Dear authors,

Thank you for your revised submission. I sent it to the two previous reviewers for comment and they felt that, while the requested technical corrections had been made, the overall limitations of the algorithm made it a borderline case for publication. I take these comments seriously, and as a result sought an opinion from another subject matter expert. Based on their comments and my own reading I do see the value in eventual acceptance of the study, for two main reasons: (1) the applicability to comparatively low-cost sensors is important and expands potential applications beyond current techniques which require more expensive sensors, and (2) to ensure that the algorithm associated with eventual data sets is available in the public peer-reviewed literature and accessible to researchers, as there is quite high interest in these systems and algorithms.

Two of the reviewers made additional suggestions for minor revisions, which I agree with. I would be grateful if you would further revise your manuscript with attention to the following points:

1. The introduction to the paper and Table 2 implies that the early work in this subject area was carried about by Long (which is the earliest cited reference about cloud cover estimation early on). However early sensors and some of the techniques discussed were pioneered earlier, in the 1980s. This early development is described in a review by Shields et al (2013): J. E. Shields, M. E. Karr, R. W. Johnson, and A. R. Burden, “Day/Night Whole Sky Imagers for 24-h cloud and sky assessment: history and overview”, Appl. Opt. 52(8), 1605-1616 (2013) https://www.osapublishing.org/ao/abstract.cfm?uri=ao-52-8-1605 I suggest rewording the introductory paragraph (and possibly Table 2) to briefly discuss this paper and provide the historical context.

2. Provide a little more detail in the concluding remarks about the limitations with respect to aerosol. The current manuscript acknowledges limitations in areas where aerosol loading is variable in time. It would be good to give more specifics on when/where that is or is not likely to be the case, as well as some quantification of changes in aerosol levels which may confound the algorithm. Section 3.2.2 references work by Ghonima et al (2012) on this topic; if directly applicable these numbers could be reported again in the conclusions in the context of their effect on the performance of ELIFAN and global patterns of aerosol loading.

Please let me know if you have any questions; I look forward to receiving your revision.

Best wishes,

Andrew

Dear authors,

Thank you for submitting this further revision of your manuscript. After reading through myself, I feel that the remaining points were addressed sufficiently, and consequently I am pleased to accept this version of your study for publication in AMT.

Best wishes,

Andrew