Evaluation of the Microphysical Assumptions within GPM-DPR Using Ground-Based Observations of Rain and Snow

Round 1

Reviewer 1 Report

This article presents an interesting investigation into the microphysical assumptions within GPM-DPR using a substantial data set of ground-based observations. The size of the data set considered should be referred to in the abstract. It is generally well constructed and clearly written.

However, the article could be improved by addressing the following points:

• Retrieval error metrics

It would be useful to include a measure of bias amongst the retrieval error metrics. Whether the microphysical assumptions result in biases is of fundamental importance. This should perhaps be expressed in percentage terms, as following the separation into stratiform and convective PSDs there is clearly a significant difference in the average rainfall rate between the two classifications. As such, it is also difficult to make comparisons between the MAEs in Tables 1 and 2 (as reported in the abstract), because the underlying distributions are significantly different (i.e. stratiform rainfall typically about 0.5mm/hr and convective about 7mm/hr comparing MAE and MAPE).

The results presented in the abstract should focus on the separated data (Table 2) in terms of percentage errors and biases. Switching between absolute values (for rain in the abstract) and percentages (for snow in the abstract) should be avoided.

• Spherical drops

The effect of assuming spherical drops is reported (line 169) to result in a reduction of less than 10% in R. It should be made clear if this is based on observations or what assumptions have been made. Does this bias depend on R?

Also, there is no mention of how the assumption of spherical drops affects the equivalent radar reflectivity factor (Eqtn 6). This could have a significant influence on the results and deserves consideration.

Minor points:

1. Line 37: isn’t it a ‘power-law’ rather than an ‘exponential’ relationship?
2. Sigma is missing the subscript ‘bsc’ in eqtns 6, 15 and 21.
3. The parameter ‘a’ should be italicised in the text (line 422, twice on line 430)

Author Response

Please see the attachment for our response to Reviewer 1's comments. 

Author Response File: Author Response.docx

Reviewer 2 Report

The authors present a very interesting, detailed, in-depth analysis of the GPM-DPR official algorithm (V06) rainfall and snowfall rate retrieval assumptions. An extensive ground-based dataset of precipitation microphysics from NASA GPM field campaigns, and from other sources, have been used. The paper is very well written, the methodology, in spite of its complexity, is very well presented. The authors provide sound evidence of important issues in the main assumptions behind the GPM-DPR (Ku- and Ka-band radar) snowfall rate retrieval algorithm, while they provide evidence of the more solid assumptions for the rainfall rate estimates.

I think that the paper deserves publication, although some minor corrections are needed first.

I think that both in the Title and in the Abstract more emphasis should be given to the main outcome of the paper, which is the verification of the assumptions behind the GPM-DPR snowfall rate retrieval algorithm (the Introduction for example, focuses mainly on the importance of retrieving snowfall). 

In the abstract, in order to compare snow and rain results, please indicate MAPE (%) both for rain and snow.

Line 37: order of magnitude. What does it mean?

There are many equations, maybe some of them could be simplified:

Eq. (2): The denominator is the LWC.

Eq. (4). The water density is constant, can be eliminated in the equation

Eq. (6) and Eq. (21) σ should be σ_bsc

Simplify Eq. (19), (21) and (22) by using a concise way to refer to N(D), e.g., N(D)=N_w f_μ,Dm (D)

Figure 2: in the caption, please clarify what are the dashed lines (rainbow) vs. the solid lines (red and blue), it is not clear. In panel b it is not clear where is the black line with ε = 1 the authors refer to (“The default relation (ε = 1) is in black, while ε > 1 is in red shades and ε < 1 is in blue shades).

Line 267: Add the reference for the ATBD and specify what ATBD it is.

Line 326 Specify that “error” is MAPE

Line 330: Correct: would like struggle

Line 451 missing parenthesis. Please correct

Author Response

Please see the attachment with our responses to Reviewer 2's comments. 

Author Response File: Author Response.docx