I appreciated the quality of this manuscript, which presents the results of analysis of the ratio between radiated energy and seismic moment of earthquakes taking into account different focal mechanism, depths and seismic regions. The results are expressed in terms of stress drop on the source.

The authors state that the results and their conclusions can be affected by uncertainties on the observation data and the used model.

In this respect, I would suggest to add a statistical analysis of the results in order to quantify the probability of rejecting the null hypothesis that the differences among the various cases may be caused just by random variations of the observations.

This manuscript describes the detailed work by the authors to investigate energy released by earthquakes globally. They compare energy obtained through various approaches and investigate systematic variation of energy to moment ratio for different fault types and hypocentral depths. Furthermore, they develop conversion relationships between the different energy estimates. The work is consistent with previous studies, but a valuable contribution as a consistent approach is used to all earthquakes that are part of the study. The paper is well written (although consistency of tense should be checked) and the application of methods and results are sound. However, one issue is that it remains unclear whether the authors computed energy themselves - my understanding is that ER is taken from IRIS (which is fine), but that the other energy values are computed in this study - this should be clearer throughout the text.

I have the following more detailed comments that may help to improve the manuscript:

line 1: title - 'based on' does not seem correct, change to e.g.  '...: Investigating dependence on type of faulting and hypocentral depth'

line 39: please check, what is difference between first and second radiated energy in this sentence?

line 53: dependence of ER

line 54: insert 'but' after ','

line 110: gCMT needs to be explained

line 116: Do I understand correctly that ER is taken from IRIS, and that EU and EO are computed in this study? This should be expressed more clearly throughout the text.

line 119: in case the results are taken from IRIS, this section is too detailed

line 120: change to 'we describe' - please check tense throughout the text, I give some more examples where I think it can be changed, but not for everything

line 121: which study? not clear so far if new calculations are done in this paper, or if results are taken from IRIS; if taken from IRIS perhaps not all equations are needed, unless important for the discussion.

line 185: show

line 195: show

line 206: insert apparent stress before tau (even though it is given above)

line 223: remove 'for strike-slip earthquakes' (as SS is given at start of sentence)

line 224: use 'higher depth'?

line 256: spell out symbol in ER E (Eu, Eo)

line 267: tsunami earthquake (rather than event)?

line 271: is 'seismogenic' correct here?

line 276: asperities or asperity patches

line 281: show

line 334: intense deformation - can you give example based on your results?

line 341: you focus on highest values as they are important in hazard context, but lowest values could also be interesting? overall, you could mention importance of results for seismic hazard

line 362: 'different types of faulting'

line 377: Our higher ER/M0 ....