Comment on amt-2021-181

Line 1: ---------------"Although ALADIN is optimized ....." Abstract: Line 11: -----------------The last line is badly worded. I suggest "This is illustrated using Saharan dust aerosol observed in June 2020". Page 1: Line 19: ----------------I find this short description awkward and not accurate enough. I suggest something like: "Two separate main optical detection channels are implemented on-board ALADIN. They are referred to as the Mie and Rayleigh channels. Both channels detect a mixture of particulate and molecular scattering. However, the primary task of the so-called Mie channel is to detect the spectrally narrow (FWHM on the order of 10s of MHz) return from atmospheric hydrometers. The Rayleigh channel primarily detects the spectrally broader (FWHM of several GHz) backscatter from atmospheric molecules." Page 2: Line 32 ---------------Delete the "In" directly after the reference to Winker et al. Page 2: Line 33 ----------------2006 was 15 years ago. I think you can delete the "already"...perhaps "previously" was meant. Page 2: Line 34 ---------------Please be specific. What is "all the available information" ? Page 2: Line 46. ---------------This is a very interesting point. Please provide a reference (even if it is only a tech note or report). Page 2: line 53 ---------------Please mention how can the general community get access to the updated L2A ATBD. Page 2: Last line -----------------"...followed by a conclusion" ==> "...followed by a conclusion section". Page 3: Line 68 ----------------I am confused by the reference to the "..previous 24 sec cycle of the burst-mode operation of the laser". Previous to what ? Was this burst mode used early on in the mission ? If so, why was it no longer used ? Or, was it something previously planned but not implemented ? Page 3: Line 75 ----------------"Fine bins.." ==> "Finer resolution bins..." Page 3: Lines 83-94 and Fig. 3 ------------------------------I found that the discussion of the spectrometers to be very confusing ! Only after reading through the L1 and L2 ATBDs, it became clear that the Rayleigh A and B signals are the result of integrating the images projected on CCD detectors. So for the Rayleigh channels, for each time-height bin two spectrally integrated measurements are available. This should be explained here. For the Mie channel I found the presentation here to be misleading. The text and Fig. 3 first had lead me to believe that in the case of the Mie channel, that the data yieled by the device was a spectrum such as that iluustrated in Fig.3. It took some time and iterating between the two documents, to realize that that for the Mie channel, that the curve shown in Fig. 3 corresponds only to the central position of the Fizeau wedge ! Only after reading through the L1 and L2 ATBDs I understood that there are 16 different spectral channels available. Further, the response of each channel is the result of integrating the spectrometer output image along the different columns (corresponding to wavelength shift). Since the central wavelength varies as a function of Fizeau wedge position, the measurement will consist of the INTEGRATED filter spectral response (e.g. as shown in the bottom right-panel of Fig. 3) with the center frequency shifted according to its position along the CCD rows. I understand that the author's would likely desire to keep the explanation concise, however, the presentation here really needs to be more detailed and accurate ! It did cost me some time to understand what was being shown here and how the instrument really functioned and I am sure this would also hold true to many other readers in the general community. Page 6: Lines 104-109 ---------------------The description of the "MCA" is likely incomprehensible to anyone not intimately involved with the data processing itself ! What does "some sort of cross-talk correction" mean ? What is the L1B-derived scattering ratio ? Either provide more details about the MCA (even references to the appropriate sections of the publicly available ATBDs would help) or, if it is deemed not essential, to the paper just skip it. The same general comments apply to the description of the ICA. Page 6: Lines 110 -----------------"At last.." ==> "Lastly,.." Section 2.2.1 --------------See my later comment (Page 13: Line 252) It would be useful for the general reader if it were to be explained what advantages (or disadvantages) the SCA method have compared to the usual method of determining extinction by calculating the log-derivative of the Rayleigh ATB profile ? Off hand. I can think that the low vertical resolution bins dealt with here may be a factor. Is this correct ? Are any multiple-scattering considerations taken into account in the retrieval. It looks like they are not. Do you expect this to have any impacts on e.g. cirrus cloud retrievals ? Page 6: Line 118-119 ---------------------"concision" is rarely used in modern English. I suggest "brevity" or "conciseness". The sentence is awkward: I suggest something like: For the sake of brevity, only an outline of the SCA algorithm is presented here. Only the main features of the algorithm, necessary to understand the subsequent sections, are covered." Page 7: Line 129 -----------------Delete the "(dR(z)=R'(z)dz)" It is trivial and does not add anything to the presentation. Page 7: Line 144 ----------------"..equation (2)" ==> "..equations (1) and (2)." Page 8: Line 180 ----------------"..thermal constraint on the primary mirror..." does not make any sense here. Do the authors mean "thermally induced distortion" or "thermal strain" ? What is meant by "orbit phase" ? Do the authors mean the "orbit position" ? Does the distortion vary predictably along the orbit or is function of the e.g. solar background ? Line 183: "...called the Instrument....(IRC) mode,..." Line 184" "...target with negligible Doppler shift due to the nadir pointing." Page 9: Line 190 ---------------(Also relevant to Eqns. 5 and 6) What is the maximum height given by the AUX_MET product. Is there any account given to the Rayleigh transmission between the top to the AUX_MET product and the top-of-atmosphere ? Page 9: Line 198 ----------------"Constraints" ...see my comment (Page 8: Line 180) Page 9: Line 210: ----------------------It would be useful if the authors could elaborate on this point a bit. For example, what order of magnitude error do they believe background aerosol levels may have on the accuracy of the calibration ? Page 10: Line 215

imparting on the reader of the paper no real feel at all for the quality of the data. To this end, the authors should include additional examples, for example, showing: -profiles and 2-D plots of the Aeolus Attenuated backscatter (both before and after cross-talk correction).
-profiles and 2-D plots of the retrieved extinction and backscatters.
-comparisons of the extinction and backscatter retrieval results for the MCA and SCA algorithm.
The above examples should, ideally, span an appropriate number of representative cases (e.g. cirrus clouds, light and heavy aerosol loadings etc..) My specific comments follow.

================= Specific Comments =================
First off, I am puzzled by the use of "correct" in the name of the algorithms refered to by this paper. There are other approaches to inverting HSRL signals to derive extinction and lidar ratio that are mathematically valid. What is special about these algorithms that make them "correct" ? It would be useful to the reader if this point was somehow addressed in this paper.
Page 1: Line 19: -----------------I find this short description awkward and not accurate enough. I suggest something like: "Two separate main optical detection channels are implemented on-board ALADIN. They are referred to as the Mie and Rayleigh channels. Both channels detect a mixture of particulate and molecular scattering. However, the primary task of the so-called Mie channel is to detect the spectrally narrow (FWHM on the order of 10s of MHz) return from atmospheric hydrometers. The Rayleigh channel primarily detects the spectrally broader (FWHM of several GHz) backscatter from atmospheric molecules." ----------------Delete the "In" directly after the reference to Winker et al.
----------------This is a very interesting point. Please provide a reference (even if it is only a tech note or report). I found that the discussion of the spectrometers to be very confusing ! Only after reading through the L1 and L2 ATBDs, it became clear that the Rayleigh A and B signals are the result of integrating the images projected on CCD detectors. So for the Rayleigh channels, for each time-height bin two spectrally integrated measurements are available. This should be explained here.
For the Mie channel I found the presentation here to be misleading. The text and Fig. 3 first had lead me to believe that in the case of the Mie channel, that the data yieled by the device was a spectrum such as that iluustrated in Fig.3. It took some time and iterating between the two documents, to realize that that for the Mie channel, that the curve shown in Fig. 3 corresponds only to the central position of the Fizeau wedge ! Only after reading through the L1 and L2 ATBDs I understood that there are 16 different spectral channels available. Further, the response of each channel is the result of integrating the spectrometer output image along the different columns (corresponding to wavelength shift). Since the central wavelength varies as a function of Fizeau wedge position, the measurement will consist of the INTEGRATED filter spectral response (e.g. as shown in the bottom right-panel of Fig. 3) with the center frequency shifted according to its position along the CCD rows.
I understand that the author's would likely desire to keep the explanation concise, however, the presentation here really needs to be more detailed and accurate ! It did cost me some time to understand what was being shown here and how the instrument really functioned and I am sure this would also hold true to many other readers in the general community.
Either provide more details about the MCA (even references to the appropriate sections of the publicly available ATBDs would help) or, if it is deemed not essential, to the paper just skip it.
The same general comments apply to the description of the ICA. It would be useful for the general reader if it were to be explained what advantages (or disadvantages) the SCA method have compared to the usual method of determining extinction by calculating the log-derivative of the Rayleigh ATB profile ? Off hand. I can think that the low vertical resolution bins dealt with here may be a factor. Is this correct ?
Are any multiple-scattering considerations taken into account in the retrieval. It looks like they are not. Do you expect this to have any impacts on e.g. cirrus cloud retrievals ?
The sentence is awkward: I suggest something like: For the sake of brevity, only an outline of the SCA algorithm is presented here. Only the main features of the algorithm, necessary to understand the subsequent sections, are covered." "Designed as a wind lidar, ALADIN does not have the ability to measure depolarization". This sentence(along with the text that follows it) implies that this wind lidar do not (can not?) measure depolarization. Is this true in general or only for the specific design of ALADIN ? What design constraint has lead to ALADIN not detecting the co-polar return.
Also, ALADIN transmits and recieves circularly polarized radiation NOT linearly polarized ! Page 13, Section 3.2 ------------------- The concept of the relationship between the extinction profile and the log-derivative of the Rayleigh attenuated backscatter profile is used throughout this section. From a mathematical view-point, it is certainly true that any approach to retrieving the extinction solely using the molecular backscatter profile (either explicitly or implicitly) involves computing the log derivative of the attenuated backscatter profile. This must be true also of the SCA approach briefly described in Section 2.2.1. It would be useful to guide the reader with regards to this point. For example, outlining how the SCA approach is related to the standard log-derivative method for retrieving extinction would be useful !