the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 19 Sep 2025
Numerical Experiments of Cloud Seeding for Mitigating Localization of Heavy Rainfall: A Case Study of Mesoscale Convective System in Japan
Abstract. This study investigated the potential of cloud seeding to mitigate extreme rainfall localization (i.e., overseeding) associated with mesoscale convective systems in Japan. Using a numerical weather prediction model, we conducted cloud seeding experiments by artificially increasing ice nuclei concentrations in a double-moment microphysics scheme for the heavy rainfall event in Hiroshima Prefecture, Japan, in August 2014. We examined the sensitivity of rainfall changes to altitudes and areas of the seeding. The results showed that seeding in the mid–upper troposphere (7.2–8.6 km), where air temperature ranged from −22 °C to −12 °C, resulted in the most pronounced changes in rainfall amount. At these levels, high supercooled cloud water content and strong updrafts favored heterogeneous freezing, resulting in a depletion of moisture and suppression of graupel growth. The cloud seeding led to reduced rainfall within the heavy rainfall region and increased rainfall downstream, demonstrating the hypothesized dispersal mechanism of “overseeding”. Expanding the seeding to cover the upstream region of the heavy rainfall area had a greater impact than increasing vertical thickness of the seeding. The most effective seeding configuration (24 km × 24 km area at 7.2 km) achieved an 11.5 % decrease in area-averaged 3-hr accumulated rainfall and a 32 % decrease as the maximum reduction in 3-hr accumulated rainfall over the heavy rainfall region. Future work should consider more realistic representations of seeding substance (i.e., transport, dispersion, and interactions) and explore a wider range of rainfall events to generalize the applicability of this approach.
- Preprint
(4237 KB) - Metadata XML
-
Supplement
(3132 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2025-3524', Anonymous Referee #1, 23 Oct 2025
-
AC1: 'Reply on RC1', Yusuke Hiraga, 13 Jan 2026
We thank the reviewer for the positive and encouraging evaluation of our manuscript. We appreciate the reviewer’s assessment that the study is interesting, well written, and suitable for publication. We are also grateful for the constructive suggestions aimed at strengthening the novelty of the work and for highlighting an important limitation and direction for future research.
In response, we have prepared the revision to better clarify these aspects and to explicitly acknowledge the additional limitation and future research needs. We have also carefully addressed the minor grammatical suggestions provided by the reviewer and conducted an overall review of the manuscript to further improve clarity and readability.
Please find our responses in the attached document.
-
AC1: 'Reply on RC1', Yusuke Hiraga, 13 Jan 2026
-
RC2: 'Comment on egusphere-2025-3524', Anonymous Referee #2, 22 Dec 2025
This manuscript presents a clear numerical case study of cloud overseeding aimed at reducing the localization of extreme rainfall during the August 2014 Hiroshima event. The sensitivity tests of seeding height and seeding area are well organized, and the results provide useful insight into when the model is most responsive. I recommend minor revisions focused on adding clearer explanation and justification of real-world feasibility, uncertainty, and the practical implications of increased rainfall downstream, so that readers can interpret the findings appropriately.
Major comments
1. Feasibility and operational realism: Please expand the feasibility discussion so readers can understand what real deployment could look like for the most effective configurations, which occur in the mid- to upper troposphere and require a large horizontal footprint. Clarify the assumed operational geometry (in cloud, cloud edge, or inflow) and the plausible delivery platforms at the target altitude. Please also discuss practical constraints near vigorous convection, including turbulence, icing, and flight safety, and how these constraints limit where seeding can be performed.
2. Targeting constraints in real experiments: Please describe limitations that arise from limited control over seeding location and timing in practice. In particular, discuss whether the reported optimal 24 km by 24 km case depends on perfect targeting of evolving upstream convective cells and how sensitive the conclusions might be if seeding must be restricted to smaller areas, shorter durations, or locations and timings offset from convective cores due to forecast and nowcast uncertainty.
3. Physical meaning of the seeding method: Please provide additional justification for representing overseeding by multiplying the ice nuclei concentration in the Meyers formulation by a very large factor within a fixed box. Readers will benefit from a direct statement of what magnitude of ice number enhancement is actually produced in the seeded volume during the seeding period (for example, typical and peak values). This would clarify how the chosen multiplier relates to a plausible range of effective ice nuclei increases and how strongly the results depend on this choice.
4. Downstream impacts and net-hazard framing: Because the mechanism redistributes rainfall downstream, the results should be discussed in a net-hazard context. Please add a concise diagnostic or justification showing whether overall extremes are reduced rather than shifted. At minimum, report whether the domain-wide maximum 3-hour rainfall increases or decreases in the seeded runs and how the area exceeding key thresholds changes when computed from each seeded run, not only using the control-defined heavy-rainfall mask.
5. Uncertainty and robustness: The manuscript notes strong nonlinearity and sensitivity to atmospheric chaos, but the results are interpreted mainly from single realizations. Please add a short justification for how robust you expect the sign and magnitude of the rainfall changes to be relative to internal model variability for this convective case. If additional ensemble simulations are not feasible, clearly state this limitation and adjust wording so conclusions are framed as conditional sensitivity results.
Minor comments
6. Clarify the seeding time protocol in the model: whether the multiplier is applied continuously at each model time step from 16:00 to 19:00 UTC, or applied intermittently.
7. For reproducibility, consider providing the specific code modification location where the multiplier is applied, as well as the key namelist settings used for these experiments.
8. Check the following typos:
- “Morrison 2-momnet” in Table 1 (should be “2-moment”)
- “upper raw, middle raw, and bottom raw” in the Figure 6 title (should be “row”)
- “folloing” in Table 2 (should be “following”)
Citation: https://doi.org/10.5194/egusphere-2025-3524-RC2 -
AC2: 'Reply on RC2', Yusuke Hiraga, 13 Jan 2026
We thank the reviewer for the positive and constructive evaluation of our manuscript. We appreciate the recognition of the clarity of the numerical case study, the organization of the sensitivity experiments, and the usefulness of the results in identifying conditions under which the model is most responsive.
In response to the reviewer’s suggestions, we have prepared the revision to improve the clarity of the discussion on real-world feasibility, uncertainty, and the practical implications of downstream rainfall increases, so that readers can more appropriately interpret the findings. We believe these revisions have strengthened the manuscript while remaining consistent with its mechanism-focused scope.
Please find our responses in the attached document.
-
AC2: 'Reply on RC2', Yusuke Hiraga, 13 Jan 2026
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 1,525 | 289 | 36 | 1,850 | 58 | 31 | 25 |
- HTML: 1,525
- PDF: 289
- XML: 36
- Total: 1,850
- Supplement: 58
- BibTeX: 31
- EndNote: 25
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
The authors present a numerical analysis of the potential for cloud seeding to mitigate the rainfall from an extreme historical MCS event in Japan. The paper is interesting and well-written and I think is suitable for publication with moderate revisions that would help strengthen the novelty of the study and to highlight an addition (in my opinion, important) limitation and direction for future work. I also have a few minor grammar suggestions. I doubt I caught all the grammar mistakes, which anyway are so small and infrequent that they do not detract from the quality of the work.
My concern regarding novelty stems from the fact that very little information is provided on numerical cloud seeding studies, particularly those aimed at rainfall mitigation and those based in Japan. This lack of information on previous work made it impossible for anyone who isn’t themselves familiar with the literature to understand how much was new about the present study. That said, the introduction section is already lengthy. Therefore, I recommend creating a new section entitled “Background” or something like that, following the introduction, that can provide further details on methods and findings of a few key previous studies, which will help frame the current work. Some points now provided in the introduction may perhaps be moved into that background section as well.
My other major concern is the lack of mention of what I see as the biggest limitation of the work and one of the biggest questions that needs to be answered if such approaches are ever to be operationalized. The authors examine the effects of cloud seeding directly into the storm area. In an operational context, this is akin to knowing exactly where the storm will take place. MCSs are notoriously difficult to forecast, particularly with respect to precise timing and location. Therefore, it seems unrealistic that such seeding could be accomplished in so localized a way. Instead, one would presumably need, with some multi-hour lead time, a prediction that a larger region is likely to develop an MCS somewhere within it, and seeding is done throughout the region. It is easy to imagine several things as a consequence of this: 1) the amount of AgI or other material needed could increase by orders of magnitude, and 2) the likelihood that some adverse outcome occurs due to rainfall being shifted spatially, rather than mitigated entirely, will increase, perhaps substantially. I suggest that the authors consider followup work that examines seeding, perhaps at lower concentrations, over larger areas, rather than just the storm location, to examine this effect. It may also be wise to perform ensemble simulations using some perturbation approach to better understand how stochastic the result may be.
Minor comments:
Throughout: “downwind” is likely better than “downstream” for atmospheric work
L17 and 18; delete “s” from “altitudes” and “areas”
L25: “… decrease as the maximum reduction…” is awkward wording
L66 and 69: delete uses of “the” except at the start of line 69
L70: delete “to date” as it is unnecessary
L88: change “in” to “on”
L89: the 100 mm/h is unclear. Is that a peak rate over some short time interval?
L97: delete “s” from convections
L103: “under a future climate”
L104: countermeasure is one word
Eqn 1: Does beta have a physical meaning? If so, describe it. Also, what is a more typical value in WRF?
L195 and Table 2: use “design” rather than “flow”
L195: change “was” to “is”
Figure 4: remind the reader what the thick black polygon indicates