Response to Reviewer A Thank you for your helpful review. In response to the your suggestions and those of the other reviewer, we have split the paper into two, closely related parts. The first (referred to as Part I below) contains the analysis of SP-CAM cloud changes and has undergone minor revisions relative to the same content in the original submission. The second (Part II) describes the column analogue and sensitivity of the results of the column analogue to different assumptions, including: (i) resolution (expanded since original submission), (ii) the effect of diurnally-varying versus diurnally-averaged insolation, (iii) the use of differing assumptions about the momentum damping timescale (vertically-uniform versus vertically-varying), and (iv) the use of a tropics-mean temperature sounding as the reference state in the column analogue. The material in Part II has expanded since the original submission, so that we could more fully explain the column analogue and address some of the issues raised in the reviews. Despite this, we have still left some questions to the future, including the application of this framework to other cloud regimes. 1. Overall Impression (All questions here relate to Part II) ------------------------------ To address questions related to the resolution-dependence, we have pursued an expanded resolution sensitivity study in Part II. Our grid refinements are isotropic, with the fixed ratio Dx = 2.5*Dz in the lower troposphere with Dx x Dz = (50x20m, 100x40m, 200x80m, and 400x160m). While anisotropic grid refinements may be useful in finding the optimal grid spacing, we have not pursued them here in order to limit the parameter space explored in this paper. This does limit our ability to pinpoint the relative roles of vertical and horizontal resolution in degrading the representation of clouds. However, experiments with a previous iteration of a column modeling framework suggested that the SWCF and LWP decreased by approximately a factor of two with a 1km horizontal grid and the SP-CAM vertical grid, which supports in part our assertion about the role of horizontal resolution in the over-prediction of cloud fraction in SP-CAM. Our resolution study indicates that consistent results are achieved across the three finest resolutions. With the 400x160m grid, either the mean state or the cloud reponse departs from the values found on the finer grids. Based on these results, we suggest that 250x100m is approximately the "minimum" (rather than "optimal") resolution for the proper simulation of this cloud regime. The consistency of the results across the three finest grids suggests that our resolution is sufficient for this cloud regime and that, indeed, the cloud fraction decreases with finer resolutions. We would speculate that finer resolution would lead to larger cloud fractions for stratocumulus clouds. In Stevens, Ackerman & Bretherton (2003), simulations of an ATEX-like case had increasing stratocumulus cloud fractions at the inversion with increasingly finer resolutions. Note that their cumulus-under-stratocumulus case differs from ours in that their subsidence was much stronger, and their simulations had no insolation. Also, Stevens et al (2005, the DYCOMS RF01 LES intercomparison paper) showed increasing LWP with decreasing Dz down to 1m. The combination of our results with these papers would suggest that SP-CAM is likely to over-estimate the trade cumulus cloud fraction and under-estimate the stratocumulus cloud fraction due to under-resolution. We do intend to study SP-CAM's representation of stratocumulus clouds using the column modeling framework in the future. (Note that much of this discussion is included in the conclusions of the revised Part II.) 3. Comments and Questions ------------------------------ 1. Introduction - Removed 'directly' from this sentence' (Part I) - Actually the idea that the feedbacks displayed by SP-CAM are of questionable realism is something shown by this study, and not necessarily obvious a priori. This latter part of this sentence was removed as this issue is taken up later in Part II. - We have performed a short 3D simulation using the same grid spacing as the finest 2D LES simulations to show that the sensitivity of the cloud properties to resolution. (Part II) - The CRM actually has 28 levels co-located with CAM levels starting at the surface. The text was edited to reflect this correction/clarification (Part I). Second part: I think this question is asking how many vertical levels in the 'low cloud' category which is defined from 700hPa to the surface. There are 9 such levels and this is added to the same paragraph (Part I). - Reference to Klein and Hartmann (1993) added (Part I). 2. SP-CAM climatology and climate sensitivity (all Part I) - Units corrected. - The low cloud definition from 700hPa to the surface is fairly standard and the correspondence of the 700hPa limit to the level used to determine LTS is somewhat coincidental. The definition comfortably includes most of the clouds in the marine boundary layer, while there are minimal clouds between the top of the MBL and 700hPa. A note about this was added to the text. - This is not a new paragraph, just a glitch caused by the insertion of the figure. Extra text about the 'embedded CRM' has been added. The threshold is intended to exclude optically thin clouds from the definition and this was added to the text. 4. Sorting by LTS (all Part I) - The differences are computed as you state. We added a sentence to clarify how they are computed. The LTS in figure 5 is the control LTS. This is stated at the end of the caption. - There were several problems with this paragraph that needed to be addressed. We replaced the % units with just 'cloud fraction' units to alleviate some ambiguity in this regard. It appears that the 0.1% per K figure was inaccurate and this has been corrected. We calculated the change in low cloud per K change in LTS bin for the perturbed case also, as suggested, and include that in the discussion. 5. Mechanism of SP-CAM +2K low cloud response - The increase in cloud fraction and cloud liquid water just below the inversion in the +2K LES runs is suggestive of a decrease in penetrative entrainment with the higher LTS of the +2K runs. It is not certain what role penetrative entrainment at the CRM4km resolution is playing in the response. The statement in Part I about this was weakened slightly to reflect this. - No, the definition of liquid-ice static energy is correct as printed (Part I). - Figure references were added to this paragraph, and the discussion was edited slightly to try and make the paragraph easier to follow (Part I). 7.- Changed 'CRM' to 'LES' in reference to Siebesma et al (2003). (Part II) - See above response regarding differences between 2D and 3D simulations. (Part II) 8.- The inclusion of a diurnal cycle has led to better agreement between the control SWCF and that of SP-CAM in both LTS bins, so that we hope that the reviewer now finds that 'agree fairly well' is appropriate. (Part II) - The differences in the cloud responses between the LTS70-80 and LTS80-90 bins are intriguing. The cloud response in the LES runs occurs mainly just below the inversion, so that the larger cloud response in the LTS80-90 bin LES runs would seem to be associated with larger amount of below-inversion cloud in that bin. The differences in the responses at coarser resolutions of CRM4km and SP-CAM are more difficult to pinpoint. 4. Editorial Comments ------------------------------- 1. For this and other similar cases noted by the reviewer, the author guidelines specify that "sect." is to be used instead of 'Section' unless starting a sentence. 2. Corrected (Part I). 3. Corrected (Part I). 4 Figure 4 (Part I): This figure (as well as figure 9) were remade with larger fonts. We prefer to maintain the large number of colors in the figure to permit some local quantitative interpretation of regions despite the wide range of values in the plots. We did adjust the color limits in several of these frames to improve the contrast. Figure 5 (Part I): The figure caption already states that the control LTS is used. We prefer not to add a secondary axis to keep the plot from getting cluttered. The caption for figure 3 already states that each symbol represents one 5% bin which should be sufficient to identify the important percentile ranges. The same text has now been added for the caption for figure 5. 5. (all Part I) -The units were removed as they were not really needed here. -Re: Sect. See point number 1. in 'Editorial Comments' above Fig9: The fonts were enlarged for this figure. See comments for Figure 4 above. 6. The Caldwell and Bretherton references (the paper is now published) have been revised appropriately (Part II). 7. - '.' deleted (Part II) - Sect. See point number 1. in 'Editorial Comments' above (Part II) 8. space was added (Part II) References: Added this reference (Part II). Checked and revised references (both parts)