Potential of Passive Microwave around 183 GHz for Snowfall Detection in the Arctic
Round 1
Reviewer 1 Report
Please see attached.
The authors have replied to my previous comments in a satisfactory manner. There are a few remaining questions and suggestions:
line 35 - what do the authors mean by `comprehensive snowfall monitoring?'
lines 98-99 - here it would be good to give the reader an idea of the advantage of a RF vs a single decision tree and also possible disadvantages.
lines 195-196 - What is meant by `random data samples' and `random variable samples'?
lines 195-198 - I don't understand what the authors mean. First it says that the outputs are averaged, then it says that a majority vote is used, and after this the way the predicted probability is thresholded is discussed. Could the authors indicate the quantities used in averaging and majority vote respectively (to say `output' is too vague)?
line 208 - what is meant by `integrated measurements'
line 208 - I suggest the authors say 'snowfall cases presented in the next section'
lines 240-241- I don't understand this part. It says that the BT goes 76.4N, and that this is noticeable because at higher latitudes the PMW measurements were masked by emission from the open ocean, which is warmer than sea ice. So I'm assuming they mean that the emissivity of water is higher than that of sea ice (which is not intuitive because it is not the case at lower frequencies - a reference should be added). But I would also think that if you are going from a radiometrically warmer region to a radiometrically cooler region, then the BT would go down? Probably there is something that is not clear to me here that has to do with the weighing functions.
lines 246-251 should be in their own section. Right now they are a continuation of case 2. Fig 5 the colour bar scale should be the same for the three panels (ideally the same as is
in Fig 4). Same for Fig 7.
line 310 statistical should be statistically? How was the most relevant split decided?
Comments for author File: 
Comments.pdf
Author Response
Please see attachment.
Author Response File: Author Response.pdf
Reviewer 2 Report
I have reviewed the revision to this manuscript, and have found that my earlier comments have been sufficiently addressed. I would be happy to recommend that it be published in its present form.
Author Response
The reviewer did recommend the manuscript to be publish in its present form and did not ask for revision.
This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.
Round 1
Reviewer 1 Report
The manuscript “Potential of passive microwave around 183 GHz for snowfall detection in the Arctic” by Edel et al. describes the results of a well conducted research on using passive microwave satellite sensors to detect snowfall at high latitudes. The research topic is at the forefront of research in the field and the results presented in the paper contribute to advance the knowledge on the subject. The article fits well the scope of the journal and deserves publication after a few issues are considered by the authors.
Main suggestions
The paper should be revised as to the English language, which is not quite adequate. The reviewer has tried to help the authors with several suggestions in the section “Minor corrections” below. However, a fine combing by a native English speaker would seem necessary.
The following paper recently published should be cited maybe in the Introduction: Skofronick-Jackson, G.; Kulie, M.S.; Milani, L.; Munchak, S.J.; Wood, N.B.; Levizzani, V. Satellite estimation of falling snow: A Global Precipitation Measurement (GPM) Core Observatory perspective. J. Appl. Meteor. Climatol. 2019. [https://doi.org/10.1175/JAMC-D-18-0124.1]
The variable SIC is never introduced and not listed in the “Abbreviations” at the end of the paper. This creates a dark corner in the explanation of the algorithm. Needs to be explained in the text.
It is my opinion that the random forest algorithm used in the work should be a bit more explained also using some literature. The way it is detailed now is not sufficient.
Minor corrections
line 27 “important” —> “relevant”
line 28 “the” needed before “hydrological”
line 31 “independance to” —> “independence from”
line 33 “is up to now the best instrument” —> “has proved to be best sensor to date”
line 33 “latitude” —> “latitudes”
line 35 “baterry” —> “battery”
line 37 “such as the Microwave Humidity Sounder (MHS) among others” —> “such as for example the Microwave Humidity Sounder (MHS)”
line 41 “challenge” —> “challenges”
line 44 remove “long”
line 46 “to” —> “for”
line 48 “multiple difficulties may arise and complexify the snowfall retrieval” —> “several problems may arise and complicate the snowfall retrieval”
line 51 “shapes, their size, as well as their density” —> “shapes, sizes and density”
line 52 “complicate” —> “substantially influence”
line 56 “algorithm” —> “algorithms”
line 59 “above” —> “over”
line 60 “indispensable” —> “unavoidable”
line 61 “Liu and Seo [21] developed…”
line 63 “Although,” —> “Although”
line 66 “[27]” —> “Rysman et al. [27]”
line 73 “additionnal” —> “additional”
line 74 “function” —> “functions”
line 80 “[32]” —> “Surussavadee and Staelin [32]”
line 103 “since it is” —> “since they are”
line 104 “CloudSat” —> “CloudSat’s”
line 107 “humidty” —> “humidity”
line 109 “sensitivy” —> “sensitivity”
line 112 “In the latter” —> “In the latter case”
line 116 a reference should be provided for RTTOV, preferably the most recent one
line 122 “slighlty” —> “slightly”
line 127-128 “It measures 125 vertical bins of 240 m thick each” —> “It measures 125 240-m thick vertical bins”
line 132 “gives” —> “includes”
line 135 “gives” —> “contains”
line 136 “permits to ensure detection of” —> “allows to detect”
line 142 “provides” —> “includes”
line 157 “not been” —> “not have been”
line 167 “with the” —_> “with”
line 191 “to —> “too”
line 214 “2” —> “two”
line 237 “significations” —> “meaning”
line 243 “mild” —> “moderate”
line 270 “the temperature are not too cold yet” —> “the temperature is not too low yet”
line 276 remove “in order”
line 277 “tend” —> “tends”
line 279 “variables” —> “variable”
line 286 “splitted” —> “split”
line 298 “attests” —> “testifies”
line 321 “in” —> “at”
line 330 “meaning that,” —> “means that”
line 343 “Spatial” —> “Space”
line 351 “incertitudes” —> “uncertainties”
line 355 “ways” —> “methods”
line 363 “arctic” —> “Arctic”
line 364-365 “… may be associated with phase attribution” Not clear what it means. Clarify.
line 365 “cases to” —> “cases for”
line 366 “environement” —> “environment”
line 369 “closer temperature at 2 meters and humidity than ERA-I compared to polar stations” —> “temperature at 2 meters and humidity closer to measurements of the polar stations than ERA-I values”
line 370 “go closer to” —> “approach”
line 373 “irrelevant” does not mean anything in this context. Please check what you want to really say.
line 374 “[18]” —> “Skofronick-Jackson and Johnson [18]”
line 378 “prevalent” Are you sure it is prevalent in the Arctic? I would remove this adjective.
Reviewer 2 Report
Please see attached
Comments for author File: Comments.pdf
Reviewer 3 Report
Line 19 and throughout. Capital "A" on Arctic . There are numerous style and grammar issues that will need to be addressed prior to publication.
Line 22 type: Proje(c)ted
Line 34 "revisit time" does not feel like the proper technical term here. Try "fly-over interval" or "orbital frequency"
Line 65: what would cause a false alarm in cold conditions?
line 82: "transfer model"
There are other passive microwave studies available at high latitude that might be of interest to you by L. Candlish, R. Raddatz.
From what I'm reading, this is an assessment of whether your method can detects "occurrence" of snowfalls. Is there a way to determine snowfall rates / intensity?
Conclusion 3: The 10.2% false negative and 8.7% being positive seem to indicate this method has some high degree of uncertainty of performance. CAn you comment on this at all?
Conclusion 5: Yes, this would be very interesting to an end-user, especially if one could infer snowfall accumulation from it.
Are your results significant by a statistical test? I think you need to look at ways to give your results some more depth.
