Evaluation of Aerosol Typing with Combination of Remote Sensing Techniques with In Situ Data during the PANACEA Campaigns in Thessaloniki Station, Greece
, Konstantinos Michailidis 1, Nikolaos Siomos 2, Anthi Chatzopoulou 1, Georgios Kouvarakis 3, Nikolaos Mihalopoulos 3, Paraskevi Tzoumaka 4, Apostolos Kelessis 4 and Dimitrios Balis 1
Round 1
Reviewer 1 Report
This study analyzes the temporal variability of PM2.5/10 and BC concentrations from the in-situ measurements and column-integrated aerosol optical properties from sunphotometer during the two PANACEA campaigns in summer and winter, and furthermore classifies aerosol types based on lidar system. The coincident observations from multi-platform instruments are valuable, especially the aerosol types identified from Lidar. However, the analyses in section 4.1 and 4.2 are too general with only simple temporal evolution, and the conclusions are not novel for the scientific community. Evaluation of aerosol type classification from lidar with other in-situ observations could be a interesting part, but looking at Fig. 11, only smoke aerosol was actually evaluated. Moreover, the selection of biomass burning episode is also not convincing (see detailed comments below). In terms of presentation quality, some of figures were not presented in a clear, concise, and well structured way. As such, I'm afraid that the manuscript cannot be accepted for publication in Remote Sensing in its current form.
Detailed comments:
Line 34: full name of BCwb
Line 82-83: how can marine particles dominate the planetary boundary layer?
Line 101 ： Whist → while / whilst ?
Line 108-110: When reading this sentence, I am confused that which observational method do the authors think is the most reliable? Seems the authors prefer lidar, but apparently in-situ is more reliable. One couldn’t ‘evaluate’ in-situ measurements with less accurate lidar retrievals.
Line 118-119: ‘ in terms of surface and column variability and layers’ → in terms of the variability of surface and column and layers.
Line 123: full name of ‘EARLINET’
Line 130: full name of ‘AEOLUS’
Line 152-153: It’s better to briefly explain the difference between Level 2.0 and 1.5. For example, automated cloud screening and quality controls, etc. Also the authors should clarify what is the fraction of Level1.5 data in total data.
Line 186: what does ‘in situ retrievals’ mean? Is it sunphotometer? If so, I don’t think sunphotometer can be termed as ‘in-situ’ retrieval, since it actually retrieves the column-integrated properties.
Line 253-254: The difference between (v) and (iv) is unclear. In climate model to chemistry transport model we classify aerosol types as sulfate, nitrate, organic/black carbon, seasalt, and dust.. What does ‘continental particles’ mean? Is it a combination of sulfate, nitrate, organic aerosols? Could the author relate the types from model and their classification a bit? Also the criteria to separate polluted and clean continent particles is confusing. Any significant difference between medium lidar ratios and medium absorption since both are ‘medium’ ?
Line 264: The authors should mention at the beginning which figure or table they are referring to.
Line 266-217: The authors only gave the ranges of PM particles here, instead, mean and standard deviation could be more useful. Moreover, I am not convinced by the argument that the higher PM10 in winter is mainly attributed to transported dust aerosols, since the difference values between PM10 and PM2.5 are very close in summer and winter. This means that the higher PM10 in winter is more contributed by the increased fine particle part while coarse particles remain almost unchanged. Again, the mean and standard deviation could be helpful in this regard.
Line 283: The WHO updated 24-h mean limits for PM2.5 and PM10 are 15 and 45 μg/m3. Also it would be great to plot these limit values in Fig. 3.
Line 295-296: To support this argument, I would recommend to mark the days affected by the dust events in Fig.4.
Line 313-314: This sentence is confusing. Where is the marked frequency? Also based on Fig. 5, since the weekends are not marked, I couldn’t see any information regarding weekly cycle.
Line 368-372/381-389: The authors discussed Ångström exponent, Lidar ratio and depolarization ratio a lot but why not show them along with AOD and FMF?
Line 426-428: As a lot of discussions are about winter, I wonder why the authors didn’t show the aerosol layer information during winter?
Table2 is nice to give an overview of the main results in this study. But as I mentioned earlier, the mean/std could be more useful to appear in the text when discussing the results instead of in the end. So I would recommend to add mean/std values in each figure, meanwhile the authors can also keep this table.
Figure 11: The colors are confusing without any legend. Please specify them explicitly.
Line 515-517: The 2nd dust cluster with high BCff concentrations is kinda weird for me. To illustrate they are indeed dust-dominated case, marking the difference of PM10-PM2.5 in the figure could be useful.
Section 4.3.2: The case in 5 Aug was identified as a biomass burning episode by the authors. In this case the THELISYS type is not smoke and also BC concentration (0.4 μg/m3) is much lower than the average level (0.8 ± 0.5 μg/m3) in summer. I doubt the plausibility of this part.
Line 595-598: I can not get the reason why the authors stated that ‘The BC concentrations are not in accordance with the thick dust layer identified by THELISYS’. Any connections or conflicts between high BC and thick dust layer?
Author Response
This study analyzes the temporal variability of PM2.5/10 and BC concentrations from the in-situ measurements and column-integrated aerosol optical properties from sunphotometer during the two PANACEA campaigns in summer and winter, and furthermore classifies aerosol types based on lidar system. The coincident observations from multi-platform instruments are valuable, especially the aerosol types identified from Lidar. However, the analyses in section 4.1 and 4.2 are too general with only simple temporal evolution, and the conclusions are not novel for the scientific community. Evaluation of aerosol type classification from lidar with other in-situ observations could be a interesting part, but looking at Fig. 11, only smoke aerosol was actually evaluated. Moreover, the selection of biomass burning episode is also not convincing (see detailed comments below). In terms of presentation quality, some of figures were not presented in a clear, concise, and well structured way. As such, I'm afraid that the manuscript cannot be accepted for publication in Remote Sensing in its current form.
We would like to thank the reviewer for his/her fruitful comments that led to the improvement of the manuscript. This is the first time that collocated in situ and remote sensing instruments are deployed in Thessaloniki in order to assess the presence of aerosols and the predominant aerosol type both situ at the surface and at elevated heights and investigate possible dependencies. The collocated in situ measurements performed on the occasion of the PANACEA campaign offered the opportunity to study the various layers of the column and especially the one close to the surface where the lidar system’s overlap is limiting the aerosol retrievals there. There are few opportunities to acquire such data sets (i.e. multiwavelength lidar measurements and in situ instrumentation) and therefore we believe that is of interest to the scientific community to what extend such measurements complement each other. In the following, answers to comments are reported just below each related comment. When needed, the part of the manuscript we modified or added to the old version is reported and figures are modified accordingly.
Detailed comments: Line 34: full name of BCwb The ΄BCwb’ term has been introduced in the revised version of the manuscript, and the introduction now reads: «separated into fossil fuel (BCff) and biomass/wood burning (BCwb) fractions».
Line 82-83: how can marine particles dominate the planetary boundary layer? Marine particles dominate the planetary boundary layer in a coastal environment i.e., however the reviewer is right that this is not the case for Thessaloniki, where usually marine particles coexist with urban particles close to the surface. Thus, the sentence has been revised in the revised version of the manuscript with: «while
marine particles are produced at the sea surface and usually exist in the lowest troposphere».
Line 101 ： Whist → while / whilst ? The sentence has been corrected in the revised version of the manuscript, which now reads: «Whilst the identification of the aerosol type in individual layers is possible with lidars».
Line 108-110: When reading this sentence, I am confused that which observational method do the authors think is the most reliable? Seems the authors prefer lidar, but apparently in-situ is more reliable. One couldn’t ‘evaluate’ in-situ measurements with less accurate lidar retrievals. We do not intent to conclude that the lidar instrument is the most reliable one, especially in the lowest troposphere (the first 500m), where for geometrical reasons, the lidar is blind and we don’t intend to use the in-situ data to evaluate the lidar retrievals. We want to emphasize the potential of the lidar multispectral information to provide accurate vertically resolved aerosol typing retrievals, especially when aerosol layers of different origin and type are observed. Moreover we want to examine the possible consistency or dependence of these retrievals with the coexisting in-situ products, which provide information in the incomplete overlap region of the lidar profile. The sentence has been modified accordingly.
Line 118-119: ‘ in terms of surface and column variability and layers’ → in terms of the variability of surface and column and layers. We thank the reviewer for this comment. The revised sentence now reads: «In Section 4 the retrievals from the two PANACEA campaigns are presented and the temporal distribution of the aerosol optical properties is discussed in terms of the variability of surface and column and layers.»
Line 123: full name of ‘EARLINET’ The sentence in the revised version of the manuscript now reads: «THELISYS is being operated for the aerosol particles detection as member station of the European Aerosol Research Lidar NETwork (EARLINET).»
Line 130: full name of ‘AEOLUS’ ESA’s Aeolus wind mission has not abbreviation.
Line 152-153: It’s better to briefly explain the difference between Level 2.0 and 1.5. For example, automated cloud screening and quality controls, etc. Also the authors should clarify what is the fraction of Level1.5 data in total data. The following sentence was added in the revised version of the manuscript «In our analysis, we used Version 3 level 2.0 data (quality-assured data), and Level 1.5 products (cloud-screened and quality controlled data) whenever the Level 2.0 were not available». Concerning the availability of the data, corresponding to the AERONET Aerosol Optical Products (v3)– Solar, Level 1.5 and Level 2.0 products provide the same
percentages over the total data). More specifically, Level 1.5/Level 2.0 is the 100% of the total data during the summer period, whilst Level1.5/Level 2.0 is the 76% of the total data, during the winter period, due to unfavorable weather conditions. On the other hand, the Aerosol Inversion (v3) differentiate, as for example during the summer period, the Level 1.5 Lidar Ratio and depolarization values are 73% of the total data, while the Level 2.0 Lidar Ratio and depolarization values are 15% of the total data. The text has been modified accordingly.
Line 186: what does ‘in situ retrievals’ mean? Is it sunphotometer? If so, I don’t think sunphotometer can be termed as ‘in-situ’ retrieval, since it actually retrieves the column-integrated properties. We thank the reviewer for this comment. The sunphotometer retrievals are not included in the in-situ retrievals. The revised sentence, now reads: «The lidar derived aerosol typing and the column-integrated properties are discussed along with the in situ retrievals; i.e. the particulate matter (PM) concentrations and the total black carbon (BC) concentrations along with the respective contribution of the fossil fuel and the biomass/wood burning. »
Line 253-254: The difference between (v) and (iv) is unclear. In climate model to chemistry transport model we classify aerosol types as sulfate, nitrate, organic/black carbon, seasalt, and dust.. What does ‘continental particles’ mean? Is it a combination of sulfate, nitrate, organic aerosols? Could the author relate the types from model and their classification a bit? Also the criteria to separate polluted and clean continent particles is confusing. Any significant difference between medium lidar ratios and medium absorption since both are ‘medium’ ? The terminology of the aerosol types we use in this paper is consistent with relevant studies (Nicolae et al., 2018, Papagiannopoulos et al., 2018), which are based on typical mixtures of components. An overview of the chemical composition of each pure aerosol type in the lidar classification schemes (e.g., NATALI, Mahalanobis classifier), that is picked up from the OPAC (Optical Properties of Aerosols and Clouds) software package (Hess et al., 1998) is presented below. The chemical composition of each aerosol type is describedbelow and refer to particle number density mixing ratios) (Nicolae et al., 2018).
Concerning the optical properties of the aerosol types, the following table summarizes the threshold values (Papagiannopoulos et al., 2018). Typically, the clean continental aerosol over Europe is a mixture of anthropogenic pollution with particles from natural sources. The clean continental type shows a low depolarizing ability with values lower than 0.07; low lidar ratio values, i.e., 20–40 sr; and relatively high Ångström exponents, i.e., 1.0–2.5. The clean continental, therefore, differentiates from the polluted continental type due to lower lidar ratio values. Below, an overview of the mean type-dependent intensive properties (S-Lidar Ratio, kβ-the backscatter related Ångström exponent and ka- extinction-related Ångström exponent) along with the standard deviation used in the Mahalanobis typing algorithm (Papagiannopoulos et al., 2018) are the following: Where CC corresponds to the Clean Continental cluster, PC to Polluted Continental, D to dust, MM to mixed marine and S to smoke cluster. The following sentence was modified in the revised version of the manuscript: (iv) small particles with low to medium lidar ratios (i.e., clean continental particles) and (v) small spherical particles, medium absorbing (i.e., Polluted continental).
References: Hess, M., Koepke, P., and Schult, I.: Optical properties of aerosols and clouds: The software package OPAC, B. Am. Meteorol. Soc., 79, 831–844, https://doi.org/10.1175/1520-0477(1998)0792.0.CO;2, 1998. Nicolae, D., Vasilescu, J., Talianu, C., Binietoglou, I., Nicolae, V., Andrei, S., and Antonescu, B.: A neural network aerosol-typing algorithm based on lidar data, Atmos. Chem. Phys., 18, 14511–14537, https://doi.org/10.5194/acp-18-14511-2018, 2018. Papagiannopoulos, N., Mona, L., Amodeo, A., D'Amico, G., GumàClaramunt, P., Pappalardo, G., Alados-Arboledas, L., Guerrero-Rascado, J. L., Amiridis, V., Kokkalis, P., Apituley, A., Baars,H., Schwarz, A., Wandinger, U., Binietoglou, I., Nicolae, D., Bortoli, D., Comerón, A., Rodríguez-Gómez, A., Sicard, M., Papayannis, A., and Wiegner, M.: An automatic observation-based aerosol typing method forEARLINET, Atmos. Chem. Phys., 18, 15879–15901, https://doi.org/10.5194/acp-18-15879-2018, 2018. Voudouri, K. A., Siomos, N., Michailidis, K., Papagiannopoulos, N., Mona, L., Cornacchia, C., Nicolae, D., and Balis, D.: Comparison of two automated aerosol typing methods and their application to an EARLINET station, Atmos. Chem. Phys., 19, 10961–10980, https://doi.org/10.5194/acp-19-10961-2019, 2019.
Line 264: The authors should mention at the beginning which figure or table they are referring to.
The reviewer is right. The sentence now reads: «The particulate matter (PM) measurements show average daily concentrations during the summer campaign much lower than those of the winter season, with extreme values observed in winter (Figure 2).»
Line 266-217: The authors only gave the ranges of PM particles here, instead, mean and standard deviation could be more useful. Moreover, I am not convinced by the argument that the higher PM10 in winter is mainly attributed to transported dust aerosols, since the difference values between PM10 and PM2.5 are very close in summer and winter. This means that the higher PM10 in winter is more contributed by the increased fine particle part while coarse particles remain almost unchanged. Again, the mean and standard deviation could be helpful in this regard. We thank the reviewer for this comment. During the winter period, a number of dust events occurred over Thessaloniki. However, it is true that the higher PM10 in winter could be more contributed by the increased fine particle part while coarse particles remain the same. The following sentence was changed in the revised version of the manuscript: «The PM2.5 daily concentrations ranged between 8 to 25 μg/m3 , with a mean value of 14 ± 5 μg/m3 during the summer campaign and between 2 and 55 μg/m3 during the winter campaign, with a mean value of 18 ± 12 μg/m3. On the other hand, the PM10 average daily concentrations ranged between 13 to 31 μg/m3 during the summer period and between 8 to 67 μg/m3 during the winter period with a mean value of 20 ± 5 μg/m3 and 28 ± 13 μg/m3, correspondingly.
Line 283: The WHO updated 24-h mean limits for PM2.5 and PM10 are 15 and 45 μg/m3. Also it would be great to plot these limit values in Fig. 3. We thank the reviewer for this correction. The sentence in the revised version of the manuscript now reads: «The levels of the particulate matter concentrations at ground level were found within the 24-h mean limits (for the majority of the days) which are set equal to 15 and 45 μg/m3, for the PM2.5 and PM10, respectively (WHO 2006).» Moreover, Fig.3 is updated with the WHO limit values.
Line 295-296: To support this argument, I would recommend to mark the days affected by the dust events in Fig.4. We thank the reviewer for this comment. Fig. 4 is updated in the revised version of the manuscript, with plotted dust events.
Line 313-314: This sentence is confusing. Where is the marked frequency? Also based on Fig. 5, since the weekends are not marked, I couldn’t see any information regarding weekly cycle. The reviewer is right. Fig. 5 is updated in the revised version of the manuscript, with the weekends marked and the word “frequency” is changed to “weekly cycle”.
Line 368-372/381-389: The authors discussed Ångström exponent, Lidar ratio and depolarization ratio a lot but why not show them along with AOD and FMF?
We thank the reviewer for this comment. The corresponding plots are added either in the revised version of the manuscript (e.g. the Ångström exponent) or as supplementary plots due to the limited number of Lidar ratio and depolarization ratio values (aerosol inversion products) during the two campaigns.
Line 426-428: As a lot of discussions are about winter, I wonder why the authors didn’t show the aerosol layer information during winter? The authors decided not to provide the aerosol layer information retrievals during the winter campaign and only provide a discussion on the results, given the low statistical significance of the lidar retrievals. During the winter period, low to mid clouds were present most of the days of the campaign and only the Planetary Boundary Height and the lowest aerosol layers could be identified. A relevant sentence has been added in the manuscript. Table2 is nice to give an overview of the main results in this study. But as I mentioned earlier, the mean/std could be more useful to appear in the text when discussing the results instead of in the end. So I would recommend to add mean/std values in each figure, meanwhile the authors can also keep this table. We thank the reviewer for this comment. We updated the figures with plotting also the mean and the standard deviation values. Figure 11: The colors are confusing without any legend. Please specify them explicitly. We thank the reviewer for this comment. We updated the figure, by adding the corresponding legend.
Line 515-517: The 2nd dust cluster with high BCff concentrations is kind a weird for me. To illustrate they are indeed dust-dominated case, marking the difference of PM10-PM2.5 in the figure could be useful. We thank the reviewer for this comment; the plot is updated in the revised version of the manuscript. The text has been modified accordingly. Section 4.3.2: The case in 5 Aug was identified as a biomass burning episode by the authors. In this case the THELISYS type is not smoke and also BC concentration (0.4 μg/m3) is much lower than the average level (0.8 ± 0.5 μg/m3) in summer. I doubt the plausibility of this part. The reviewer is right. The case is characterized as a pollution case, where biomass particles coexist with continental ones as neither of the instrumentation denote the dominance of high absorbing particles (i.e., biomass burning particles). The following sentence is corrected in the revised version of the manuscript: «A case of transported pollution that occurred on 5 August 2019 is analyzed.» Moreover, the following sentence is modified: «The 5-day backward trajectories calculated using the Hybrid SingleParticle Lagrangian Integrated Trajectory model (HYSPLIT; https://www.arl.noaa.gov/hysplit/) in conjunction with fire spots from MODIS satellite product FIRMS (Fire Information for Resource Management System; https://firms.modaps.eosdis.nasa.gov/map/) indicate the areas of the air parcels
traveling below 3.5 km before reaching the study area. Continental particles, along with biomass burning particles, were transported from central Europe in the region of Thessaloniki (Figure 13b).»
Line 595-598: I can not get the reason why the authors stated that ‘The BC concentrations are not in accordance with the thick dust layer identified by THELISYS’. Any connections or conflicts between high BC and thick dust layer? The reviewer is right. The sentence is deleted in the revised version of the manuscript and the following sentence was added: «Even if the monitored layer is identified as an elevated one in the FT, both THELISYS retrievals and PM concentration show the dominance of coarse particles.»
Author Response File: 
Author Response.pdf
Reviewer 2 Report
1. The campaigns showed different periods for PM (July 16 – Aug. 10, Jan. 2 – Feb. 29), BC (July 17 – Aug. 27, Dec. 18 – Feb. 19) and AOD (July 17 – Aug. 26, Jan. 10 – Feb. 26). Please discuss the availability of data.
2. Figure 6: The peak BC_traffic time of summer and winter was different. Was it similar to the rush hours of the two seasons?
3. Line 521: “A case of a biomass burning episode that occurred on 5 August 2019 is analyzed.” Why it was a biomass burning episode if BC_wood concentration was low? (Fig. 14)
4. Line 553: “The total BC values ranged between 0.16 and 0.73 μg/m3,” however, some of the concentration showed in Fig. 14 were higher than 0.8 μg/m3.
Author Response
We would like to thank the reviewer for his/her comments that led to the improvement of the manuscript. In the following, answers to comments are reported just below each related comment. When needed, the part of the manuscript we modified or added to the old version is reported and figures are modified accordingly.
- The campaigns showed different periods for PM (July 16 – Aug. 10, Jan. 2 – Feb. 29), BC (July 17 – Aug. 27, Dec. 18 – Feb. 19) and AOD (July 17 – Aug. 26, Jan. 10 – Feb. 26). Please discuss the availability of data.
The sunphotometer measurements were available during the summer campaign only after 18 of July, after the instrument’s calibration finished. 17 of July was also the day of installation of the aethalometer in the Laboratory, whilst PM concentrations were available after 15 of July. Concerning the winter period, all instruments were installed at the same period, however providing measurements according to the weather conditions, as rain and low level clouds are not favorable for both the sunphotometer and the lidar instrument.
The following Table has been added in the revised version of the manuscript:
|
|
Summer PANACEA campaign |
Winter PANACEA campaign |
|
Aethalometer |
July 17 – Aug. 27 |
Dec. 18 – Feb. 19 |
|
Sunphotometer |
July 18 – Aug. 26 |
Jan. 10 – Feb. 26 |
|
MP101M |
July 15 – Aug. 10 |
Jan. 01 – Feb. 29 |
|
Lidar |
Daytime and nighttime measurements, according to the weather conditions |
Daytime and nighttime measurements, according to the weather conditions |
- Figure 6: The peak BC_traffic time of summer and winter was different. Was it similar to the rush hours of the two seasons?
It is true that the peak BC_traffic is mainly affected by the rush hours of the two seasons. Similar BC diurnal cycle is reported in Liakakou 2020a Atmos.Environ.
Liakakou, E.; Stavroulas, I.; Kaskaoutis, D.G.; Grivas, G.; Paraskevopoulou, D.; Dumka, U.C.; Tsagkaraki, M.; Bougiatioti, A.; 1354 Oikonomou, K.; Sciare, J.; Gerasopoulos, E.; Mihalopoulos, N.: Long-term variability, source apportionment and spectral 1355 properties of black carbon at an urban background site in Athens, Greece, Atmospheric Environment, Volume 222, 117137, 1356 ISSN 1352-2310, 2020.
- Line 521: “A case of a biomass burning episode that occurred on 5 August 2019 is analyzed.” Why it was a biomass burning episode if BC_wood concentration was low? (Fig. 14)
The reviewer is right. The case was wrongly characterized as a biomass only from the air masses affected by the fire spots. In the revised version of the manuscript the case is characterized as a pollution case, where biomass particles coexist with continental ones. The following sentence is corrected in the revised version of the manuscript: «A case of transported pollution that occurred on 5 August 2019 is analyzed.»
Moreover, the following sentence is modified:
«The 5-day backward trajectories calculated using the Hybrid SingleParticle Lagrangian Integrated Trajectory model (HYSPLIT; https://www.arl.noaa.gov/hysplit/) in conjunction with fire spots from MODIS satellite product FIRMS (Fire Information for Resource Management System; https://firms.modaps.eosdis.nasa.gov/map/) indicate the areas of the air parcels traveling below 3.5 km before reaching the study area. Continental particles, along with biomass burning particles, were transported from central Europe in the region of Thessaloniki (Figure 13b).»
- Line 553: “The total BC values ranged between 0.16 and 0.73 μg/m3,” however, some of the concentration showed in Fig. 14 were higher than 0.8 μg/m3.
The reviewer is right. The following sentence is corrected as follows in the revised version of the manuscript: «The total BC values ranged between 0.09 and 0.91 μg/m3, through the day, with a mean value of 0.4 μg/m3.»
Author Response File: Author Response.pdf
Reviewer 3 Report
Optical characterization of aerosols in the lower atmosphere is of high importance to understand e.g. long-range transport of different aerosol types. There is no question that this is relevant and topical and the subject matter is appropriate for the journal of Remote Sensing. However, I am worry how "relevant" a local study is with one data point and whether such research will be relevant to many people. I strongly suggest authors to improve their paper upon the above comment.
Other comments:
[line 34] do not use abbreviations before it is defined (e.g. BC_wb).
[lines 75-98] I appreciate short overview provided in this paragraph.
[line 167] Are the values AAE really constant for any site? Did you validated this premise using independent source(s) of information? Otherwise, how accurate are findings and how relevant are interpretations regarding wb and ff summarized in Sec. Results?
[line 185] Figure 1 is missing; Figure 2 is not cited in the paper. I think something is corrupted with the numbering.
[line 212] What you mean under the term “intensive properties”
[line 245] “ontains”-> “contains”
[line 268] Is this what you expect here or is there a proof that the large amplitudes of PM2.5 are associated with biomass burning particles?
[line 270] the same as above, but for transported dust particles.
[Fig. 3a] What does it mean that PM10 almost perfectly mimics the trend of PM2.5? One could expect that the PM concentrations in such a case are determined by PM2.5 particles, while the number of particles larger than 2.5 microns scales up linearly with PM2.5. But what does it mean? How to interpret this result?
[Table 2] No optical properties of aerosols are provided here.
Author Response
Reviewer #3
Optical characterization of aerosols in the lower atmosphere is of high importance to understand e.g. long-range transport of different aerosol types. There is no question that this is relevant and topical and the subject matter is appropriate for the journal of Remote Sensing. However, I am worry how "relevant" a local study is with one data point and whether such research will be relevant to many people. I strongly suggest authors to improve their paper upon the above comment.
We would like to thank the reviewer for his/her fruitful comments that led to the improvement of the manuscript. This is the first time that collocated in situ and remote sensing instruments are deployed in Thessaloniki in order to assess the presence of aerosols and the predominant aerosol type both situ at the surface and at elevated heights and investigate possible dependencies. The collocated in situ measurements performed on the occasion of the PANACEA campaign offered the opportunity to study the various layers of the column and especially the one close to the surface where the lidar system’s overlap is limiting the aerosol retrievals there. It is true that the analysis presents a local study, however, there are few opportunities to acquire such data sets and the current analysis demonstrates to what extend such measurements (i.e. multiwavelength lidar measurements and in situ instrumentation) complement each other. It is not a global study in the sense of not existing network that performs similar collocated measurements. In the following, answers to comments are reported just below each related comment. When needed, the part of the manuscript we modified or added to the old version is reported.
Other comments:
[line 34] do not use abbreviations before it is defined (e.g. BC_wb).
The ΄BCwb’ term has been introduced in the revised version of the manuscript, and the introduction now reads: «separated into fossil fuel (BCff) and biomass/wood burning (BCwb) fractions».
[lines 75-98] I appreciate short overview provided in this paragraph.
We thank the reviewer for this comment. A shorter version of this paragraph is provided in the revised version of the manuscript.
[line 167] Are the values AAE really constant for any site? Did you validated this premise using independent source(s) of information? Otherwise, how accurate are findings and how relevant are interpretations regarding wb and ff summarized in Sec. Results?
The AAE values applied in this study are in consistency with the corresponding values applied in a corresponding study in Athens (Liakakou et al., 2020). Even if there are no simultaneous radiocarbon measurements (Zotter et al., 2017) or ion m/z 60 samples to evaluate the applied values, previous studies have supported that the AAE values for ff and wb assumptions for BC absorption are considered fairly robust (e.g., Dumka et al., 2018, section 3.5; Kaskaoutis et al., 2021, section 4.5), especially in the longer wavelengths.In the latter study, a sensitivity analysis for AAEBC = 0.9–1.1 was performed, leading to changes of AbsBrC absorption up to 10.3–11.1% at 370 nm. Moreover, in the study of Rajesh et al., 2021, Table 1 summarizes the AAE values that have been applied in various studies, with applied values for wood burning (AAEwb in the range of 1.6 –2.2) and fossil fuel combustion (AAEff in the range of 0.9–1.1).
The following sentence is added in the revised version of the manuscript: «Source specific BC fractions, namely those related to wood burning (BCwb) and fossil fuel combustion (BCff) were calculated by Aethalometer, based on the absorption Ångstrom exponent (AAE470-950) values, assuming a fixed value for wood burning (AAEwb = 2) and fossil fuel combustion (AAEff = 1). This assumption is also applied in previous studies that support the AAE values for ff and wb assumptions for BC absorption are considered fairly robust »
References
Dumka, U.C., Kaskaoutis, D.G., Tiwari, S., Safai, P.D., Attri, S.D., Soni, V.K., Singh, N.,Mihalopoulos, N., 2018. Assessment of biomass burning and fossil fuel contributionto black carbon concentrations in Delhi during winter. Atmos. Environ. 194, 93–109.
Kaskaoutis, D.G.,Grivas, G.,Stavroulas, I.,Bougiatioti, A.,Liakakou,E.,Dumka, U.C. Gerasopoulos, E., Mihalopoulos, N., Apportionment of black and brown carbon spectral absorption sources in the urban environment of Athens, Greece, during winter, Science of The Total Environment, Volume 801, 2021, 149739, ISSN 0048-9697, https://doi.org/10.1016/j.scitotenv.2021.149739.
Liakakou, E., Kaskaoutis, D.G., Grivas, G., Stavroulas, I., Tsagkaraki,M., Paraskevopoulou, D., Bougiatioti, A., Dumka, U.C., Gerasopoulos, E., Mihalopoulos, N., 2020b. Long-term brown carbon spectral characteristics in a Mediterranean city (Athens). Sci. Total Environ. 708, 135019.
Rajesh, T.A., Ramachandran, S.,Vishnu K. Dhaker, Black carbon aerosols: Relative source strengths of vehicular emissions and residential/open wood burning over an urban and a semi-urban environment, Atmospheric Pollution Research, Volume 12, Issue 6,2021,101060, ISSN 1309-1042, https://doi.org/10.1016/j.apr.2021.101060.
Zotter, P., Herich, H., Gysel, M., El-Haddad, I., Zhang, Y., Mocnik, G., Hüglin, C.,Baltensperger, U., Szidat, S., Prévôt, A.S.H., 2017. Evaluation of the absorption Ångströmexponents for traffic and wood burning in the aethalometer-based source apportionmentusing radiocarbon measurements of ambient aerosol. Atmos. Chem.Phys. 17, 4229–4249.
[line 185] Figure 1 is missing; Figure 2 is not cited in the paper. I think something is corrupted with the numbering.
The reviewer is right. We reorder the numbering in the revised version of the manuscript.
[line 212] What you mean under the term “intensive properties”
The lidar derived intensive properties are type-sensitive and provide information for classification for the detected layers.
The revised sentence in the manuscript now reads: « The intensive properties per layer (aerosol type-sensitive properties), relevant to this study are the extinction-related Ångström exponent (AE), the backscatter-related Ångström exponent (BAE), the ratio of the backscatter coefficients profiles (color ratios - CRs), the lidar ratio (LR), and the ratio of the lidar ratios (RLR).»
[line 245] “ontains”-> “contains”
The sentence has been corrected in the revised version of the manuscript.
[line 268] Is this what you expect here or is there a proof that the large amplitudes of PM2.5 are associated with biomass burning particles?
The following sentence was changed in the revised version of the manuscript: «The PM2.5 daily concentrations ranged between 8 to 25 μg/m3 , with a mean value of 14 ± 5 μg/m3 during the summer campaign and between 2 and 55 μg/m3 during the winter campaign, with a mean value of 18 ± 12 μg/m3. »
[line 270] the same as above, but for transported dust particles.
During the winter period, a number of dust events occurred over Thessaloniki. However, the reviewer is right that the high PM10 can be contributed as well by the increased fine particle part while coarse particles remain the same. The following sentence was changed in the revised version of the manuscript: «On the other hand, the PM10 average daily concentrations ranged between 13 to 31 μg/m3 during the summer period and between 8 to 67 μg/m3 during the winter period with a mean value of 20 ± 5 μg/m3 and 28 ± 13 μg/m3, correspondingly. »
[Fig. 3a] What does it mean that PM10 almost perfectly mimics the trend of PM2.5? One could expect that the PM concentrations in such a case are determined by PM2.5 particles, while the number of particles larger than 2.5 microns scales up linearly with PM2.5. But what does it mean? How to interpret this result?
The reviewer is right that the PM10 and PM2.5 values are very close during the summer and winter period. This means that the higher PM10 in winter is more contributed by the increased fine particle part while the coarse particles remain almost unchanged. A relevant sentence is added in the revised version of the manuscript.
[Table 2] No optical properties of aerosols are provided here.
The reviewer is right. The revised label now reads: «Mean in-situ, layering and columnar retrievals, along with their standard deviation during the two PANACEA campaigns. »
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
I would like to thank the authors for the changes made in the article and the answers to my comments. Reading the new version of the article, I am confident that the changes improved the results, the readability, and the understanding for the readers. Although the results and conclusions are not novel, the co-located in situ and remote sensing observations are valuable, which might make this manuscript worthy of publication.
Reviewer 3 Report
I am fine with the present version of the manuscript.
