Impact of Vertical Profiles of Aerosols on the Photolysis Rates in the Lower Troposphere from the Synergy of Photometer and Ceilometer Measurements in Raciborz, Poland, for the Period 2015–2020
Round 1
Reviewer 1 Report
This manuscript investigates the impacts of vertical distribution of aerosols which is retrieved from combination of ceilometer and sun-photometer on photolysis rate in lower atmosphere. The topic is interesting and suit for the remote sensing journal. There are some questions need to be resolved before it can be published.
- Are the distributions for j values in Fig.5 typical in all cases? Should give some analysis about what causes them.
- Table 2 only compares statistical characteristics using GRASP and REFERENCE input setup in the TUV model. Why not use extremely different input values from GRASP/MERRA2 and benchmark values from STANDARD?
- Statistics of j values seems depending on altitude rather than TUV input options in Table 2, while the relative differences obtained for various TUV input configurations are significant in Table 3. How to explain the differences and relations between Table 2 and Table 3?
- Are the differences acceptable between CIMEL observations and columnar values calculated by vertical AOC profiles? Are MERRA-2 profiles for SSA and ÅE checked with columnar SSA and ÅE?
- In Table 4-5, there is complete STANDARD input while it lacks GRASP or GRASP/MERRA2 input. The values of vertical profiles of SSA and ÅE are not shown, so comparison of different cases and explanation of the impact might be unclear and confused.
- The extreme negative cases seem to appear in spring and summer, while extreme positive cases usually appear in autumn and winter. Does it suggest seasonal effects on the results?
- Please be careful to check for text errors and revise the summary.
Author Response
Dear Reviewer,
thank You for your insightful comments. Below we provide our responses:
“Are the distributions for j values in Fig.5 typical in all cases? Should give some analysis about what causes them.”
Old Figure 5 (and Fig.6), showing the histograms, were deleted and replaced by new Figure 6, which provides changes of the photolysis rate for all considered levels (surface, 0.5 km and 2 km above). Our results are for the period 2015-2020, i.e. the period is too short to discuss if the distribution shown in new Fig.6 represent typical (climatological) variability over the site. New Fig.7 is added to illustrate possible sources of the photolysis rate variability (changes of aerosol optical depth, vertical profiling of extinction coefficient, single scattering albedo, and Angstrom exponent).
“Table 2 only compares statistical characteristics using GRASP and REFERENCE input setup in the TUV model. Why not use extremely different input values from GRASP/MERRA2 and benchmark values from STANDARD?”
We select REFERENCE as the extreme input because all aerosols characteristics are fixed to the long-term means for the period 2015-2020 from the CIMEL observations. STANDARD is not extreme input as it contains “real” measured column values of the aerosol characteristics. We decide to use GRASP input to compare with REFERENCE, not GRASP/MERRA because GRASP input consists of the values adjusted to the measured (by CIMEL) column values. In case the GRASP/MERRA input, SSA and AE come from the reanalysis not corresponding to the long-term columnar SSA and AE means (from CIMEL) used in REFERENCE input. The additional effect of the SSA and AE profiling is further discussed in the text.
“ Statistics of j values seems depending on altitude rather than TUV input options in Table 2, while the relative differences obtained for various TUV input configurations are significant in Table 3. How to explain the differences and relations between Table 2 and Table 3?”
Table 2 shows statistics of j values in absolute unit s-1. Based on Table 2 results and new Fig.2 we can guess that altitude is the basic driver for j variability. However, when we compare the relative differences between j values (in percent, Table 3) for moments of the CIMEL observations (during perfect weather conditions) we can reveal sources of j variability at each considered level (see Fig.7).
“ Are the differences acceptable between CIMEL observations and columnar values calculated by vertical AOC profiles? Are MERRA-2 profiles for SSA and ÅE checked with columnar SSA and ÅE?”
The vertical extinction profile is adjusted to the observed aerosol optical depth (vertical integral of the extinction coefficients). MERRA-2 profiles for SSA and AE are not checked with columnar SSA and AE as the vertical integral of SSA and AE do not correspond with the columnar values measured by the CIMEL. This is also discussed in the revised manuscript:
“MERRA-2 profile for SSA and ÅE are not normalized by the corresponding observed values. GRASP/MERRA results should be treated with caution, but we decide to use this option to estimate extreme j effects caused by the vertical variations of SSA and ÅE, even if they do not correspond to the column values.” L.259-261
“In Table 4-5, there is complete STANDARD input while it lacks GRASP or GRASP/MERRA2 input. The values of vertical profiles of SSA and ÅE are not shown, so comparison of different cases and explanation of the impact might be unclear and confused.”
The result in Tables 4 and 5 are only to discuss if it is possible to guess the occurrence of the extreme in the relative differences if we know details of characteristics of the aerosols. In practice, columnar values of aerosol properties (from CIMEL observations) are only available for the site (i.e. STANDARD input). In addition, GRASP provides a vertical profile of extinction coefficient but it is constrained by the CIMEL measured aerosol optical depth (AOD). Therefore only AOD, SSA, AF, AE values are in Tables 4 and 5. GRASP/MERRA values of SSA and AE cannot be characterized by pertaining columnar values of SSA and AE, as a simple vertical integral (e.g. without weighting to account for the aerosol density profile) of SSA or AE does not correspond to the columnar values measured by the CIMEL sunphotometer.
The extreme negative cases seem to appear in spring and summer, while extreme positive cases usually appear in autumn and winter. Does it suggest seasonal effects on the results?
Maybe the seasonal effect exists but we do not have enough data (only 6 years) to discuss the problem
Please be careful to check for text errors and revise the summary.
Yes, we did our best to improve the text errors and the summary section has been rewritten.
Reviewer 2 Report
See attached
Comments for author File: 
Comments.pdf
Author Response
Dear Reviewer,
thank You for your insightful comments. Below we provide our responses:
“- It is not so clear to this reviewer how the aerosol vertical distribution significantly affects the photolysis rate through statistical analysis here.”
We clarify this point. New Figure 7 has been added in the revised manuscript and in Summary and Conclusion section we state:
“Comparison of the output of the STANDARD model with the model accounting for vertical changes in AOC (GRASP and GRASP/MERRA) allows to estimate effects caused by vertical changes in AOC. Above the surface, the boxes and whiskers, shown in Fig. 7 for the GRASP versus STANDARD and GRASP/MERRA versus STANDARD input pair, are comparable with those obtained for REFERENCE versus STANDARD (the benchmark set) input. This illustrates that the vertical profiling of AOC causes variations in j values on a scale similar to that due to columnar AOC changes.” L. 426-432
“- It would be very useful to show case studies showing the model output results with different model input settings compared with real measurements of the photolysis frequency.”
Results for the case studies are in new Figure 5.
In the revised manuscript, we explain why the comparison with measured photolysis frequencies was not possible
“Here we present only the results based on model simulations. The ground-level measurements of the actinic flux were not available at the site. Previous studies showed that the TUV model is able to simulate j (O1D) and j (NO2) at the ground level (see for example the comparison in Fig.5 by Wang et al. [7]).” L.391-394.
Minor details:
Line 41-42: Please provide reference for this formula. Done
Line 49: Should be “j(NO2)” same for line 279. Done
Line 55: I suggest changing “in the troposphere was ...” to “in the troposphere is ...” Here
using the present tense is more appropriate. There are other places throughout the article where
using present tense is better (like Line 87). Done
Line 85: Change “provides” to “indicates”. Done
Line 117: Change “than” to “then”. Done
Line 139-140: Please change the order of figure 1 (a) and (b), otherwise it is a little bit
confusing when you mention “A stronger signal originating from aerosol layering was observed
in the former case.”
Order of Figure 1 a and 1b has been change and the corresponding text.
Line 146: Delete “number of ”, change “size distribution and spectral..” to “size distribution,
spectral”. Done
Line 158: Change “and” to “,”. Done
Line 160: Change to “...allow retrieval of vertical.”. Done
Line 162: Delete “its”. Done
Line 165-166: Change to “.examples of the GRASP É‘-profile and the Eltermans’s É‘-profile..”.Done
Line 170-172: What is the significance of using log scale for the y-axis (Altitude) in figure 2? It
is customary to use linear scale for altitude.
In the revised manuscript, linear scale for the y-axis was in all Figures.
Line 255: Change to “a wide range, implying that...”. Done
Line 256: Change to “for various weather conditions at the site”. Done
Line 279: Should be “j(NO2)”. Done
Line 289: Change to “Standard deviations also increase significantly with height...”. Done
Line 321:Change to “CIMEL”. Done
Line 365: Please expand on “for the cases with the extreme columnar AOC it is difficult to
predict the emergence of the extreme differences. ”
The old text was deleted. In the revised manuscript, we explain
“Cases with extreme columnar AOC values are not always associated with extreme variations in j values due to the vertical AOC profiling. Therefore, it is impossible to predict the extreme influence of the AOC profile on j values based on the specific configuration of the columnar AOC properties, solar elevation, total column ozone, and temperature” l. 372-376.
Line 298, 349, 376,379: I recommend changing these two tables to figures for more clarity, for
example, using boxplot to describe the output data distributions.
Box plots (Fig.6 and Fig.7) have been added to present the output data. However, the tables remain because they contain also the results not shown in these Figures.
Line 408-416: it does not appear that including the “real” α-profile resulted in significant changes comparing to the REFERENCE benchmark value, except there is only a few percent difference for the standard deviation, which is quite obvious, because for the REFERENCE the authors used a fixed Elterman’s α-profile and for the last 2 settings they used “real” α-profiles from ceilmeters and GRASP which varies for different days and cases.
New statement has been added (Section 4. Summary and Conclusions “) showing importance of “real” α-profile: “The corresponding ranges between the 2.5th and 97.5th percentile are 13.9% and 7%, respectively. This gives almost two times larger span for the cases using “real” α-profile. Such the increase was also found for j (NO2).” l.439-442. See also the answer to the first general problem mentioned at the beginning of the review no.2 .
This does not support the statement in the introduction (line 66-72), that the aerosols can significantly affect the photolysis rates ~10% reduction over the Europe region.
This problem has been explained in the revised manuscript. The following statement has been added:
“At the ground level for j (O1D), standard deviation, 2.5th and 97.5th percentile are 6.7%, -10%, and 12%, respectively (Table 3). The corresponding values for j (NO2) are 7.8%, -14.3%, and 16.6%, respectively. These estimates agree with ~10% effect of the atmospheric aerosols on the photolysis rates, which was previously mentioned for Europe [6, 21].” l.416-420
It would be more convincing if you could present some special cases, for example a polluted day and a clear sky day and calculate their photolysis rates with different settings and compare with the observations and give a more in-depth discussion of these examples.
New Figure 5 has been added showing vertical profiles (0-14 km) of j values for the days with weak and strong aerosols layering to illustrate the performance of various versions of the TUV model. These cases are discussed throughout Section 2.
Reviewer 3 Report
The paper is of interest because it shows that information on the vertical distribution of aerosols is useful for the photolysis rates in the lower troposphere.　However, it difficult to understand the contents because there was not enough explanation on how to present the results and how to reach the conclusion.　 Therefore, it is recommended that the results be presented more clearly in Chapter 3. Such a revision would make the paper more understandable.
Author Response
Dear Reviewer,
thank You for your reviewing our contribution. Following your suggestions, we did our best to improve the manuscript. New Figures (5, 6, 7) are added to illustrate the model performance and to corroborate the statement that vertical profiles of the aerosol characteristics strongly affect the photolysis rates. Substantial changes have been made in Sections 3 and 4.
Round 2
Reviewer 3 Report
The manuscript has been revised well.
