Research: Solid Earth, 118(6), 2865–2877. https://doi.org/10.1002/jgrb.50178

Zaliapin, I., & Ben‐Zion, Y. (2021). Perspectives on Clustering and Declustering of Earthquakes. Seismological Research Letters, 93(1), 386–401. https://doi.org/10.1785/0220210127