Stratospheric Water Vapor from the Hunga Tonga–Hunga Ha’apai Volcanic Eruption Deduced from COSMIC-2 Radio Occultation

Round 1

Reviewer 1 Report (Previous Reviewer 1)

The authors' rebuttal to the reviews for the previous version of the paper inidicate that authors addressed the reviewers' concerns, including mine. Therefore, the current version can be published.

Author Response

Thank you for the positive comments, and there were no suggested changes to address.

Reviewer 2 Report (New Reviewer)

Comments for author File: Comments.pdf

Author Response

We thank the Reviewer for positive comments and suggestions. Responses to specific comments are included below, and we have made corresponding changes to the paper.

General comments:

The paper studies the stratospheric water vapor after the Hunga Tonga – Hunga Ha’apai (HTHH) volcanic eruption that occurred on January 15 2022 using refractivity profiles from COSMIC-2 GNSS radio occultation data, MLS temperature retrievals as well as nearby radiosone measurements. The study shows that HTHH eruption injects water vapor up to 110 +/- 14 Tg which is about 8% of total background mass of stratospheric water vapor. The authors also show that water vapor enhancement is about 2500-3500 ppmv in the stratosphere (~29-33 km) in the days following the eruption and the plume traveled westward and dispersed in about 10 days following the HTHH eruption. The foundation of an article that might appear in the Remote Sensing is already present. The scientific methods and analysis are sound, and the article is well-written and logically organized. Therefore, I recommend for a prompt publication of the paper. I also have a few comments for the authors that hopefully will help the readers of their paper to better understand the work and compare their work with the existing literature.

Other comments:

- In my view it is useful to add a few sentences in the first paragraph explaining the differences in the amount of water vapor injection to the stratosphere via the land-locked volcanic eruptions such as El Chicon in 1982 and Pinatubo in 1991 and the recent submarine HTHH volcanic eruption. For example, an injection of water vapor to the stratosphere was not observed after Pinatubo eruption in 1991 while this study along with previous studies suggests there is a huge injection of water vapor to the stratosphere following HTHH volcanic eruption. The large values of water vapor estimates in the stratosphere can be explained by pointing out that the HTHH volcanic outburst was a submarine eruption.

We agree with this suggestion and we have modified the beginning Introduction text as follows: ‘The January 2022 volcanic eruption of Hunga Tonga – Hunga Ha’apai (~20^o S, 175^o W; hereafter HTHH) was the most explosive eruption in the satellite era, spanning the last four decades. Because HTHH was a submarine eruption, it injected large amounts of water vapor (H₂O) directly into the stratosphere. This behavior is distinctive from other large eruptions from land-locked volcanoes such as El Chichon in 1982 and Pinatubo in 1991, where enhanced stratospheric H₂O was not observed’.

- The existing abstract is fine and informative but from the reviewer’s perspective it is interesting to also mention that the estimation of 110+/- 14 Tg for the total mass of water vapor equals to about 8% of the background mass of stratospheric water vapor.

Good suggestion.  We have modified the text to: ‘The total mass of H₂O injected by HTHH is estimated as 110 +/- 14 Tg from measurements in the early plumes on January 16-18, which equates to about 8% of the background global mass of stratospheric H₂O’.

-In section 4 (discussion) it is helpful to compare your estimate of water vapor injection following HTHH volcanic eruption (110+/- 14 Tg) with previous works. For example, Khaykin et al (2022) estimated the global stratospheric water mass enhancement by 13% relative to climatological valus following HTHH eruption. While no observational report of an increase in the amount of water vapor after Pinatubo eruption in 1991 was reported, the study of Pitari and Mancini (2002) reported 37.5 Tg injection of water vapor to the stratosphere but their work was based on modeling and not observation.

We had previously included comparisons of our H₂O mass calculations with MLS and radiosonde estimates in Section 3.3, but now have moved this to the Discussion section (lines 374-379). We have also included a reference to the model results of Pitari and Mancini (2002).

- I found the statement in line 486-487 weird as it judges the comments before even receiving them!

This statement refers to referee comments on the original submission of the paper, but it also applies to these additional helpful comments.

Khaykin, S., Podglajen, A., Ploeger, F. et al. Global perturbation of stratospheric water and aerosol burden by Hunga eruption. Commun Earth Environ 3, 316 (2022). https://doi.org/10.1038/s43247-022-00652-x.

Pitari, G. and Mancini, E.: Short-term climatic impact of the 1991 volcanic eruption of Mt. Pinatubo and effects on atmospheric tracers, Nat. Hazards Earth Syst. Sci., 2, 91–108, https://doi.org/10.5194/nhess-2-91-2002, 2002.

Reviewer 3 Report (New Reviewer)

The eruption of the Hunga Tonga-Hunga Ha’apai (HTHH) volcano on January 15, 2022 injected large amounts of water vapor into the stratosphere, which could affect the global climate for several years after the eruption. It is meaningful to monitor the HTHH water vapor signal and its propagating patterns. However, there are a few concerns about this submission. I would expect authors consider all concerns below in order to ensure a statistically meaningful conclusion.

1.      P2L68-69: According to your description, you use nearby MLS measurements to separate the influences of temperatures. However, the MLS profiles are much less than C2 profiles in this region. Can you explain the details how to match the COSMIC2 and MLS profiles?

2.      P4Figure 1: How do you get the sigma value in the figure 1? Is it calculated based on the historical C2 profiles?

3.      P6L186 “within 600 km and 6 hours”: Why you choose this range? Have you quantified the effect of co-located temperature profile within different range? The temperature may have large differences within 600 km and 6 hours.

4.     P8Figure 5: Which height of the temperature data are you shown in figure 5? The different number and vertical resolution of MLS and C2 profiles are taken into account in the figure 5?

5.     P10L317-318: “ 98+/- 11, 103+/-14 and 102+/-14 Tg” How to use the C2 measurements get this value?

6.      P12L371-372: How to get the value “110+/- 14 Tg”? Why is different with the calculated value in P10L317-318?

Author Response

We thank the Reviewer for positive comments and suggestions. Responses to specific comments are included below, and we have made corresponding changes to the paper.

The eruption of the Hunga Tonga-Hunga Ha’apai (HTHH) volcano on January 15, 2022 injected large amounts of water vapor into the stratosphere, which could affect the global climate for several years after the eruption. It is meaningful to monitor the HTHH water vapor signal and its propagating patterns. However, there are a few concerns about this submission. I would expect authors consider all concerns below in order to ensure a statistically meaningful conclusion.

1. P2L68-69: According to your description, you use nearby MLS measurements to separate the influences of temperatures. However, the MLS profiles are much less than C2 profiles in this region. Can you explain the details how to match the COSMIC2 and MLS profiles?

We used the nearest MLS profile within 600 km and 6 hours for each of the COSMIC2 profiles, as described in the text (line 193). This is as good of co-location as could be achieved with the available data.

2. P4Figure 1: How do you get the sigma value in the figure 1? Is it calculated based on the historical C2 profiles?

The sigma values are derived from pre-volcanic (January 1-14) C2 measurements in the region near HTHH. This is stated on lines 147-148.

3. P6L186 “within 600 km and 6 hours”: Why you choose this range? Have you quantified the effect of co-located temperature profile within different range? The temperature may have large differences within 600 km and 6 hours.

We tested several different co-location criteria, and the 600 km and 6 hour criteria was as good as can be achieved with available MLS sampling. Choosing a smaller co-location range results in cases with no available MLS data; a larger range give identical results to those shown.

4. P8Figure 5: Which height of the temperature data are you shown in figure 5? The different number and vertical resolution of MLS and C2 profiles are taken into account in the figure 5?

Sorry that we omitted this information.  We have changed the Fig. 5 caption to include: ‘Temperature anomalies are calculated at the altitude of the maximum retrieved H₂O in each profile.’

5. P10L317-318: “ 98+/- 11, 103+/-14 and 102+/-14 Tg” How to use the C2 measurements get this value?

This is described in the text in Section 3.3. We calculate the plume horizontal areas (as shown in Fig. 2), and then vertically integrate (over 25-35 km) the derived H₂O profiles (e.g Fig. 3) within the specified plume areas. Multiplying the plume areas by the integrated H₂O gives the H₂O mass within the plume volume.  We have added this latter statement in Section 3.3.

6. P12L371-372: How to get the value “110+/- 14 Tg”? Why is different with the calculated value in P10L317-318?

This is described in the text (lines 330-332). The average of our calculations for January 16-18 results in 100 +/- 14 Tg, and we then take account for an approximate 10% underestimate of H₂O from the local retrieval (described in the Appendix line 438 and Fig. A1) to give a final estimate of 110 +/- 14 Tg.

Round 2

Reviewer 3 Report (New Reviewer)

I am fine with the revised version submitted by the authors.

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

General Remarks.

The paper investigates the sensitivity of RO observations to water vapor injections from an extreme event (Hunga Tonga – Hunga Ha’apai volcanic eruption). The author find examples of large anomalies in retrieved refractivities associated with stratospheric water vapor. Especially large perturbations are observed when the RO sounding intersects the volcanic plume. These results are interesting and deserve publishing.

However, the paper needs some improvements. The goal of the study conforming with the paper text must be clearly formulated in the title and in the abstract. There are some doubts regarding the humidity retrieval algorithm. My conclusion is that the text needs a major revision.

 

Specific Comments

 

L. 48. The objective of this paper is to explore the retrieval of the extreme stratospheric H2O amounts from HTHH in the first week after the eruption using COSMIC-2 (C2) GNSS radio occultation (RO) data.

 

What is the meaning of "exploring the retrieval"? The authors must clearly formulate the goal of the study. What do they study: the atmospheric conditions caused by an extreme event (volcanic eruption), or the retrieval scheme? If the object of the study is the atmospheric conditions, then more discussion and references are needed: How does it compare to other similar events? How does it influence the atmospheric dynamics and radiation transfer?

 

The Results and Discussion sections address the sensitivity of RO observations to stratospheric H2O. This is a good point. But then this must be reflected in the title and abstract. Shortly, in the current form the paper misses a clear formulation of its goal, and it doesn't have any Conclusions section.

 

L. 125. The sensitivity of N to isolated large stratospheric H2O perturbations (1000 and 1500 ppmv anomalies at 30 km) is illustrated in Fig. 1, showing positive N anomalies of ~2% and 3%, respectively, compared to an unperturbed background.

 

The authors write about "localized" perturbations without explicitly specifying the spatial scale of the perturbation. Figure 1 suggests that the anomaly has a scale of 4 km, which can hardly be termed "localized". How realistic is such a perturbation? How was this figure obtained? Did the authors performed an end-to-end simulation? If so, did they use geometric-optical or wave-optical simulation?

 

L. 146. Anomalies are calculated as % differences from the non-volcanic background January average.

 

I guess, the average of C2 data is meant here. How exactly did you evaluate the average? Explicitly formulate the value of sigma. Did you average the refractivity after the statistical optimization or the results of the raw retrieval?

 

L. 164. We calculate a simplified estimate of H2O based on Eq. 1, incorporating C2 N observations and MLS temperatures as independent information, and using dry pressure p from the C2 atmPrf files. Specifically, for each profile we take the full C2 N profile (not anomalies) and find the nearest co-located MLS temperature profile (within 600 km and 6 hours). We then solve Eq. 1 for water vapor pressure e and convert to H2O mixing ratio.

 

The dry pressure is obtained by integrating the hydrostatic equation under the assumption of dry atmosphere, which implies a simple linear relation between the refractivity and density. Dry pressure is not the atmospheric pressure p in Eq. (1), unless the atmosphere is dry. The standard hybrid inversion implies using the dependence e(p,T,N) from Eq. (1), which is substituted into the state equation and hydrostatic equation. Then, given N(z) and T(z), the hydrostatic equation can be integrated to produce p(z) and, therefore, e(z). Note, one of the authors of the paper is also one of the authors of Report No. 199 of Max-Planck Institute for Meteorology:

https://www.mpimet.mpg.de/fileadmin/publikationen/Reports/Report119.pdf

Chapter 7 "Deriving tropospheric humidity from refractivity and temperature" describes a simple iterative solution of the hydrostatic equation and the state equation.

 

Reviewer 2 Report

 

This manuscript demonstrates that stratospheric water vapor anomalies from volcanic injection which exceed ~1000 ppmv are readily detectible using COSMIC-2 radio occultation (RO) data.   In the 25-35 km layer, where the Hunga Tonga-Hunga Ha'apai eruption led to westward-travelling, layered water vapor anomalies, it proved optimal to reduce the half-width of the vertical response function by half, to 0.75 km, relative to standard retrievals, without increasing noise. This technique is complementary to information gained from MLS temperature (vertical resolution 3-4 km) and radiosondes (vertical resolution of 100 m, but the burst altitude is ~30 km altitude).   The paper is clearly written and well-organized.   The use of radiosondes over Australia, estimates of westward rate of travel, and evolution of the water vapor anomalies' vertical extent during January 16-24, 2022 are some of the interesting aspects of the paper.   I recommend publication with minor revision.  I have a few comments and questions below.

 

1.  Is it possible to make a slightly more specific statement about effective resolution in the 25-35 km layer?  Line 85 suggests that normal RO soundings have a resolution of 1-2 km.  Lines 93-97 state that the half-width of the response function was reduced from 1.5 to 0.75 km.   Is it fair to say that the new effective resolution is 0.5-1 km?

 

2.  An 8 K anomaly probably wouldn't happen in the tropics and one that large in the extratropics would probably be vertically thick, so this would seem to make it unlikely that there is an ambiguity of interpretation for water vapor > 1000 ppmv,  and that non-isolated 3 sigma events in delta N are due to water vapor anomalies.   Is this a correct interpretation?

 

3.  Figs. 2 and 6:  It looks like U ~ - 22 m/s near 30 km.  It might be interesting to readers to include an estimate and relate it to the phase of the QBO.

 

4.  Fig. 3 and 7 and lines 175-176:  Why does the mean altitude of the water vapor anomalies descend with time?  Is there any effect due to vertical shear of the easterly flow?  (e.g., Strong flows at one level would advect water vapor anomalies somewhere at that level first.)   If the pattern change is not due to re-shaping in some sense by the winds, how could radiative cooling due to excess water vapor lead to such a quantitatively large excursion of water vapor across isentropes?

 

5.  Lines 214-216:  Please explain a bit more about why dry retrievals should show negative temperature anomalies and the significance of Fig. 5. 

 

6.  Lines 247-249 and Fig. 7:  Why would the descent rate be 160 m/day and then change?   How certain are you that the evolution of the spread in plume height isn't due to differential advection in the vertical?   Why choose the first of the two linear fits?  Prior to January 18 the data look like a scatter shot without strong evidence of descent.   Why is only the vapor near 29 km evident after 1/19?  Is this evidence that there are strong variations in horizontal wind vector with height?

 

 

 

Reviewer 3 Report

In this manuscript, the extremely large stratospheric H2O perturbations during the volcanic eruption of Hunga Tonga – Hunga Ha’apai on January 2020, are detected using the Radio Occultation measurement. This manusrcipt further enhances the RO applications in space weather research. However, an issue still needs to be improved before pubulication. 

As shown in the Figure 2 of this manuscript, the refractivity N anomalies  and H2O perturbations are obviously detected for some RO events, while are still not captured for many RO events in the region of the HTHH eruption. Thus, in this manuscript the authors at least provide the ratios between the detectable and undetectable RO events , and try to give some reasonable explanations for the undetectable RO events.