Vertical distribution of aerosols over the Maritime Continent during El Niño

Cohen, Jason Blake; Ng, Daniel Hui Loong; Lim, Alan Wei Lun; Chua, Xin Rong

doi:https://doi.org/10.5194/acp-18-7095-2018

Articles | Volume 18, issue 10

https://doi.org/10.5194/acp-18-7095-2018

© Author(s) 2018. This work is distributed under
the Creative Commons Attribution 3.0 License.

https://doi.org/10.5194/acp-18-7095-2018

© Author(s) 2018. This work is distributed under
the Creative Commons Attribution 3.0 License.

Articles | Volume 18, issue 10

Research article

|

23 May 2018

Research article |

| 23 May 2018

Vertical distribution of aerosols over the Maritime Continent during El Niño

Jason Blake Cohen, Daniel Hui Loong Ng, Alan Wei Lun Lim, and Xin Rong Chua

Download

Final revised paper (published on 23 May 2018)
Supplement to the final revised paper
Preprint (discussion started on 16 Feb 2017)
Supplement to the preprint

Interactive discussion

Status: closed

AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

- Printer-friendly version

- Supplement

RC1: 'Referee comment: acp-2016-1179', Anonymous Referee #2, 15 Mar 2017
- AC1: 'Response to Reviewer Number 2', Jason Cohen, 27 Mar 2017
RC2: 'Interactive comment on “Vertical distribution of aerosols over the Maritime Continent during El Nino”', Anonymous Referee #1, 18 Mar 2017
- AC2: 'Response to Reviewer 1', Jason Cohen, 01 Apr 2017

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision

AR by Jason Cohen on behalf of the Authors (23 May 2017) Author's response Manuscript

ED: Referee Nomination & Report Request started (28 May 2017) by Aijun Ding

RR by Anonymous Referee #3 (13 Jun 2017)

Suggestions for revision or reasons for rejection

Review of “Vertical distribution of aerosols over the maritime continent during El Nino” by Cohen et al.

This paper uses CALIPSO profiles to investigate the plume heights of fires during the El Nino season of October 2006. By comparing with and adjusting a plume rise model match with observation, the authors claim that current fire emission power is underestimated, which is the main conclusion of their study. While the paper offers some interesting results, I feel the authors still fail to provide enough evidence to support that their finding is “significantly” different from previous studies. In particular, the authors did not sufficiently address the data quality issues raised in the first round review. My specific comments are listed below:

1. The plume height analysis entirely depends on CALIPSO extinction profiles and vertical feature mask product. However, both these two products bare uncertainties that could contaminate the results. Especially that the feature classification is based on lidar ratio which is highly uncertain. The author should provide evidence that these data are validated against ground based lidar for the region and period of interest.
2. The fire region is selected according to MISR AOD. While I agree that MISR cloud screen might be better than MODIS, its sampling is rather poor over low latitudes. This means that many small scale plumes may not be captured by MISR at all. I suggest the authors compare both MODIS and MISR and maybe also OMI which is good in measuring absorbing aerosols to better determine the fire region or to confirm their defined region.
3. The authors state that the CALIPSO statistics is based on more than 10,000 profiles. In my opinion, it is not the number of profiles that matters but the number of plumes or fire events that the sampling represents. Usually one plume or event may contain tens or even hundreds of profiles. So are these 10,000 profiles complete or representative or the entire biomass burning season?
4. The conclusion that fire power is underestimated is solely based on matching the plume rise model with CALIPSO observed mid plume height. However, the model still shows quite different low and high plume heights. Given that the latter two is also related to emission power, the conclusion seems unsound.
5. I still have some problem with the claim that the result of the current paper is very different from previous studies. This is also the main issue raised by the first round of reviewers. The authors cited Tosca et al. (2011) who stated that the plumes are mostly confined within the boundary layer. The Tosca et al. results are based on the entire 2001-2009 period rather than just October 2006, therefore not directly comparable with the current study. The authors need to be more careful about their statement and conclusion and provide direct comparisons with previous research.
6. Lines 162-163: is the resolution 10km or 1km? Also I suggest the authors change the “x” in “10kmx10km” to the real multiplication signs by inserting symbols.
7. Lines 183-185: please provide references that AOD and other products are validated.
8. Lines 258-260: I don’t understand this sentence. Please rephrase.
9. Line 289: “this work’s underlying Kalman Filter plus variance maximization inversely modeled fields”, is this used in the plume rise model? If so, please describe more specifically in the model description section.
10. Line 325: lower temperature should correspond to “lower emission factor” rather than “higher aerosol emission factor”.
11. Line 400-401: This statement is too strong. Given the data quality problem and model mismatch problem, I don’t think the work “comprehensively quantifies …”.
12. The paper still has many grammar mistakes, such as verbs associated with plural or single forms. Please double check.

Hide

RR by Anonymous Referee #4 (25 Jun 2017)

Suggestions for revision or reasons for rejection

The manuscript uses satellite-derived AOD to spatially and temporally constrain the sampling of smoke aerosols with an aim to examine the aerosol vertical distributions, measured from CALIOP, over the maritime continent during the 2006 El Nino. The observed aerosol vertical distributions were then used to compare with the results of a simple plume rise model. The study provides some insights into the aerosol signature in terms of vertical distribution during El Nino conditions and the limitations of plume rise models. But additional evidences and analysis are required to support the conclusions. Several major comments have to be addressed. Furthermore, the manuscript readability and clarity has to be improved before the publication in ACP.
Major comments:
1. More elaborations and descriptions are required for CALIPSO data processing. Which version and level of CALIOP product? Is each individual measurement under cloud-free conditions? If yes, which cloud mask data was used? How many samples in total? What is the threshold value of extinction form CALIOP data (please consider the daytime background solar illumination by Winker et al., 2013)? In addition, an analysis of the uncertainties of the CALIOP-derived vertical aerosol extinction, in particular over this region, is needed.
2. The effects of the uncertainty in boundary layer depth need to be considered. The authors simply use the 1000 m to approximate the boundary layer height. Assuming the boundary layer has +/- 300 m uncertainties during the CALIPSO overpass, which is totally possible, what are the uncertainties of the percentage of free atmosphere aerosol estimated by your method? When taking this into account, how does your result compare with previous studies?
3. The author uses aerosol-induced in-situ stabilization as a possible explanation to the underestimation of plume height by model. But the rationale seems problematic. The model does not account for the effect of aerosol-induced stabilization which actually happens in the real atmosphere. The stabilization causes weaker buoyancy, thus lower plume height. Therefore the model that misses such stabilization should overestimate, not underestimate, the plume height.
4. The comparison between model and observation is insufficient. The observation misses 3 days and the model misses 9 days with only 18 days left. This is rather a small sample. Since the fires lasts from September to November, it is worthwhile to expand the analysis to September and November. In addition, the analysis is primarily limited to the monthly averages. The authors do show the comparisons in daily basis in Figure 4, but do not analyze them. In particular, the three special days mentioned in section 3.2 are good example cases to analyze in order to shed more light on the observation-model comparisons. Such more comprehensive analysis is very worthwhile in order to support the conclusions of how to reduce model bias which is actually not well examined or indicated in the manuscript.

Specific comments:
Line 15: “measurements and modeling”. Please specify which measurement and which model.
Line 16: “underestimated” by what?
Line 51: Sentence not readable
Line 53 ~ 54: “underestimation” in “spatial, and temporal distribution”?
Line 60: Change “show” to “shown”
Line 75: Show full name of “CALIOP”
Line 77: Show full name of “SSA”
Line 77: “go with each pass”. Is it scientific language?
Line 82: grammar error
Line 85: Show full name of “MISR”
Line 157: Please provide the reference.
Line 158: delete one “are”
Line 162~167: Show full name of “AERONET”, “NOAA”,”RMS”, “RCP”,”GDED”
Line 167: what is R2 statistic?
Line 204~206: Please provide new plot to show them more directly.
Line 218: Show full name of “BC”
Line 272: add “of” after “and”
Line 285~286: Why not examining the hypothesis by looking at precipitation data from A-Train?
Line 290: Show full name of “MERRA”
Line 360-361: “vertical distribution” is not a parameter to be “estimated”.

Figure 1: What does the color stand for? Please add a title to the colorbar.
Figure 4: The comparison is really not readable. Please make it more clear.

Referee Report: PDF

Hide

ED: Reconsider after major revisions (02 Jul 2017) by Aijun Ding

AR by Jason Cohen on behalf of the Authors (09 Mar 2018) Author's response Manuscript

ED: Referee Nomination & Report Request started (14 Mar 2018) by Aijun Ding

RR by Anonymous Referee #4 (18 Mar 2018)

RR by Anonymous Referee #3 (26 Mar 2018)

ED: Publish subject to minor revisions (review by editor) (29 Mar 2018) by Aijun Ding

AR by Jason Cohen on behalf of the Authors (07 Apr 2018) Author's response Manuscript

ED: Publish as is (16 Apr 2018) by Aijun Ding

Download

Article (1877 KB)
Full-text XML

Short summary

Measured aerosol heights over the Maritime Continent are higher than previously thought, with 61 to 83 % of aerosols above the boundary layer. These aerosols should hence have a larger impact on the climate. The use of a plume rise model cannot match the measurements, unless the measured fire energy is increased by 0–60 %. Furthermore, the model is too spread, indicating the importance of including convection and aerosol–radiation interactions. Significant model improvements will be required.