This article provides a fairly comprehensive review of the 6-year oscillations in the Earth system and proposes that the core processes, widely believed as the driver of the 6-year oscillations in LOD and magnetic field, is also responsible for the newly discovered similar 6-year oscillations in the climate system. This article is well written, and the authors have demonstrated a thorough understanding of the current progress in the field. I only have a few minor comments.

• In the abstract (line 18), the newly discovered 6-year oscillations are not just in the atmosphere, but also in the hydrosphere, ocean and cryosphere. I would suggest replacing “atmosphere” with “climate system” or “surface geophysical fluids”.
• Lines 44-45, “A recent study by Watkins et al. (2018) …”: It was a study published almost 5 years ago and should not really counted as a recent study. Also there was a parallel paper (Ding and Chao, 2018) published at about the same time on the same topic (6-year oscillation in surface deformation). The authors may want to cite this one as well, which provides some insights on the likely coupling mechanism.

Ding, H. & Chao, B. F. (2018). A 6-year westward rotary motion in the Earth: Detection and possible MICG coupling mechanism, Earth and Planetary Science Letters., 495, 50-55, https://doi.org/10.1016/j.epsl.2018.05.009

• Unlike the well-known 6-year oscillations in Earth rotation and magnetic field, the similar 6-year oscillations in the climate system have only been recently reported. Can the authors elaborate a little bit on how significant those signals are or how confident people are on detection of those signals?
• Line 92, “As LOD changes are well explained by deep Earth processes”: This is not really true. Although we believe that deep Earth processes are responsible for LOD changes at several years and longer time scales, the mechanisms are still not clear. The authors may want to rephrase the wording here.

There does seem to be a whole range of phenomena in the Earth system with a "period" of about 6 years. My own work suggested a very clear 6 year period (actually 5.88 year period) for \Delta LOD from 1962-2011. This fit appeared very imptessive - certainly enough to solidify my career! In retrospect, I think I did this work and published the paper at exactly the right time - but for my career, not for broader scientific understanding! By looking at the LOD series in detail post 2011, it seems that the 6 year period breaks down, with peaks in 2010,2014 and 2019 or so. So it was a good thing that I investigated when I did, otherwise things would have been much more complicated!

More generally, looking at LOD prior to 1962 (occultation data constrained) shows power at around 6 year, but generally at periods slightly greater than 6 years. This means that an assumption of a modal interaction seems more complex than I had suggested. Perhaps this is not surprising - with so many different elements making a contribution at about this period, there is a likelihood of a disruption of the simple oscillation. and most significantly, this disruption is uneven - suggesting that the signal is non-stationary. In this case, more complex Fourier analysis will not help with further understanding - I do not suggest wavelets aren't wonderful, but when the changes happen on the same time scale as the oscillation, overinterpretation is a risk.

As a result, this paper is to be welcomed as a start, but only as a start. With no clear methodology as to a mechanism of the coupling, a phenomonological approach is necessary - while correlation does not imply correlation, lack of correlation is certainly likely to allow rejection of causation. So it is interesting to start with the different phenomena with the similar periods, but what is the phase difference bewteen these "oscillations"? And is that phase diffrence consistent over time? Hopefully this paper will drive more detailed and intensive analysis of the data time-series to address this question.