Marine Stratus—A Boundary-Layer Model

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Dr Peter Taylor attempted to design a 1D model to simulate and understand the marine stratus cloud. While the idea is innovative, the demonstration of the topic requires substantial enhancement. This read like a technical report. Also there are unfound generalisations in the manuscript text. The figure quality is also not good. The figures are also not properly described in the text. 

Comments on the Quality of English Language

ok

Author Response

While I have made revisions in response to other referees comments I generally ignored this unduly negative and dismissive review.

Reviewer 2 Report

Comments and Suggestions for Authors

This paper presents a simple one-dimensional model used to study marine stratus. I find the paper

marginally acceptable due to its simplicity and lack of summary of previous similar efforts which

surely must exist. The impression it gives is of someone new to the field, but there may be some value to

having the simplest possible model of these clouds. Apart from the lack of review of previous work, another

deficient area is that the methodology and model are not fully enough described for anyone to be able to

repeat this work. I also noticed some other possible errors in the description. I will detail all these

points below. I classify this as a major revision due to the number of issues I am listing that are important

to address. But the results are credible, so it can become publishable.

 

Specific Points

 

1. The paper lacks a review of previous literature and consequently also lacks any comparisons in the

summary discussion. These are important to include for publication.

 

2. The models of this type also can potentially include a mean W vertical velocity profile as many real

cases may have large scale subsidence or convergence. The fact that this model assumes it is zero should

be mentioned.

 

3. The results imply that there is surface friction but the momentum equation shown only has a vertical

mixing term. What is done at the surface to provide friction?

 

4. Eq 10. The shortwave scheme has no reflection term which is why the upward part shows no albedo effect

above the cloud. This is a simplification that should be mentioned.

 

5. Eq. 11 and above. I am concerned about the dimensions of variables in these equations. Eq. 9 implies

RFU is W/m2 so Eq. 11 is W/m3, but RFDIV  also appears in Eq. (2) as a heating term. I sense some

inconsistency in units. 

 

6. The experiments seem to have surface fluxes of heat and moisture but none are described.

 

7. Line 150 and others. The pdf I have shows d(Chi)/dz instead of d(Theta)/dz. There is a problem with

the Greek character.

 

8. Discussion. While the results look credible, there is far too little on other similar work and what

is learned through this exercise. It comes across as just an exercise of a new model and not really

an advancement in science.

 

Author Response

These were useful comments and good specific points. I have added more discussion of previous work (points 1, 8), including Oliver et al, 1978, and some more detail of the "adjust" procedure. I also added a statement that W = 0 (point 2). There are comments later, p13, that subsidence may be a factor in dissipating cloud and at some point we could look at this in an approximately 1-D model.

Regarding points 3 and 6, we assume U,V = 0 at the surface and so surface stress is determined by the roughness length, and the eddy viscosity/diffusivity formulations (Eq 8). Similar factors determine surface fluxes of heat, Q and QL.

Point 4 is a concern and we need to deal with backscatter of solar radiation and cloud albedo.  This problem is noted (p 5, 13) in the text.

Point 5 is valid and I have corrected Equn 2 in the text by dividing (RFDIV+LHT) by ρcp and pointing out that RFDIV and LHT are per unit mass of air.  It is correct in the code!

Point 7.  Corrected.  Partly a problem with different fonts in equations and text.

Point 8.  I have added some comments on differences with earlier work by Oliver et al.

Reviewer 3 Report

Comments and Suggestions for Authors

Today enormous importance is given to complex and highly articulated numerical applications. However, it is often extremely important to return to using idealized, or less complex, models in order to be able to improve those components that then characterize Earth System Models (or complex numerical systems).

The simplifications applied in this article are not trivial. They can crucially characterize the use, tuning and application of certain parameterizations of state-of-art models. The proposed work is a brief application of 1D-RANS model for PBL evolution includes water vapor and cloud droplets. Nevertheless, the work is described in a somewhat approximate way, especially formally. There is no summary description of the conclusions in the abstract. In the paper the discussion of the results is limited to a few lines, so are the conclusions. Although thepaper is well structured, interesting and well described (where it is!) I suggest reevaluating the paper after discussion and conclusions have been expanded and thoroughly described.

Author Response

I have added more discussion of previous work, including Oliver et al, 1978, and some more detail of the "adjust" procedure for vapor-droplet transitions. I have extended the conclusions section, but not by very much. More work on the treatment of solar radiation is needed along with other improvements to the model, but given the deadlines I cannot include these in this paper.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors made no comments to my review. I strongly reject the manuscript.

Comments on the Quality of English Language

Ok.

Author Response

I am sorry that this anonymous reviewer has problems with this paper but chose not to explain why.

Reviewer 2 Report

Comments and Suggestions for Authors

I am not yet satisfied with what has been added to the paper based on my previous review.

1. It is very important to include enough information in the paper for the experiments to be repeated by others. As in the first round, the paper lacks a description of how surface fluxes were handled even though they were in  the reply to the reviewer. This should be added to the paper for acceptance.

2. Similarly the radiation terms, RFU etc. will need surface or upper boundary conditions that don't appear to be mentioned. Solar downward radiation is a critical term as is longwave upward radiation from the surface.

3. Also, although clear-sky longwave radiation terms are omitted, some comment on this is needed. Longwave cooling is important over multiple hours.

Again, overall this paper must include complete descriptions of methods to be accepted.

Author Response

Thankyou for useful suggestions

Concerning Point 1: Details of the treatment of the lower boundary have been added, saying,

 Lower boundary conditions on the water surface are U = 0, Θ = T­_surf , the surface water temperature, Q = QSAT (T_surf) and QL = 0. The surface can thus be a source of water vapor but is assumed to be a sink for cloud droplets as they collide and coalesce. Fluxes of momentum, heat, water vapor and liquid water evolve as a part of the solution and depend on the assumed roughness lengths. These can differ, z_0m, z_0h etc. but are presently all set as z₀ = 0.001m.

Point 2: For radiation there is some detail in section 3.3 but I have added more  in section 2, saying,

Boundary conditions are needed on up-welling radiant fluxes at the surface and down-welling fluxes at the model top. These are specified in Section 3.3, based on black body RFU at the water surface and a relatively low albedo (0.05 in cases here) for solar irradiance. At the top boundary (3000 m in this case) we specify a typical, clear sky value for RFD (200 Wm^-2 here) and must specify SFD. In realistic simulations this will have a strong diurnal cycle but in the test case considered in section 3.3 we simply set SFD = 250 Wm^-2 and hold it constant.

Point 3:   Clear sky downward LW fluxes are included and LW cooling at the cloud top is evident in Fig 9a. Throughout most of the cloud layer upwelling and downwelling LW radiation are equal. in the discussion of Figure 9 I say...

 Results at the 60 h point show upwelling and downwelling long wave radiation both equal to black body emisssions at the cloud water temperature through most of the cloud layer (Figure 9c). Near the cloud base (~500m) where QL is lower there still some unabsorbed radiation from the underlying water surface so that RFU > RFD. Clouds appear at around 30 h, as in the case with no radiation, after which the liquid water content of the cloud increases with time (Figure 9b) and has a maximum near the cloud top where radiational cooling is lowering temperatures as illustrated in Figure 9a.

Further investigation of the radiative flux divergence contribution to cooling of the cloud top will be undertaken in the future but the difference between RFU and RFD and the gradient d(RFU-RFD)/dz just at the cloud top will lead to negative values of RFDIV and, via Equation 2, cloud top cooling. Text added is...

 In Figure 9c one can see that, just at the cloud top level, d(RFD-RFU)/dz will be negative, and, with no solar component, RFDIV is also negative and will cause the cloud top cooling.

Thanks for drawing my attention to these issues.