General Comments:

The authors perform an extensive and thorough investigation on the environmental controls of the shallow-to-deep convective transition during the Amazonian wet season at the T3 site during the GoAmazon campaign. The regime classification methodology was easy to follow and robust. The findings on the importance of low-level wind shear and their estimated moisture flux convergence for each regime are novel and fascinating. The authors’ ambition to investigate myriad potential forcings may have hindered the depth of certain components of the analysis, such as the water budget closure, which could have been a study of its own. The uncertainty of rain gauge measured precipitation and surface fluxes should also be addressed.

Specific Comments:

Line 137 (Equation 2): Add a symbolic error term after CONV and address that the residual estimated CONV will contain the net of measurement errors from each instrument. Although beyond the scope, an error budget analysis within each regime would be useful to convey uncertainty in the final CONV term. The sensitivity of the water budget closure to another precipitation source would be interesting to see for similar reasons as above, since precipitation is likely the highest source of systematic error (e.g., intense precipitation from gauges is often underestimated). Lastly, it would be interesting to apply the constrained variational analysis estimated moisture flux convergence to further test your water budget closure. This could be an entire study on its own; therefore, authors are simply encouraged to mention some of these ideas in text as suggestions for future research to minimize error in CONV.

Figure 6: Panels a-b are hard to see, suggest restricting panels a-b) to <4 km and tighten x-axis or removing those panels and focus on panels c-d. For c-d), use horizontal error bars of standard deviations of differences to infer statistical significance at each level.

Lines 305-306: Would similar relationships have been found using the rain gauge averaged precipitation between 14-20 LST? That would be an interesting appendix figure or finding to mention and not show if differences are negligible.

Technical Corrections:

Lines 222-223: CWV difference at 8 LST doesn’t appear statistically significant for Deep – Cong

Lines 226-227: Change “A similar, albeit less pronounced, pattern is also observed in the lower free troposphere.”

To e.g.,:

“A similar, albeit less pronounced diurnal moistening is also observed in CWV_mid for all regimes.”

Lines 230-231: Remove “, and associated with low-level moisture divergence, which would explain the slower accumulation of moisture below 700 hPa despite slightly larger latent heat.” Current logic is too speculative.

Lines 239-242: Avoid comparing Cong vs. Deep when differences are negligible. Consider reframing certain comparisons as Cong+Deep vs. ShCu in this section in general e.g.,

“In the afternoon, MLCAPE is higher for the Deep regime (1237 J kg−1), a few hours before the late afternoon STD transition, slightly surpassing the value for the Cong regime (1111 J kg−1) and significantly exceeding the value for the ShCu regime (671 J kg−1).”

could be changed to:

“At 14 LST, MLCAPE for the Deep regime (1237 J kg−1) and Cong regime (1111 J kg−1) significantly exceed the value for the ShCu regime (671 J kg−1).”

Line 258: CWP should be CWV

Lines 259-260: Combine these 2 sentences. “By the end of the day, the accumulation of water vapor in the column is negligible. This term is also nearly zero during Cong days.”

Lines 261-262: Change “which might affect the convective regimes developing the following day.”

To e.g.,:

“which might act as a positive feedback for the continuation of nocturnal deep convection into the following day after Deep regime conditions”

Lines 265-266: Change “There is a characteristic low-level jet during the wet season. Anselmo et al. (2020) also reported an Amazonian low-level jet, occurring 10-40% of the time during March-May 2014-2015.”

To e.g.,:

“The wind profiles for all regimes peak between the 900-800 hPa layer, characteristic of a low-level jet also reported by Anselmo et al. (2020), which they observed 10-40% of the time during GoAmazon between March-May 2014-2015.”

Lines 265-272: Avoid describing relative wind speed maxima of a point sounding as “jets”.

Line 286: Word choice of “important subsidence”

Lines 288-289: Remove sentence beginning “However, solely through differences in…”