Investigating the Large-Scale Transport of a Volcanic Plume and the Impact on a Secondary Site

Round 1

Reviewer 1 Report

This paper examines datasets in South Africa and Chile for the presence of volcanic ash and SO2 in the atmosphere during the June 2011 eruption of the Puyehue-Cordon Caulle volcanic center (PCCVC).  The paper also uses output from flexpart model simulations of the 2011 PCCVC eruption to examine the path of the plume as it passed near Cape Point, South Africa, where some measurements were being made.    Key findings of the study are that an increase in aerosol loading was observed between June 10 and 20 at Cape Point, South Africa, presumably as a result of fine ash.  Both the model results and VIIRS satellite retrievals showed that the main plume travelled east-west at latitudes south of Cape Point, although the ash plume was more dispersed.

Overall I think that the methodology and results of the paper are well presented, and the significance is great enough to merit publication.  However there are many minor issues and a few more significant ones that I think should be addressed before publication.  The more significant ones are:

• Several aspects of the methodology could be clearer. For example, SO2 column loading in Figures 5 and 6 come from Modern Era Retrospective Analysis (MERRA) model output.  What SO2 datasets were ingested into the MERRA model that formed the basis of the SO2 output?  Are there any published validation studies of this output?  What time period was used to come up with mean values of UVI in the OMI data presented in Fig. 2?  Or for the mean AOD measurements in Fig. 3?
• Model setup and inputs for the flexpart simulations are not well described.   There is no description for example of eruption source parameters, only a vague statement that “no a priori information was used”.  If you used an inverse method such as that of Eckhardt et al. (2008) to derive source parameters, this should be mentioned, and the satellite or other data used in the inversion should be described.
• There seem to be a few omissions in the figures. The caption to Figure 2 for example describes sub-figures (a) and (b), but only figure is shown.  The caption to Figure 4 describes error bars, but there are no error bars on the plot. 
• You say in the abstract that there was an observed increase in SO2 at Cape Point on June 10-20, 2011. I can’t see this in Fig. 6b.  In the text you’re a little more vague about this, saying (line 185) that there was large variability in daily mean values of SO2 column loading at Cape Point based on MERRA model output.  In Fig. 6b, it’s hard for me to see any significant anomaly in SO2 over this time period.  But it’s also hard to identify the time period of June 10-20 in Fig. 6b, because there are no tick marks on the x axis indicating the start of each month.  Adding tick marks to the x axis, and a box around the time period June 10-20 would help.

More specific comments are below.  Overall I think these issues can be addressed with minor modifications to the manuscript.  I look forward to seeing the paper in print.

Larry Mastin

Minor and technical comments, some of which duplicate or expand upon points above:

Line 29:  remove the period after “investigated”

Line 55: change “has” to “as”, and “effect” to “affect”

Line 56: change the comma to a period.

Line 57: change “as particles” to “ash particles”

Line 59:  capitalize “Mountains” in “Andes Mountains”

Line 73: change the period at the end of the line to a comma.

Line 93:  the ultraviolet index should be more clearly defined.  Does it have units?

Line 75:  Looks like a tracked change was not accepted on this line.

Line 78:  add “site” after “secondary”

Lines 111-115:  presumably the MERRA model output of SO2 column mass is based on some SO2 datasets that have been ingested into the model.  Do you know what these datasets are?  Or their frequency, spatial resolution, uncertainty etc?  Are there any published studies that test their accuracy?

Line 114:  add “, and” after “[22]”.

Line 118:  change “in which calculates” to “, which calculates”

Line 120:  add “radiation with wavelengths” between “scatters” and “between”.

Line 121: change “spectra” to “spectral band”

Lines 136-145: many aspects of the flexpart model setup are not mentioned here.  Is there another publication that can provide the details?  What is meant by “no a priori information was used in the FLEXPART model simulation”?  Did you use an inverse method similar to Eckhardt et al. (2008) to derive the inputs?  If so, perhaps you should cite Eckhardt et al. (2008) in this sentence (as well as on line 143), and say more explicitly that you used that methodology.    

Lines 139-140: “In this study, forecast meteorological data from the European Centre for 139 Medium Range Forecasts (ECMWF) was used as input for the model”.  You used forecast data, not reanalysis data, for the dispersion modeling?  Perhaps you’re just confusing the two data types?  Forecast data are generated from models that forecast into the future, and are generally only available in real time, or within a few weeks of a given forecast date.  Reanalysis data are based on reanalysis of meteorological conditions that occurred in the past.

Lines 145-146: “The particles were released at heights between 2 -14 km above ground-level and at 6-hour 145 intervals [10].”  The model does not allow for continuous ash emission?  Only pulses, every six hours?

Figure 2:  the caption refers to subfigures (a) and (b), but there is only one image here.  In addition, it’s not clear to me what “UVI anomalies for Monthly mean and 1 SD of solar noon UVI” means.  You took the monthly mean UVI, added one standard deviation to it, and then plotted the deviations from that?  Also, over what time period were the mean and SD UVI calculated?

 

Line 157:  clarify “the monthly mean and 1 SD of PFR” to note that the error bars represent +/-1 SD in Fig. 3.  also, here you mention a weak seasonal cycle in AOD in Fig. 3, but it’s hard for me to see one.   Which part of the year do you think has lower AOD?  Any seasonal differences in mean value are less than one standard deviation from one another.

Line 157:  over what time period was the mean calculated for AOD at this station? 

Figure 4:  in the caption you say that the error bars represent one SD, but there are no error bars on this plot.  How were the points connected by the solid lines derived? Are they the 24-hour average of the 1-minute values of AOD taken throughout the day?  Also, it would help to plot the standard deviation from the mean monthly values.

Figure 5a:  what do the error bars represent?  The standard deviation of SO2 column mass through the month in 2011?  Or the standard deviation of monthly values for each month, from the long-term record?  And over what time period is the long-term record (if it’s relevant)?  The same questions pertain to Fig. 6.

Figure 5b and 6b: please add tick marks to the x axes of these plots to indicate where each month starts.

Lines 184-185: “In June and July 2011 (Figure 6b), the daily mean values were below the climatological daily mean values”.  By “climatological mean values” are you referring to the monthly means in Fig. 6b?  If so, perhaps add (monthly) to this sentence to clarify.

Line 185 and Fig. 6b:  please put a box or something around the dates 10-20 June 2011.  I can’t see any anomalous SO2 in June, but because there are no tick marks on the x axis denoting the start of each month, it’s not possible to tell when the 10-20 June time period on this plot is.

Lines 198-199:  I don’t know whether you want to speculate on the cause of differences between flexpart and IASI results in SO2, but there could be several causes of the higher near-source SO2 dispersion in the flexpart results than IASI.  Errors in the vertical distribution of SO2 mass in the initial flexpart plume and/or errors in the modeled wind field are the obvious candidates.

Line 219 (and elsewhere): I don’t think it’s necessary to capitalize “Ash”.

line 222:  by “zonal transport” are you referring to the confinement of the cloud to a narrow band of latitudes?

Line 229: add a date after “lower volcanic emission”

Line 239:  clarify this sentence.  It seems to be missing some key words.

Line 250:  it sounds here like you’re saying that ash dispersed over the secondary site is responsible for increased SO2 column mass, which doesn’t make sense to me. Perhaps reword.

Lines 253-254: the Author Contributions could be more specific about who analyzed each dataset, and about who did the flexpart modeling.

References:

Eckhardt, S., Prata, A.J., Seibert, P., Stebel, K. and Stohl, A., 2008. Estimation of the vertical profile of sulfur dioxide injection into the atmosphere by a volcanic eruption using satellite column measurements and inverse transport modeling. Atmospheric Chemistry and Physics, 8: 3881-3897.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

It is a kind of good study. This manuscript is well written and organized; can be accepted for publication after some minor corrections. 1. Line 82: Correct the longitudes and latitudes. Add either N/S or E/W. 2. Captions in Figs 3 and 4: It should be written as mean ± 1 SD. Also do this correction throughout the ms. 3. Add your study's highlights in the abstract. Good luck.

Author Response

Please see the attachment.

Author Response File: Author Response.docx