General Comments

This manuscript nicely quantifies the influence of moisture in the portion of the lower troposphere that is just above the boundary layer on the development of convection over dry or wet soils. The authors use radiosonde observations from the US Southern Great Plains field site along with an entrained parcel buoyancy model. Their results demonstrate an important role for lower tropospheric humidity in convective triggering in dry-coupling convective events; they also show that this is far less important in wet-coupling events.

I have two main points I think should be addressed before this paper is ready for publication. First, I don’t think these results should be presented as contrary to the CTP-HI_low framework, but as a quantification and elucidation of the negative feedback behavior that is discussed at length in Part II of the CTP-HI_low framework papers (Findell and Eltahir, 2003b, citation provided below). Second, the process quantification in this paper centers around your reconstructed land-coupled humidity profiles. I think you need to provide much more information (along with figures) about these profiles. I feel like they are at the heart of your story, and yet very little information is provided about them. Are they systematically different from the observed humidity profiles? (For example, are they systematically more humid?) Can you show us a plot of the data that the regression is based on? This seems central to your results, so I feel it is important to be clear and transparent about what information is actually captured by the term that leads to attribution of land-driven coupling. Some of this information is revealed through Figure 5, but I think you need to be explicit about b_LC as soon as the concept is introduced.

I offer a few additional minor suggestions below. I feel that this paper will be a nice contribution to the literature once these issues are addressed. I believe that all the suggestions qualify as minor revisions. I look forward to seeing this work in print!

Respectfully submitted,

Kirsten Findell

Specific comments

• Abstract: I think the abstract would be improved by clarification of what you mean by lower troposphere since HI_low does capture some of the lower troposphere. On line 51 you say “lower troposphere (LT) above the PBL”: this distinction should be in the abstract, too.
• To further the point mentioned above about consistency between your results and the CTP-HI_low framework, the discussion of physical mechanisms leading to the negative feedback region shown on the map in Figure 2 of Findell and Eltahir (2003b) focuses on the topographic and dynamical circumstances that commonly lead to the presence of a layer of moist air coming off of the Mexican plateau with its base at about 850 mb. As you note, since HI_low is calculated from the humidity deficit 50 and 150 mb above the ground surface, it might not capture this layer of moist air. Nevertheless, I think the results you find are quite consistent with the process understanding that was enabled by Parts I and II of the CTP-HI_low framework papers. Your sentence about lateral entrainment of moisture on lines 55-57 seems to echo the discussion of processes leading to the negative feedback regime provided in Findell and Eltahir (2003b).
• The SGP site is not in the negative feedback region shown in Findell and Eltahir (2003b), but in the transitional region just to the east. In this region, negative and positive feedback days were shown to occur with approximately the same frequency. This is also consistent with your results.
• Interestingly, while working on the project that eventually led to Findell et al. (2011), I searched at length for an improvement to HI_low, trying to determine the best atmospheric levels to consider for a humidity deficit metric. In the end, I did not find a perfect level or set of levels. Instead, I found that some measure of the humidity deficit was necessary, but I could not conclude that inclusion of higher atmospheric levels would improve on the insights gained from HI_low. Your work clearly indicates that there are times when higher-level moisture information is needed; I wonder if other times the higher-level information actually muddies the water. (No recommended action here, just some interesting things to think about.)
• On line 148 you indicate that “a strong increase in moist static energy from the land surface moisture” is the same as a small HI_low. This is not accurate. That quoted phrase is consistent with a large latent heat flux.
• Around this point in the paper (~line 148), it became clear that your treatment of small HI_low values as a surrogate for wet soil advantage days neglects the portion of the CTP-HI_low framework that is so humid (such low HI_low values) that any surface flux partitioning can trigger convection (Labeled “Atmospherically controlled days with convection over wet or dry soils” in the CTP-HI_low framework schematic).
• Figure 4 has only 1 panel, but in the paragraph beginning on line 216 you discuss panels a, b, and c. I think you mean to refer to different line types.
• Final paragraph of the results section: You talk about differences between deep, shallow and convective congestus. I think you should show some figures to help highlight these results.

References

Findell, Kirsten L., and E A B Eltahir, 2003b: Atmospheric controls on soil moisture-boundary layer interactions. Part II: Feedbacks within the continental United States. Journal of Hydrometeorology, 4(3), 570-583.

Findell, Kirsten L., Pierre Gentine, Benjamin R Lintner, and Christopher Kerr, June 2011: Probability of afternoon precipitation in eastern United States and Mexico enhanced by high evaporation. Nature Geoscience, 4(7), DOI:10.1038/ngeo1174.

Please find my comments in the attachment.