General comments

This paper presents comprehensive physics-based database of injection-induced seismicity with descriptions related to its physical and statistical implications for understanding and forecasting of induced seismicity. I think the database adequately covers the vast range of physical parameters from worldwide injection-related geo-energy applications. The manuscript is well written and structured. I see the paper as a useful contribute to ESSD, if the below minor comments are handled appropriately. 

Specific comments

1. This study seems not to discriminate the term “induced” and “triggered”, although both terms indicate different nucleation mechanism in some studies (Dahm et al., 2015; McGarr et al., 2002, Ellsworth et al. 2019). Particularly, Ellsworth et al. (2019) classified the mainshock of the Mw 5.5 Pohang earthquake as triggered seismicity, which was initiated by anthropogenic forcing and propagated beyond the bounds of the stimulated region. Other seismic events that occurred during hydraulic stimulations were termed as induced seismicity, of which magnitudes are limited to within the spatial dimension of the stimulated volume (Kim et al., 2022). As mentioned in the manuscript, Mw 5.5 Pohang earthquake is regarded as a representative counter-example of magnitude scaling relations driven by McGarr (2014), but McGarr and Majer (2023) argue that the relationship is intended for earthquakes induced, not triggered. Thus, some descriptions regarding the term “induced” and “triggered” might be needed in the manuscript for the further understanding of the physical mechanism of seismicity.

2. In Figure 5, parameters with the logarithmic scale of y-axis such as permeability, maximum injection rate, maximum injection volume show that the mean value is plotted outside of the boxplot probably due to extremely large or small value of outliers. Particularly, the permeability for “research” is generally one or two order lower than the permeability for “geothermal”, but mean values of both types indicate contrasting result. Mean values calculated by excluding the outliers can better represent the characteristics of the given parameters. 

Technical corrections

1. L172. The project number is missing in first column of the given excel file.

2. Host rock properties in Figure 2. There is a typo in the unit of thermal expansion coefficient. (1/°K → 1/K)

3. L193: “hydrogeological” properties needs to be changed to “thermal and hydrogeological” properties, as thermal conductivity and thermal expansion coefficient are included in the reservoir rock properties.

References

Dahm T, Cesca S, Hainzl S, Braun T, Krüger F. Discrimination between induced, triggered, and natural earthquakes close to hydrocarbon reservoirs: a probabilistic approach based on the modeling of depletion-induced stress changes and seismological source parameters. J Geophys Res Solid Earth. 2015;120(4):2491–2509. https://doi.org/10.1002/2014JB011778.

McGarr A, Simpson D, Seeber L. Case Histories of Induced and Triggered Seismicity in International Handbook of Earthquake And Engineering Seismology. San Diego, CA: Acad. Press; 2002.

Ellsworth WL, Giardini D, Townend J, Ge S, Shimamoto T. Triggering of the Pohang, Korea, earthquake (Mw 5.5) by enhanced geothermal system stimulation. Seismol Res Lett. 2019;90(5):1844–1858. 

https://doi.org/10.1785/0220190102.

McGarr A. Maximum magnitude earthquakes induced by fluid injection. J Geophys Res Solid Earth. 2014;119(2):1008–1019. https://doi.org/10.1002/2013JB010597.

McGarr A, Majer EL. The 2017 Pohang, South Korea, Mw 5.4 main shock was either natural or triggered, but not induced. Geothermics. 2023;107. 102612.

The paper by Kivi et al introduces a global physics-based database of injection-induce seismicity.

I recommend acceptance with the option of making the following very small number of minor tweaks.

The authors provide a publicly available database which is very positive and will without doubt be welcomed by diverse stakeholders and researchers. Providing information to fill in critical gaps in the HiQuake database for well-documented cases is a very positive step.

The early part of the paper provides a good concise review of the subject. The paper continues on to explain the rationale of the database and details of its contents. The work was clearly thought through very carefully and reasons for decisions are given.

The paper is authoritatively written and the relevant descriptions of data categories thorough. The database accords with official international standards. The illustrations and tables are elegant and clear.

The English is excellent throughout the main body of the paper. I noticed a small number of awkward/wrong expressions in the Abstract that would benefit from a light editing: 1) “if intensely shaking the ground” should be reworded, 2) “these data are hardly gathered” would be better “these data are challenging to gather”, 3) “data aims” should be “data aim”, 4) “Conclusively” could be deleted.

Line 125: It is not clear what is meant by “We rely on HiQuake in the consensus that anthropogenic adtivities induced all reported earthquakes.” Please rephrase this. HiQuake lists all earthquakes proposed, on the basis of scientific arguments, to have been human-induced. Simple inclusion in HiQuake says nothing about the likelihood that the proposals are true, and cases vary from highly unlikely to virtually certain to be human induced. It would be very helpful if this critical but widely unappreciated feature of HiQuake could be made clear here in the text.

Line 186: Foulger et al. (2018) mention three CCS projects that are thought to be seismogenic.

Section 2.2.6: It would be useful to cite errors in the b-values mentioned as the errors in b are frequently larger than the differences between the b-values under discussion. It is thus common to read interpretations of variations in b that are not statistically significant. The text at line 548 onwards also needs attention in this respect.

Gillian R. Foulger