A Re-Evaluation of the Swiss Hail Suppression Experiment Using Permutation Techniques Shows Enhancement of Hail Energies When Seeding

Round 1

Reviewer 1 Report

Review on Revised manuscript of

A Re-Evaluation of the Swiss Hail Suppression Experiment using Permutation Techniques shows Enhancement of Hail Energies when Seeding

Authors: Armin Auf der Maur and Urs Germann

Having considered the authors' response to my review and the changes in the content and conclusion of the article introduced by the authors, I believe that the article “A Re-Evaluation of the Swiss Hail Suppression Experiment using Permutation Techniques shows Enhancement of Hail Energies when Seeding ” can be published in journal Atmosphere.

Reviewer 2 Report

The authors have significantly improved the article and improved its content in many ways. Key scientific messages are now clearer. Generally, we can agree with the current version and support its publication.

Please correct below citation as following:

Abshaev M.T., Abshaev A.M., Sulakvelidze G.K., Burtsev I.I., Malkarova A.M. etc. Development of rocket and artillery technology for hail suppression. – In the book “Achievements in Weather Modification”, UAE, Abu Dhabi, 2006. – P. 109 – 127. [Available online at https://www.researchgate.net/publication/
705 343691402_ACHIEVEMENTS_IN_WEATHER_MODIFICATION_-_UAE_PRIZE_FOR_WEATHER_
706 MODIFICATION_Article_Development_of_Rocket_and_Artillery_Technology_for_Hail_Suppression_
707 Copyrights_C_Department_Atmospheric_Studies_Minist].

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

Study objective:

The work aims to re-evaluate data from the of Swiss Hail Suppression

Experiment using additional statistical methods to investigate the overall impact of the AgI seeding on the hail storm intensity (i.e. hail energy on ground).

Major Comments:

• Given that this is a re-evaluation of a well-documented and historical campaign, its must be carried out with careful consideration to not only the probalistic and numerical procedures, but also to couple it with the corresponding physical mechanisms. The authors state that microphysical modeling is beyond the scope of this work, however it is a crucial aspect to justify the statistical evaluations, and in this case re-evaluations which are more critical to readers to avoid compiling additional doubts to a longstanding campaign.

• If the modeling work cannot be pursued, additional physical investigation is needed by reporting the background concentration (and size range) of ice nucleating particles, expected concentration (and size range) of AgI seeding particles, updraft/downdraft magnitude ranges, thermodynamic structure of targeted clouds and how they varied throughout the campaign. These key variables should be further incorporated in the statistical and regression methods carried out.

Technical Comment/Corrections:

• Line 12-13: Revise the sentence “Damage on crops and other objects by hail is a disaster which lead especially farmers to protect 13 themselves by various means.”
• Line 16: Remove the word “still” and add a comma after “nowadays”
• Line 17-18: What is meant by “A convincing proof of the benefit of such actions is not known the present authors?” specify in what context since silver iodide seeding is well documented in the literature – see for example:

 French, J.R., Friedrich, K., Tessendorf, S.A., Rauber, R.M., Geerts, B., Rasmussen, R.M., Xue, L., Kunkel, M.L. and Blestrud, D.R., 2018. Precipitation formation from orographic cloud seeding. Proceedings of the National Academy of Sciences, 115(6), pp.1168-1173.

Friedrich, K., Ikeda, K., Tessendorf, S.A., French, J.R., Rauber, R.M., Geerts, B., Xue, L., Rasmussen, R.M., Blestrud, D.R., Kunkel, M.L. and Dawson, N., 2020. Quantifying snowfall from orographic cloud seeding. Proceedings of the National Academy of Sciences, 117(10), pp.5190-5195.

Rauber, R.M., Geerts, B., Xue, L., French, J., Friedrich, K., Rasmussen, R.M., Tessendorf, S.A., Blestrud, D.R., Kunkel, M.L. and Parkinson, S., 2019. Wintertime orographic cloud seeding—a review. Journal of Applied Meteorology and Climatology, 58(10), pp.2117-2140.

Tessendorf, S.A., French, J.R., Friedrich, K., Geerts, B., Rauber, R.M., Rasmussen, R.M., Xue, L., Ikeda, K., Blestrud, D.R., Kunkel, M.L. and Parkinson, S., 2019. A transformational approach to winter orographic weather modification research: The SNOWIE Project. Bulletin of the American Meteorological Society, 100(1), pp.71-92.

• Line 18-23: the outcome of the work should not be mentioned in the onset of the introduction. Instead, additional literature should be cited regarding AgI seeding for weather modification in general, the conceptual chain of events impacting the involved microphysical processes, and then referencing the uncertainty particularly in hail suppression evaluations to justify the motivation for the current work.
• Line 25: for readers not familiar with “Grossversuch IV”, please add additional information for clarity? Location? Duration? Experiment? Operational program? And add “(see Section 2)”
• Line 32: revise to “The magnitude of seeding coverage (sc) treatment variable ….”
• Line 33: “...contains information on how well…”
• Line 33-34: revise to “…how well seeding was done. Instead of using the objective values of sc, it was replaced by planned coverages while some 20% of the cells…
• Line 36: specify briefly why this is misleading and add the section or page where this is addressed in more detail in the paper e.g. “(see page/section xx for more details)”
• Line 42: “The 1986 study [3] also contained an explanatory…”
• Line 44: add the definition of the C(alpha) test or its brief objective
• Line 46: The review of hail suppression experiments by [1] does actually mention “The outcome of this experiment was not encouraging: in 205 seeding cases the seeding produced only 5% less risk”
• Line 47-48: Again, please try to keep result statements for the discussion and conclusion sections. Please rephrase to something like: The present re-evaluation aims to expand on the work of [3]…
• Line 53: add comma after “slightly different”
• Line 53: revise to “…will emerge with different ways…”
• Line 54-57: Remove or move all results from the current work to the results/discussion/conclusion sections i.e. after you state the methodology followed.
• Line 82: add parenthesis to (E_GR)
• Move Figure 1 to after Line 86 – after the first mention of Figure 1 in that paragraph
• Line 92: “...based on a Soviet era concept…”
• Line 99: remove comma and add parenthesis to “(see Figure 2)”
• Line 117: please specify the target error or uncertainty margin after these calibrations
• Line 126: please justify further why you think the logarithmic transformation removes importance for severe cases. In fact, logarithmic transforms are generally more useful for representing data with large magnitude differences instead of linear scale.
• Line 148: change “not transformed” to “non-transformed”
• Lines 148-162 should be moved to the discussion/conclusion
• Change the title of Section 3 to “Methods”
• Line 171: provide justification for “four times less effective PGIM type rockets”, since in Line 91 you mention that 4 PGIM rockets were sometimes used instead of the Oblako types.
• Also, it is unclear how your analysis accounts for the difference in rocket types?

Reviewer 2 Report

Review on the article

A Re-Evaluation of the Swiss Hail Suppression Experiment using Permutation Techniques shows Enhancement of Hail Energies when Seeding

Authors: Armin Auf der Maur and Urs Germann

The problem of evaluation of the physical and economic efficiency of weather modification projects is very relevant. In this regard, the article under consideration is of scientific and practical interest. The authors of the article conducted a thorough re-evaluation of the randomized Grossversuch IV experiment, in which the first version of the Soviet rocket method of hail suppression  [1] used in Moldova and Georgia was tested.

In Grossversuch IV experiment in the period 1977-1982, there were 83 experimental days, 253 convective cells were recorded that met the seeding criteria. According to the randomization procedure, 113 cells were selected for seeding, of which 20 cells were not seeded, and the seeding of 22 cells was below 1/3 of the required dosage of rockets. Hailstorms seeding were carried out with the help of Soviet rockets “Oblako” and “PGI-M”. The seeding objects were convective cells whose radar reflectivity exceeded 45 dBZ. During the period when the reflectivity exceeded 45 dBZ, one rocket “Oblako” or 4 rockets “PGI-M” was launched into the center of the cell every five minutes [2].

The statistical evaluation of the seeding efficiency was carried out by comparing the total kinetic energy of hail E_GR (MJ) falling from seeded and unseeded hailstorms, measured by the radar method verified by the haipad network readings and hail spectra [2]. As a result of this assessment, it was found that the kinetic energy of hail from the seeded group of convective cells is 3 times higher than the kinetic energy of hail that fell from the unseeded ones. I.e., the method of seeding hailstorms instead of preventing hail led to an increase in the kinetic energy and area of hail precipitation.

The reasons for such a negative result is that, along with a good scientific component and documentation in the Grossversuch IV experiment, seeding operations did not reproduce the method being tested and could not provide hail suppression:

1. First, in contrast to the method being tested, in which not only hailstorms were seeding, but also developing hail danger clouds, and frontal part of hailstorms was seeded, in Grossversuch IV only hailstorms with Z > 45 dBZ were sown in their central part, where hail has already formed and is growing. Such seeding can lead to an increase in updrafts due to the release of heat from the crystallization of cloud water and the intensification of hail growth due to coagulation in the medium of cloud droplets and crystals.
2. Secondly, the consumption of rockets in Grossversuch IV was 5-10 times less than in the method being tested (the annual consumption of rockets “Oblako” and “PGI-M” in Moldova or Georgia was stretched for 6 years).
3. Thirdly, the assumption of the method being tested about the limited water content of the hail growth zone, as a result of which hail growth can be prevented at a high concentration of artificial embryos, is not acceptable. A powerful updraft continuously brings a huge amount of droplet moisture into the hail growth area, and only part of this moisture turns into precipitation, since the precipitation-forming efficiency of hailstorms reaches 17-30%. It is especially low in supercell storms. Therefore, an increase in the concentration of hail embryos cannot significantly slow down the growth of hail due to the lack of liquid water, since it is introduced in excess by updrafts and will rather lead to an increase in the precipitation-forming efficiency of the cloud, the mass and kinetic energy of the hail. This is also noted by the authors of the reviewed article in section 4.3. Possible mechanisms.
4. Fourth, it is known [4] that the natural concentration of ice and snow graupels in the zones of hail origin and growth is 10³-10⁴ per m³ with a total mass of about 1 g/m³. This means that the natural concentration of hail embryos is always excessive, but this does not prevent the growth of some part of the embryos to dangerous sizes, since the growth of hail is selective.

Thus, the seeding criteria, the place of seeding and the amount of the introduced rockets did not fully meet the requirements of the being tested method of hail suppression - a method that really could not prevent hail even if the front part of the hail growth zone was sown instead of the central part of the hail growth zone.

Therefore, in the USSR in the 1980s switched from the concept of competition to a new concept of "acceleration precipitation" [3], refused to seeding the front part of the hail growth zone, and began to seeding of new growth areas of storms where are the conditions for the formation of hail embryos formed, including:

* Area of the first radar echo of new developing potentially hail-dangerous convective cells that originated above the isotherm -8 ^oC;

* Windward and frontal flank of developing really hail-dangerous convective cells;

* Windward and frontal flank of the radar echo overhang of mature hailstorms.

A massive seeding of the entire area of these zones with a series of new rockets that create an initial concentration of ice-forming particles along the seeding route of about 4×10¹⁰ - 10¹¹ m^-3 is envisaged. The distance between the rocket trajectories is no more than 0.5 - 1.0 km, in order to ensure the fusion of seeding volumes due to turbulent diffusion for a time of no more than 1 min.

It should be noted that in 1972-74, the US National Hail Experiment (NHRE) was conducted, in which the Sulakvelidze artillery method oh hail suppression [4] was evaluated, based on the concept of competition and seeding of the hail origin and growth zone. In NHRE, seeding was carried out with silver iodide particles from the aircraft in the area of the strongest updrafts [5], which can only accelerate the growth of hail due to the coagulation of a mixture of droplets and crystals.

In the experiments of Grossversuch IV and NHRE, quite naturally, negative results were obtained, which gave rise to pessimism about the hail suppression. Despite the fact that these experiments tested old versions of hail suppression methods that are no longer used, this pessimism persists to this day, and is reflected in the statement of the World Meteorological Organization [6]. A similar pessimism has also appeared in the field of artificial precipitation increase and is supported by those who are not carry out such projects and do not delve into the essence of the problems. Unfortunately, this pessimism is holding back investment in the development of weather modification projects in some countries.

However, climate change and a multiple increase of frequency and intensity of such dangerous phenomena as catastrophic hailstorms and accompanying dangerous thunderstorms, flush floods and mudslides of showers origin, strong winds and tornadoes require capacity - building in the field of their monitoring and prevention. The UN Sendai Framework Program on Disaster Risk Reduction for 2015-2030 also calls for this, which provides for a set of measures to significantly reduce the death rate, the number of victims and economic losses through the use of new technologies.

Conclusion and recommendations

1. The reviewed article convincingly shows that the previously used methods of seeding the central part of hail clouds with Z > 45 dBZ lead to an increase in the kinetic energy and the area of hail precipitation, and not to its prevention. This is of scientific and practical interest.
2. The article is recommended for publication in the journal Atmosphere after a small adjustment in terms of excluding further increase in pessimism in the field of hail prevention associated with the Grossversuch IV experiment.

To do this, it is recommended to indicate in the conclusion that such a negative result of hail suppression operations applies only to the seeding method used in Grossversuch IV, and does not apply to all other hail prevention methods.

References

1. Bukhnikashvili A.V., Gaivoronsky I.I., Kartsivadze A.I., Kiziriya B.I., Okudzhava A.M., Ordzhonikidze A.A., Seregin Yu.A. Methodology of hail suppression and results of experiments in the Alazan Valley // Proceedings of the All-Union Meeting on AV for hail processes. Tbilisi, 1964. pp. 76-87.
2. Federer, B.; Waldvogel, A.; Schmid, W.; Schiesser, H.H.; Hampel, F.; Schweingruber, M.; Stahel, W.; Bader, J.; Mezeix, J.F.; Doras, N.; DAubigny, G.; Der Megreditchian, G.; Vento, D. Main results of Grossversuch IV. J. Appl. Meteor. Climatol. 1986, 25, 917–957.
3. Sulakvelidze, G.K., 1967: Rainstorms and Hail. Hydrometizdat, Leningrad., 412 p. Translated by Israel Program for Scientific Translations (Available National Technical Inform. Service, Springfield, Va., NTIS TT-68-50466)
4. Heymsfield A.I., 1983: Case study of a hailstorm in Colorado. Part IV. Graupel and hail growth mechanisms deduced from particle trajectory calculations.- J. At. Sci., V.40, No 6, 1482-1509.
5. Abshaev M.T., 1994: The new concept of hailstorm modification // Sixth WMO Sci. Conf. on Weather Modif., Paestum, Italy, 1994. Vol. 1. P. 139-142.
6. Abshaev M.T. Russian rocket technology: seeding concept, criteria of hail detection, seeding schemas, radar control, rocket systems, efficiency// Report No 6. WMO Meeting on Hail Suppression, Nalchik, 2003. P. 91-98.

Reviewer 3 Report

There are some inaccuracies in the text of the article, for example:

1) L. 24-25 - instead of the former Russia, it should be written the former USSR. During the period of the Grossversuch IV, it was the USSR, and now its legal successor is Russia.

2) L. 25 Number of countries practiced hail suppression those time was much larger ( + Central Asia, China, Mongolia, Argentina, Brasil etc.). 

3) L. 25-27. Here are the postulates of only the concept of beneficial competition proposed by Georgy Sulakveldize (USSR). To complete the introduction it is suggested to add brief information on alternative concepts - the concept of precipitation acceleration from feeder clouds proposed by Abshaev (USSR) and the concept of hail trajectories lowering by Browning and G.Foote (USA), both are widely mention in the literature. Appropriate references are needed. 

4) L. 32 - I suggest to change phrase as "The magnitude of treatment variable called seeding coverage sc, the treatment variable called seeding coverage     varying from 0 to 1, ....

Lines 33 - 35 are excellent, thanks for them !

5) L. 31 - I suggest to use phrases like "to our knowledge", "most probably", etc. before points 1-4, instead of claiming that these particular points were exactly problematic, although I support it.

6. L. 31 - 41. I recommend adding one more very problematic side - the quality of the implementation of the hail suppression concept in the Grossversuch IV project was extremely unsatisfactory. The seeding norms per hailstorm were not followed, the seeding place also did not correspond to what it should have been. Unprepared students were involved in the work, who did a lot of spontaneous operations, etc.  

6) L. 82 - I suggest to give a formula for radar estimation of hail energy on ground EGR.

7) what is 1+ on horizontal and vertical axis labels in fig. 1 ?

Common comments:

1) Taking into account modern concepts, obtained including with the help of dual-pol radars, seeding areas with reflectivity > 45 dBZ, can really stimulate the hail growth, since in this area hail is already growing naturally and it is too late to introduce the glaciogenic aerosol here to compete with natural hailstones.

2) It is not easy to read an article, the main analyzed data of which is in another article (original source from Bruno Federer). But in general, this is permissible.

3) In fact, given the current understanding of the hail nucleation and growth, the seeding of hail storms carried out in the Grossversuch IV project did indeed stimulate the enlargement of the growing hail rather than its reduction. As the author correctly noted, in hail storms it is unlikely to create competition for growing hailstones due to artificial ice particles. This is exactly what can be found in the 45 dbz zone. That is the reason why current hail suppression technology in Russia is based on seedon of newly formed feeder clouds with reflectivity below 35 dBZ.

4) Although I'm not an expert in statistical data processing, this part seems to me to be pretty good. It is upsetting that the unseeded 28 cells are included in the seeded storm statistics. And yet it seems to me that it is more logical and understandable to operate with cells, rather than days, because in one day the hailstone pillow can be hitted by hailstones of different cells.

5) To operate with the calculation of hail energy from radar data, it would be necessary to give the average statistical error in determining this parameter, which can be significant for many reasons. Then perhaps it will become clearer why the data on the hailpads and the radar are so different.

6) Sub-chapter 4.3 is good

7) L. 495. Larger hail size(number), larger hailfall area and longer hailfall duration.   

8) Generally I am against using term "quality of seeding" (sc) (for instance in Line 516) determined by simple rate of seeding. It is more complex and determined by quality of realiation of the scientific hypotheses, quality of seeding materials used, quality of delivering seeding materials, time delays, etc. 

9) Line 519 -    + hailfal duration . For seeding of cell core with Z > 45 dBZ) we actually increase hailstones size(number), area of hailfall and its duration.