the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 13 Oct 2022
The recovery and re-calibration of a 13-month aerosol extinction profiles dataset from searchlight observations from New Mexico, after the 1963 Agung eruption
Abstract. We report the recovery and re-calibration of an extensive dataset of vertical profile measurements of the 1963/64 stratospheric aerosol layer measured from a two-site searchlight measurement facility at White Sands missile base and Sacramento Peak observatory, in New Mexico, US. The recovered dataset comprises 105 profiles of 550 nm aerosol extinction (βp (z)) and is part of a longer program of measurements with the US Air Force Cambridge Research Laboratories (AFCRL) searchlight facility that began in February 1963. The recovered series of βp (z) profiles span the 13-month period December 1963 to December 1964 and provide a unique record of the altitude and vertical extent of the Northern Hemisphere dispersed portion of the aerosol cloud from the March 1963 Agung volcanic eruption. The data recovery methodology involved first re-digitizing the 105 original βp (z) profiles (βpOrig (z)) from individual Figures within an AFCRL research report (Elterman, 1966a). The re-calibration involves inverting the original equation used to compute βp (z) in Elterman (1966a; 1966b) to retrieve a normalized detector response ((Erp (z))/(Erp (35))) profile for each of the 105 re-digitized βpOrig (z) profiles. An iterative procedure was then used to compute the re-calibration βpOrig (z) profiles (βpRecal (z)), with the molecular extinction profile calculated with the corresponding daily molecular extinction profile calculated from local soundings, rather than the US Standard Atmosphere 1962 in the original dataset. Two-way molecular and aerosol transmittance corrections are applied using the MODTRAN (MODerate resolution atmospheric TRANsmission) code in transmission mode, applying a best-estimate aerosol phase function calculated from measurements, applied for the entire 2.76 to 35.2 km column. For the tropospheric aerosol transmittance, the AERONET aerosol phase function from White Sands High Energy Laser Systems Test Facility (HELSTF) was applied (2.76 to 10.7 km), a separate stratospheric phase function applied between 11.2 and 35.2 km, calculated from a set of particle size distributions measured by the U-2 high-altitude aircraft over a US region in the vicinity of White Sands in early 1964. Errors were estimated taking as a reference the errors determined in the computation of βpOrig (z). Using available tabulated data from the original procedure the errors in the re-digitalization of βpOrig (z) and in the retrieval of the (Erp (z))/(Erp (35)) procedures were calculated and later added to the original estimates. Both the βpRecal (z) and the stratospheric aerosol optical depth magnitudes showed higher magnitudes than βpOrig (z) and the original stratospheric aerosol optical depth, however their magnitudes show a reasonable agreement with other contemporary observations. Both the original and re-calibration datasets are being submitted to PANGAEA data repository for its storage and public access.
- Preprint
(1397 KB) - Metadata XML
-
Supplement
(443 KB) - BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on essd-2022-272', Anonymous Referee #1, 14 Nov 2022
This study reported the recovery and re-calibration of an extensive dataset of vertical profile measurements of the 1963~1964 stratospheric aerosol layer measured from a two-site searchlight measurement facility at White Sands missile base and Sacramento Peak observatory, in New Mexico, US. This work has scientific significance for the observation of stratospheric aerosol increase due to historical volcanic eruptions. However, the authors heavily cite figures and tables from previous studies and do not have a comprehensive flow chart of this study, and the manuscript is not written with sufficient standardization, making it difficult to read. In addition, calibration or re-calibration is often done with reference to higher standards and is a bottom-up effort to seek higher accuracy. This is also the guarantee of traceability of measurement results. Only in this way can the results after calibration or re-calibration be more reliable and comparable. If more uncertainty is introduced into the re-calibration process, the results are hard to convince.
There are some detail comments as followed:
- According to the introduction, the detector response and aerosol extinction profile are given in the original literature table, why do the authors need to invert the detector response using the aerosol extinction profile and then compare the accuracy of the detector response with the original table?
- I can understand that many parameters need to be reacquired when recovering historical data, however, the use of these data in section 4.4 requires more rigorous argumentation and validation to prove that they are plausible and do not introduce too much error into the re-calibration data. The authors have not argued enough in this regard and suggest a fuller justification of the uncertainty analysis of the data used and its impact on the results, e.g., scattering phase functions, etc.
- For the aerosol extinction profile, did the authors re-digitize the figure data from the original literature in order to obtain observations for each layer and thus use Fernald's algorithm? However, it is possible that the large uncertainty of the data in the near-surface layer caused a large error of the re-calibrated results from the literature digitization results (Fig. 6). In this regard, the authors should give other supporting information to show that the reader can trust the rescaled data.
- From the AOD assessment in Figure 7, the recalibration data only systematically increased the value of AOD, while the correction of the overall trend was more problematic. Why not use AOD as a constrain to retrieve the atmospheric column aerosol extinction when simultaneous AOD data are available?
- There are also some puzzling descriptions in the text, such as
- Line 268-269: “…… an order of magnitude lower than the values in figure 1”, but the figure 1 is the “searchlight scenes geometry”.
- Line 323: “…… in the digitization procedure (see section 2.3 above)”, but section 2.3 is not included in this manuscript.
- Line 638: “…… from one of the major volcanic eruption of the XX century.”, what does this XX mean? I did not list all of these problems. Please revise them carefully.
Citation: https://doi.org/10.5194/essd-2022-272-RC1 - AC1: 'Reply on RC1', Juan Carlos Antuna-Marrero, 05 Mar 2023
-
RC2: 'Comment on essd-2022-272', Anonymous Referee #2, 30 Nov 2022
This study shows the recovery and recalibration of an extensive dataset of stratospheric aerosol vertical profiles originally measured during 1963/64 by two searchlight measurement sites in New Mexico, US. A description of the original retrieval process and discussion of the source of its sources of uncertainty are provided. The different steps, including assumptions and additional processing of the recovery and recalibration processes are described. The goddess of the new vertical products is shown through comparison against independent stratospheric and column integrated AOD of different sites.
The complex methodological structure, with lots of references to specific tables, figures and data in the literature, makes it difficult to follow some parts of the manuscript. Inputs, outputs and assumptions have to be clearly indicated in the description of each step of the recalibration, recovery or retrieval.
Use of English needs to be deeply revised. Lots of typos and poorly constructed sentences can be found along the text, making some paragraphs hardly understandable.
Some specific comments can be found below:
- section 2 needs to be revisited, it is not clear for me what is being described, the original method of the measurements or the new methodology applied after re-digitalization or both are the same. A better explanation is required of how the iterative process is made. Please, state clearly which are the inputs, outputs and assumed parameters. A table or an explanatory box chart will be very useful.
- line 184: Table 1 is referred but further details are needed. Some column names are not totally clear (ToD (LST) and DD).
- line 207: Tp(z)init first mentioned here. Clarification of its meaning is necessary.
- In line 242 is stated that only aerosol profiles are going to be considered in this work. And in the same paragraph some lines after it is said that a two component atmosphere is assumed. Thus, I do not understand the need of refilling the aerosol profile with Rayleigh values. If Rayleigh is already accounted for in the molecular profile, these points corresponding to aerosol should be zero according to my understanding. Unless the authors are referring to total extinction, in this case it should be clearly indicated.
- line 366: It is not clear what is the second molecular phase function that was applied to the top part of the profile.
- line 460: It is difficult to understand what is represented in figure 4. This sentence needs to be revised.
- What is the wavelength corresponding to AERONET AOD represented in figure 7?
- What is the meaning of the whiskers in figure 7? Standard deviation?
- It is surprising the spiky behaviour of the monthly averages of AOD of the recalibrated and original datasets in the months of October, November and December. Some clarification is needed in the text.
- To my understanding the objective of section 5.5.1 is not clear. From the figure and the explanations in the text it seems that the original AOD is more adequate than the recalibrated values. If the message is that the reference stratospheric AOD and sAODorig do not retain information of stratospheric aerosol load because of different reasons, it should be clearly stated.
- line 569: “A second study reported the climatology of the mixing height (MH)” What is the reference for this study?
Citation: https://doi.org/10.5194/essd-2022-272-RC2 - AC2: 'Reply on RC2', Juan Carlos Antuna-Marrero, 05 Mar 2023
-
RC3: 'Comment on essd-2022-272', Anonymous Referee #3, 07 Dec 2022
General comments
This manuscript deals with the interesting idea of estimating vertical profiles of aerosol extinction in the troposphere and stratosphere from ground-based active remote sensing measurements with a search light facility. The measurement principle is quite interesting and unusual and the re-evaluated measurements provide information on the stratospheric aerosol loading during the declining phase after the Agung eruption in 1963. In find the basic approach of re-evaluating these older measurements very good and the initiative commendable.
I do have, however, several major concerns regarding the applied methodology:
- You determine the Tp profiles required for the re-calculation of the aerosol extinction profiles now with Modtran using climatological extinction profiles, if I understood correctly? (The descriptions are not entirely clear, e.g. the magnitudes with asterisks TR*, TO3* and Tp* appear in section 4.4.3.1 without an explicit introduction or additional explanations. I assume these are the quantities that you finally used to calculate the aerosol extinction profiles from the reconstructed normalized detector response?). The main issue here is that you made different assumptions to calculate the aerosol effect on the transmission than in the original retrieval by Elterman. Of course one would expect different results then, right? We don’t really know which assumption is better.
- Elterman used an iterative approach to determine the aerosol extinction profile, but you did not. It would be very important to estimate the impact of this difference. The iterative approach is certainly more realistic and should provide better results. Perhaps the differences between your extinction and AOD values and the Elterman values are mainly due to this difference? In my opinion this is an essential aspect that needs to be tested and can probably be tested with little effort. At the moment the paper presents two different datasets (the old one and your new one) that differ significantly and the reader cannot tell which one is better. Perhaps the Elterman values are more realistic, because they are based on the more correct iterative approach. If this is the case, then your retrievals would have little value, right? But the reader cannot judge, which dataset is better, because this important aspect has not been investigated.
- The error estimation provided is incomplete and neglects some important effects, such as the effect of uncertainties in the aerosol phase functions on the retrieved aerosol extinction profiles. You used different aerosol scattering phase functions than in the original studies and my guess is that this leads to significant differences in the obtained extinction values. The different transmission calculations and the related uncertainties will probably also be an important source of error. Right now the differences between the “old” and “new” aerosol extinction profiles are larger than the respective errors. The authors simply state that this is the case, but no explanation is given and no attempt to improve on this is made. This can be considered a major issue or weakness of the presented results and the reader does not know how to deal with it. To me the differences mean that at least one of the two retrievals does not work well or that the error estimates are incomplete or not robust.
In addition, the manuscript is in a poor general state. There are many typos, incomplete sentences and statements whose meaning is not fully clear. Several of the co-authors are native speakers and it looks like none of them has actually read the paper. That should not be the case!
I apologize for the somewhat harsh judgement, but in my opinion the manuscript requires at least a major revision to become acceptable. It may well be that the original aerosol extinction profiles are in better agreement with reality than the new ones, because of the issues mentioned above and below. I would be glad to be convinced otherwise.
Specific comments
Line 91: “by modulating a searchlight beam”
How was this done? By modulating the intensity? Or by focusing/defocusing the beam?
Line 115: “within the same ESSD paper”; please cite the paper.
Line 132: “Figure 2 and 3”
In this paper or an another, older paper? Your Figures 2 and 3 show profiles for a different date.
Line 140: “referring to the component Rayleigh/molecular”
Does this mean Rayleigh scattering and molecular absorption? Please state explicitly.
Line 143: “the precursor words “molecular/Rayleigh””
Does “molecular/Rayleigh” mean:
(a) Rayleigh scattering by molecules only or
(b) Raleigh scattering and molecular absorption?
This is not clear and should be explicitly mentioned.
Lines 165 to 163: These two paragraphs were presented above more or less in identical form. This information fits better here in my opinion and I suggest reducing the text in the previous section.
Line 166: “Synchronous demodulation” ; what does this mean here?
Same line: what does “for those times” refer to? This is not clear.
Line 224: “The second is due to the convergent iteration procedure applied to solve the same equation (1), introducing new correction factors (a new profile of ðððð(ð§ð§)) in each step.”
Why is this an "issue"? This appears to be the correct way to treat this problem?
Please explain, what the problem is here. I guess the issue is that you do not carry out this iteration? This should then be stated explicitly.
Lines 226 and following: Perhaps I'm missing a point, but if the final T_p(z) is not known, then the Erp ratio cannot be calculated by inverting equation (1), right? Are the “final” T_p profiles also available?
Line 231: “The iteration convergent procedure .. was not conducted ..”
This may be a big issue, because the extinction profiles directly depend on this, right?
The question is, how much a limitation this is. I think you should at least test, how the results after a 2nd or 3rd iteration differ from the results after the first iteration. Otherwise, differences to the original retrievals by Elterman cannot really be interpreted.
Last part of the same sentence: I don’t fully understand it, to be honest and the critical piece of information here is, whether Elterman also tabulated the “final” Tp profiles or not. Is this the case?
Line 248: “Most of the missing levels were filled in with the daily aerosol extinction profile averages, and the very few remaining were filled by linear extrapolation.”
Sorry, but this is confusing. In the sentence before you said that missing data were flagged with a value of -999.99 and now you write that averaged profiles/interpolations are used. Perhaps these steps apply to different things, but this is not well explained here?
Line 269: “in figure 1” ? Should this be “figure 2”? Figure 1 shows something else.
Line 270: “The absolute differences .. is” -> “The relative differences .. are”
Line 273: “The relative errors .. has” -> “The absolute error .. has”; “absolute” and “relative” were mixed up here and above.
Line 274: “while its mean value is lower than 1%.”
This was already mentioned two sentences above.
Equation (2): Please state, whether the “final” Tp profiles are also available from the older papers or not.
Line 291: “and the aerosol (ðð(ð§)Orig) transmissions”
Again, are those the transmission values after the final iteration?
Line 357: “Then the pressure and temperature in each sounding, from 31 to 36 km, were filled using monthly mean values from the COSPAR International Reference Atmosphere – COSPAR-86 (Fleming et al., 1990).”
If you filled in gaps in this way, does this lead to discontinuities in the T, p and hence density profiles? Or are the resulting profiles smooth?
In addition, the quality of the CIRA-86 atmosphere, particularly in the troposphere & stratosphere is questionable, I think. It would be good to provide some error estimates here.
Line 361: what does “modern conventional algorithms” mean?
Section 4.4.2.1: It would be interesting to show a sample of the tAPF and also the derived size distribution parameters.
Line 377: “Finally, the resulting tAPF at the two cited wavelengths were used to calculate the tAPF at 550 nm by interpolation.”
How was this done exactly? I'm not sure there is a straight-forward approach to do that?
Section 4.4.2.2.: It would also be interesting for the reader to show these PSDs here?
What are the size distribution parameters, or typical values of the mean/median or effective radius?
Line 400: “ðð *(ð§), and ðð3* (ð§)”
These asterisk quantities suddenly appear, but they are not introduced, I think. Please explain what they are.
Line 401: “In the case of ðð∗(ð§), the aerosol extinction profile for desert aerosols embedded in the MODTRAN-4 code was used.”
Why? this will be different from the actual Tp profile, right? If no iterations are performed this will introduce errors in the re-calculated aerosol extinction profiles, right?
Next sentence: the second “showed” should be deleted.
Line 416: “deg K”; Kelvin does not come with a degree sign.
Line 421: “The lower altitude of the monthly mean tropopause”
What does this mean, “lower altitude of the monthly mean tropopause” ?
Line 427: “the 36 daily profiles of ð½ð *(ð§) and ðð *(ð§)”
Why was TR*(z) this used for the retrieval “without” transmission correction? Does "without transmission correction" in the previous sentence only refer to the effects of the aerosols on the transmission? This is not clear.
Line 445: “Together with the 36 daily profiles ð¸ðp(ð§)/ð¸ðp(35) in equation (1) these changes produced a set of 36 daily-re-calibration aerosol extinction profiles …”
What Tp profiles were used here? This is not clear to me at all? Please explain in detail.
And what Pp profiles are used?
Line 448: “Both in the second and third steps”
What is the third step? Please mention it explicitly.
Line 453: “That precludes applying any new normalization-correction ..”
I don't understand this statement? Why would you want to apply a new normalization?
Lines 478/479: extinction values without units
Line 504: what does “mixing layer” refer to here? Is this the planetary boundary layer?
Line 504: “The agreement in winter ..” ; sentence incomplete.
Line 509: “Average mixing layer height diurnal cycle for December showed a maximum of 929 m”
What does this mean? Statement is incomplete and unclear.
Line 513: “these maximum averaged values”
Unclear what you mean by “maximum averaged”
Section 5.5.1: Please mention briefly what the Sato and Stothers values are based on, i.e. which methods are used.
Section 5.6, errors: What about the contribution of the uncertainty in the extinction profiles used to determine Tp? This uncertainly will also contribute to the overall errors. If I understand correctly you are now using a climatological extinction profiles for dust/sand from Modtran?
And what about the errors caused by uncertainties in the assumed aerosol phase function? This may also be a potentially larger source of error.
These contributions could be estimated relatively easily.
Section 6.2: You should also discuss possible reasons for the large differences between the original and the new aerosol extinction profiles. In my opinion important reasons are/may be:
(a) different extinction profiles used to calculate the Tp profiles
(b) Different phase functions
(c) missing iteration approach in case of your retrievals
Line 587: “which is significantly higher than the estimated error in ð½ðð ecal between 37% and 52%.”
This means that the retrievals or the error estimates are not robust, right?
Line 597: “Photometric measurements on twilight photographs”
What does this mean?
Line 602: “The mean of the monthly means, in table 2”
The monthly mean results are shown in table 3, not in table 2.
Line 644: “That procedure ..”
Which one? The original one by Elternman or the one used here?
Same sentence: I'm sorry, but I don't get the meaning of this sentence: "introduced changes that were not possible to be accounted for ..." ?? Did you introduce these changes or not?
Typos, grammar etc.
The paper contains many typos and inconsistencies. In the following I’m only mentioning some of them. Please read the paper carefully.
Titel: “profiles dataset” -> “profile dataset”
Introduction, line 1: “large-magnitude explosive volcanic aerosol physico-chemical processes”
Does this phrase make sense? “Large magnitude” should refer to volcanic eruptions, not to “processes”, right?
Introduction, first sentence: the sentence is incomplete
Line 63: “Of these eruptions ..” ; Sentence is incomplete
Line 68: “to the SAGE-II” -> “due to the SAGE-II”
Line 95: “Measured brightness .. were” -> “Measured brightnesses .. were”
Line 108: “be rescued and re-calibration” -> “be rescued and re-calibrated”
Overall many cases were “re-calibration” should be “re-calibrated”. Please search the entire document.
Line 108: the last part of this sentence is incomplete.
Line 112: “re-calibration” -> “were re-calibrated”
Line 116: “the two datasets both published” -> “and both datasets were published”
Line 122: “complimenting” -> “complementing”
Line 124: “.. are derived” -> “.. is derived”
Line 140: “used when” -> “was used when”
Line 161: “who’s” -> “whose”
Line 179: “its original processing” -> “their original processing”
Line 187: Sorry, but what does "textually" mean here? Do you mean verbatim or a word-by-word citation?
Line 241 and several other lines: “There-digitalization” -> “The re-digitalization”; please search the entire document.
Line 254: “the the”
Line 256: “two three” ?
Line 265: “Their values are ..” ; sentence is incomplete
Line 268: “an order of” -> “, i.e. an order of”
Line 295: “on table” -> “in table”
Line 305: “shown on” -> “shown in”
Line 310: “when it addresses ..”; statement unclear; “when” doesn’t appear to fit here.
Line 321: “with also” ?
Line 411: “Then each of the 36 profiles .. were -> was”
Line 439: “Next” -> “Next,”
Line 470: “Looking at the lower levels ..” ; Sentence is incomplete and does not make sense.
Line 479: “maximums” -> maxima” (this occurs several times; please check entire paper).
Line 530: “respect to” -> “with respect to” (also several cases throughout the paper)
Line 554: “It includes” -> “They include” (referring to “considerations”)
Line 637: “conducted byre-digitalization”
Line 638: “has been” -> “have been”
Same sentence: Please check this sentence again carefully. The wording is not very precise or at least misleading.
Citation: https://doi.org/10.5194/essd-2022-272-RC3 - AC3: 'Reply on RC3', Juan Carlos Antuna-Marrero, 05 Mar 2023
Status: closed
-
RC1: 'Comment on essd-2022-272', Anonymous Referee #1, 14 Nov 2022
This study reported the recovery and re-calibration of an extensive dataset of vertical profile measurements of the 1963~1964 stratospheric aerosol layer measured from a two-site searchlight measurement facility at White Sands missile base and Sacramento Peak observatory, in New Mexico, US. This work has scientific significance for the observation of stratospheric aerosol increase due to historical volcanic eruptions. However, the authors heavily cite figures and tables from previous studies and do not have a comprehensive flow chart of this study, and the manuscript is not written with sufficient standardization, making it difficult to read. In addition, calibration or re-calibration is often done with reference to higher standards and is a bottom-up effort to seek higher accuracy. This is also the guarantee of traceability of measurement results. Only in this way can the results after calibration or re-calibration be more reliable and comparable. If more uncertainty is introduced into the re-calibration process, the results are hard to convince.
There are some detail comments as followed:
- According to the introduction, the detector response and aerosol extinction profile are given in the original literature table, why do the authors need to invert the detector response using the aerosol extinction profile and then compare the accuracy of the detector response with the original table?
- I can understand that many parameters need to be reacquired when recovering historical data, however, the use of these data in section 4.4 requires more rigorous argumentation and validation to prove that they are plausible and do not introduce too much error into the re-calibration data. The authors have not argued enough in this regard and suggest a fuller justification of the uncertainty analysis of the data used and its impact on the results, e.g., scattering phase functions, etc.
- For the aerosol extinction profile, did the authors re-digitize the figure data from the original literature in order to obtain observations for each layer and thus use Fernald's algorithm? However, it is possible that the large uncertainty of the data in the near-surface layer caused a large error of the re-calibrated results from the literature digitization results (Fig. 6). In this regard, the authors should give other supporting information to show that the reader can trust the rescaled data.
- From the AOD assessment in Figure 7, the recalibration data only systematically increased the value of AOD, while the correction of the overall trend was more problematic. Why not use AOD as a constrain to retrieve the atmospheric column aerosol extinction when simultaneous AOD data are available?
- There are also some puzzling descriptions in the text, such as
- Line 268-269: “…… an order of magnitude lower than the values in figure 1”, but the figure 1 is the “searchlight scenes geometry”.
- Line 323: “…… in the digitization procedure (see section 2.3 above)”, but section 2.3 is not included in this manuscript.
- Line 638: “…… from one of the major volcanic eruption of the XX century.”, what does this XX mean? I did not list all of these problems. Please revise them carefully.
Citation: https://doi.org/10.5194/essd-2022-272-RC1 - AC1: 'Reply on RC1', Juan Carlos Antuna-Marrero, 05 Mar 2023
-
RC2: 'Comment on essd-2022-272', Anonymous Referee #2, 30 Nov 2022
This study shows the recovery and recalibration of an extensive dataset of stratospheric aerosol vertical profiles originally measured during 1963/64 by two searchlight measurement sites in New Mexico, US. A description of the original retrieval process and discussion of the source of its sources of uncertainty are provided. The different steps, including assumptions and additional processing of the recovery and recalibration processes are described. The goddess of the new vertical products is shown through comparison against independent stratospheric and column integrated AOD of different sites.
The complex methodological structure, with lots of references to specific tables, figures and data in the literature, makes it difficult to follow some parts of the manuscript. Inputs, outputs and assumptions have to be clearly indicated in the description of each step of the recalibration, recovery or retrieval.
Use of English needs to be deeply revised. Lots of typos and poorly constructed sentences can be found along the text, making some paragraphs hardly understandable.
Some specific comments can be found below:
- section 2 needs to be revisited, it is not clear for me what is being described, the original method of the measurements or the new methodology applied after re-digitalization or both are the same. A better explanation is required of how the iterative process is made. Please, state clearly which are the inputs, outputs and assumed parameters. A table or an explanatory box chart will be very useful.
- line 184: Table 1 is referred but further details are needed. Some column names are not totally clear (ToD (LST) and DD).
- line 207: Tp(z)init first mentioned here. Clarification of its meaning is necessary.
- In line 242 is stated that only aerosol profiles are going to be considered in this work. And in the same paragraph some lines after it is said that a two component atmosphere is assumed. Thus, I do not understand the need of refilling the aerosol profile with Rayleigh values. If Rayleigh is already accounted for in the molecular profile, these points corresponding to aerosol should be zero according to my understanding. Unless the authors are referring to total extinction, in this case it should be clearly indicated.
- line 366: It is not clear what is the second molecular phase function that was applied to the top part of the profile.
- line 460: It is difficult to understand what is represented in figure 4. This sentence needs to be revised.
- What is the wavelength corresponding to AERONET AOD represented in figure 7?
- What is the meaning of the whiskers in figure 7? Standard deviation?
- It is surprising the spiky behaviour of the monthly averages of AOD of the recalibrated and original datasets in the months of October, November and December. Some clarification is needed in the text.
- To my understanding the objective of section 5.5.1 is not clear. From the figure and the explanations in the text it seems that the original AOD is more adequate than the recalibrated values. If the message is that the reference stratospheric AOD and sAODorig do not retain information of stratospheric aerosol load because of different reasons, it should be clearly stated.
- line 569: “A second study reported the climatology of the mixing height (MH)” What is the reference for this study?
Citation: https://doi.org/10.5194/essd-2022-272-RC2 - AC2: 'Reply on RC2', Juan Carlos Antuna-Marrero, 05 Mar 2023
-
RC3: 'Comment on essd-2022-272', Anonymous Referee #3, 07 Dec 2022
General comments
This manuscript deals with the interesting idea of estimating vertical profiles of aerosol extinction in the troposphere and stratosphere from ground-based active remote sensing measurements with a search light facility. The measurement principle is quite interesting and unusual and the re-evaluated measurements provide information on the stratospheric aerosol loading during the declining phase after the Agung eruption in 1963. In find the basic approach of re-evaluating these older measurements very good and the initiative commendable.
I do have, however, several major concerns regarding the applied methodology:
- You determine the Tp profiles required for the re-calculation of the aerosol extinction profiles now with Modtran using climatological extinction profiles, if I understood correctly? (The descriptions are not entirely clear, e.g. the magnitudes with asterisks TR*, TO3* and Tp* appear in section 4.4.3.1 without an explicit introduction or additional explanations. I assume these are the quantities that you finally used to calculate the aerosol extinction profiles from the reconstructed normalized detector response?). The main issue here is that you made different assumptions to calculate the aerosol effect on the transmission than in the original retrieval by Elterman. Of course one would expect different results then, right? We don’t really know which assumption is better.
- Elterman used an iterative approach to determine the aerosol extinction profile, but you did not. It would be very important to estimate the impact of this difference. The iterative approach is certainly more realistic and should provide better results. Perhaps the differences between your extinction and AOD values and the Elterman values are mainly due to this difference? In my opinion this is an essential aspect that needs to be tested and can probably be tested with little effort. At the moment the paper presents two different datasets (the old one and your new one) that differ significantly and the reader cannot tell which one is better. Perhaps the Elterman values are more realistic, because they are based on the more correct iterative approach. If this is the case, then your retrievals would have little value, right? But the reader cannot judge, which dataset is better, because this important aspect has not been investigated.
- The error estimation provided is incomplete and neglects some important effects, such as the effect of uncertainties in the aerosol phase functions on the retrieved aerosol extinction profiles. You used different aerosol scattering phase functions than in the original studies and my guess is that this leads to significant differences in the obtained extinction values. The different transmission calculations and the related uncertainties will probably also be an important source of error. Right now the differences between the “old” and “new” aerosol extinction profiles are larger than the respective errors. The authors simply state that this is the case, but no explanation is given and no attempt to improve on this is made. This can be considered a major issue or weakness of the presented results and the reader does not know how to deal with it. To me the differences mean that at least one of the two retrievals does not work well or that the error estimates are incomplete or not robust.
In addition, the manuscript is in a poor general state. There are many typos, incomplete sentences and statements whose meaning is not fully clear. Several of the co-authors are native speakers and it looks like none of them has actually read the paper. That should not be the case!
I apologize for the somewhat harsh judgement, but in my opinion the manuscript requires at least a major revision to become acceptable. It may well be that the original aerosol extinction profiles are in better agreement with reality than the new ones, because of the issues mentioned above and below. I would be glad to be convinced otherwise.
Specific comments
Line 91: “by modulating a searchlight beam”
How was this done? By modulating the intensity? Or by focusing/defocusing the beam?
Line 115: “within the same ESSD paper”; please cite the paper.
Line 132: “Figure 2 and 3”
In this paper or an another, older paper? Your Figures 2 and 3 show profiles for a different date.
Line 140: “referring to the component Rayleigh/molecular”
Does this mean Rayleigh scattering and molecular absorption? Please state explicitly.
Line 143: “the precursor words “molecular/Rayleigh””
Does “molecular/Rayleigh” mean:
(a) Rayleigh scattering by molecules only or
(b) Raleigh scattering and molecular absorption?
This is not clear and should be explicitly mentioned.
Lines 165 to 163: These two paragraphs were presented above more or less in identical form. This information fits better here in my opinion and I suggest reducing the text in the previous section.
Line 166: “Synchronous demodulation” ; what does this mean here?
Same line: what does “for those times” refer to? This is not clear.
Line 224: “The second is due to the convergent iteration procedure applied to solve the same equation (1), introducing new correction factors (a new profile of ðððð(ð§ð§)) in each step.”
Why is this an "issue"? This appears to be the correct way to treat this problem?
Please explain, what the problem is here. I guess the issue is that you do not carry out this iteration? This should then be stated explicitly.
Lines 226 and following: Perhaps I'm missing a point, but if the final T_p(z) is not known, then the Erp ratio cannot be calculated by inverting equation (1), right? Are the “final” T_p profiles also available?
Line 231: “The iteration convergent procedure .. was not conducted ..”
This may be a big issue, because the extinction profiles directly depend on this, right?
The question is, how much a limitation this is. I think you should at least test, how the results after a 2nd or 3rd iteration differ from the results after the first iteration. Otherwise, differences to the original retrievals by Elterman cannot really be interpreted.
Last part of the same sentence: I don’t fully understand it, to be honest and the critical piece of information here is, whether Elterman also tabulated the “final” Tp profiles or not. Is this the case?
Line 248: “Most of the missing levels were filled in with the daily aerosol extinction profile averages, and the very few remaining were filled by linear extrapolation.”
Sorry, but this is confusing. In the sentence before you said that missing data were flagged with a value of -999.99 and now you write that averaged profiles/interpolations are used. Perhaps these steps apply to different things, but this is not well explained here?
Line 269: “in figure 1” ? Should this be “figure 2”? Figure 1 shows something else.
Line 270: “The absolute differences .. is” -> “The relative differences .. are”
Line 273: “The relative errors .. has” -> “The absolute error .. has”; “absolute” and “relative” were mixed up here and above.
Line 274: “while its mean value is lower than 1%.”
This was already mentioned two sentences above.
Equation (2): Please state, whether the “final” Tp profiles are also available from the older papers or not.
Line 291: “and the aerosol (ðð(ð§)Orig) transmissions”
Again, are those the transmission values after the final iteration?
Line 357: “Then the pressure and temperature in each sounding, from 31 to 36 km, were filled using monthly mean values from the COSPAR International Reference Atmosphere – COSPAR-86 (Fleming et al., 1990).”
If you filled in gaps in this way, does this lead to discontinuities in the T, p and hence density profiles? Or are the resulting profiles smooth?
In addition, the quality of the CIRA-86 atmosphere, particularly in the troposphere & stratosphere is questionable, I think. It would be good to provide some error estimates here.
Line 361: what does “modern conventional algorithms” mean?
Section 4.4.2.1: It would be interesting to show a sample of the tAPF and also the derived size distribution parameters.
Line 377: “Finally, the resulting tAPF at the two cited wavelengths were used to calculate the tAPF at 550 nm by interpolation.”
How was this done exactly? I'm not sure there is a straight-forward approach to do that?
Section 4.4.2.2.: It would also be interesting for the reader to show these PSDs here?
What are the size distribution parameters, or typical values of the mean/median or effective radius?
Line 400: “ðð *(ð§), and ðð3* (ð§)”
These asterisk quantities suddenly appear, but they are not introduced, I think. Please explain what they are.
Line 401: “In the case of ðð∗(ð§), the aerosol extinction profile for desert aerosols embedded in the MODTRAN-4 code was used.”
Why? this will be different from the actual Tp profile, right? If no iterations are performed this will introduce errors in the re-calculated aerosol extinction profiles, right?
Next sentence: the second “showed” should be deleted.
Line 416: “deg K”; Kelvin does not come with a degree sign.
Line 421: “The lower altitude of the monthly mean tropopause”
What does this mean, “lower altitude of the monthly mean tropopause” ?
Line 427: “the 36 daily profiles of ð½ð *(ð§) and ðð *(ð§)”
Why was TR*(z) this used for the retrieval “without” transmission correction? Does "without transmission correction" in the previous sentence only refer to the effects of the aerosols on the transmission? This is not clear.
Line 445: “Together with the 36 daily profiles ð¸ðp(ð§)/ð¸ðp(35) in equation (1) these changes produced a set of 36 daily-re-calibration aerosol extinction profiles …”
What Tp profiles were used here? This is not clear to me at all? Please explain in detail.
And what Pp profiles are used?
Line 448: “Both in the second and third steps”
What is the third step? Please mention it explicitly.
Line 453: “That precludes applying any new normalization-correction ..”
I don't understand this statement? Why would you want to apply a new normalization?
Lines 478/479: extinction values without units
Line 504: what does “mixing layer” refer to here? Is this the planetary boundary layer?
Line 504: “The agreement in winter ..” ; sentence incomplete.
Line 509: “Average mixing layer height diurnal cycle for December showed a maximum of 929 m”
What does this mean? Statement is incomplete and unclear.
Line 513: “these maximum averaged values”
Unclear what you mean by “maximum averaged”
Section 5.5.1: Please mention briefly what the Sato and Stothers values are based on, i.e. which methods are used.
Section 5.6, errors: What about the contribution of the uncertainty in the extinction profiles used to determine Tp? This uncertainly will also contribute to the overall errors. If I understand correctly you are now using a climatological extinction profiles for dust/sand from Modtran?
And what about the errors caused by uncertainties in the assumed aerosol phase function? This may also be a potentially larger source of error.
These contributions could be estimated relatively easily.
Section 6.2: You should also discuss possible reasons for the large differences between the original and the new aerosol extinction profiles. In my opinion important reasons are/may be:
(a) different extinction profiles used to calculate the Tp profiles
(b) Different phase functions
(c) missing iteration approach in case of your retrievals
Line 587: “which is significantly higher than the estimated error in ð½ðð ecal between 37% and 52%.”
This means that the retrievals or the error estimates are not robust, right?
Line 597: “Photometric measurements on twilight photographs”
What does this mean?
Line 602: “The mean of the monthly means, in table 2”
The monthly mean results are shown in table 3, not in table 2.
Line 644: “That procedure ..”
Which one? The original one by Elternman or the one used here?
Same sentence: I'm sorry, but I don't get the meaning of this sentence: "introduced changes that were not possible to be accounted for ..." ?? Did you introduce these changes or not?
Typos, grammar etc.
The paper contains many typos and inconsistencies. In the following I’m only mentioning some of them. Please read the paper carefully.
Titel: “profiles dataset” -> “profile dataset”
Introduction, line 1: “large-magnitude explosive volcanic aerosol physico-chemical processes”
Does this phrase make sense? “Large magnitude” should refer to volcanic eruptions, not to “processes”, right?
Introduction, first sentence: the sentence is incomplete
Line 63: “Of these eruptions ..” ; Sentence is incomplete
Line 68: “to the SAGE-II” -> “due to the SAGE-II”
Line 95: “Measured brightness .. were” -> “Measured brightnesses .. were”
Line 108: “be rescued and re-calibration” -> “be rescued and re-calibrated”
Overall many cases were “re-calibration” should be “re-calibrated”. Please search the entire document.
Line 108: the last part of this sentence is incomplete.
Line 112: “re-calibration” -> “were re-calibrated”
Line 116: “the two datasets both published” -> “and both datasets were published”
Line 122: “complimenting” -> “complementing”
Line 124: “.. are derived” -> “.. is derived”
Line 140: “used when” -> “was used when”
Line 161: “who’s” -> “whose”
Line 179: “its original processing” -> “their original processing”
Line 187: Sorry, but what does "textually" mean here? Do you mean verbatim or a word-by-word citation?
Line 241 and several other lines: “There-digitalization” -> “The re-digitalization”; please search the entire document.
Line 254: “the the”
Line 256: “two three” ?
Line 265: “Their values are ..” ; sentence is incomplete
Line 268: “an order of” -> “, i.e. an order of”
Line 295: “on table” -> “in table”
Line 305: “shown on” -> “shown in”
Line 310: “when it addresses ..”; statement unclear; “when” doesn’t appear to fit here.
Line 321: “with also” ?
Line 411: “Then each of the 36 profiles .. were -> was”
Line 439: “Next” -> “Next,”
Line 470: “Looking at the lower levels ..” ; Sentence is incomplete and does not make sense.
Line 479: “maximums” -> maxima” (this occurs several times; please check entire paper).
Line 530: “respect to” -> “with respect to” (also several cases throughout the paper)
Line 554: “It includes” -> “They include” (referring to “considerations”)
Line 637: “conducted byre-digitalization”
Line 638: “has been” -> “have been”
Same sentence: Please check this sentence again carefully. The wording is not very precise or at least misleading.
Citation: https://doi.org/10.5194/essd-2022-272-RC3 - AC3: 'Reply on RC3', Juan Carlos Antuna-Marrero, 05 Mar 2023
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 737 | 287 | 41 | 1,065 | 81 | 34 | 28 |
- HTML: 737
- PDF: 287
- XML: 41
- Total: 1,065
- Supplement: 81
- BibTeX: 34
- EndNote: 28
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1