Hiraga et al. (2026) Numerical experiments of cloud seeding for mitigating localization of heavy rainfall: a case study of Mesoscale Convective System in Japan

Identification

• Journal: Natural hazards and earth system sciences
• Year: 2026
• Date: 2026-03-11
• Authors: Yusuke Hiraga, Jacqueline Muthoni Mbugua, Shunji Kotsuki, Yoshiharu Suzuki, Shu-Hua Chen, Atsushi Hamada, Kazuaki Yasunaga, Takuya Funatomi
• DOI: 10.5194/nhess-26-1287-2026

Research Groups

• Department of Civil and Environmental Engineering, Tohoku University, Sendai, Japan
• Center for Environmental Remote Sensing, Chiba University, Chiba, Japan
• Institute for Advanced Academic Research, Chiba University, Chiba, Japan
• Department of Civil and Environmental Engineering, Hosei University, Tokyo, Japan
• Department of Land, Air and Water Resources, University of California, Davis, CA, USA
• Faculty of Sustainable Design, University of Toyama, Toyama, Japan
• Academic Center for Computing and Media Studies, Kyoto University, Kyoto, Japan

Short Summary

This study numerically investigated the potential of cloud overseeding to mitigate localized heavy rainfall from a mesoscale convective system in Japan, finding that an optimal seeding configuration could reduce area-averaged 3-hour accumulated rainfall by 11.5% and maximum rainfall by 32% in the heavy rainfall region.

Objective

• To systematically investigate the effectiveness of cloud overseeding for mitigating localized heavy rainfall associated with a Mesoscale Convective System (MCS) in Japan, examining the sensitivity of rainfall responses to various seeding conditions (altitude, horizontal extent, vertical thickness).

Study Configuration

• Spatial Scale: Three nested computational domains with horizontal resolutions of 25 km, 5 km, and 1 km. The innermost domain (1 km resolution) covered the Chugoku and Shikoku regions of Japan. Cloud seeding experiments were conducted over areas ranging from 6 km × 6 km to 30 km × 30 km.
• Temporal Scale: Simulations ran from 00:00 to 19:00 UTC on 19 August 2014, with a 16-hour spin-up. Cloud seeding was applied from 16:00 to 19:00 UTC on 19 August 2014, with analysis focused on 3-hour accumulated rainfall during this period.

Methodology and Data

• Models used: Advanced Research version of the Weather Research and Forecasting model (WRF), version 4.1.2, coupled with the Morrison 2-moment cloud microphysics scheme. Other physics parameterizations included Kain–Fritsch (cumulus), RRTMG (shortwave/longwave radiation), MYNN 2.5 (planetary boundary layer), Revised MM5 (surface layer), and Noah-MP Land Surface Model.
• Data sources: Initial and boundary conditions were derived from NCEP Global Data Assimilation System (GDAS) FNL operational global analysis (0.25° horizontal resolution, 34 vertical levels, 6-hour intervals). WRF-simulated rainfall was verified against Radar/Rain gauge-Analyzed Precipitation (RA data) (1 km spatial and hourly temporal resolution).

Main Results

• Seeding in the mid–upper troposphere (7.2–8.6 km altitude), where air temperatures ranged from −22 to −12 °C, resulted in the most pronounced changes in rainfall due to high supercooled cloud water content and strong updrafts favoring heterogeneous freezing.
• Cloud seeding led to reduced rainfall within the heavy rainfall region and increased rainfall downwind, quantitatively demonstrating the hypothesized dispersal mechanism of "overseeding."
• Expanding the seeding area to cover the upstream region of the heavy rainfall area had a greater impact on rainfall modification than increasing the vertical thickness of the seeding.
• The most effective seeding configuration (24 km × 24 km area at 7.2 km altitude) achieved an 11.5 % decrease in area-averaged 3-hour accumulated rainfall and a maximum reduction of 32 % in 3-hour accumulated rainfall over the heavy rainfall region.
• While this optimal configuration also induced greater rainfall increases further downwind (up to 23.7 mm, 310.5 %), reductions in extreme 3-hour rainfall remained dominant, indicating mitigation of localization rather than simply shifting comparable hazards.
• A rough estimation suggests that maintaining the observed increase in ice nuclei concentration for 3 hours over the 24 km × 24 km × 500 m seeding volume would require approximately 3240 kg of dry ice.

Contributions

• Systematically investigated the effectiveness of cloud overseeding for mitigating localized heavy rainfall in a Mesoscale Convective System (MCS) in Japan.
• Provided a quantitative demonstration of the hypothesized overseeding mechanism, showing rainfall reduction in the target area and downwind dispersion.
• Clarified the sensitivity of rainfall responses to seeding altitude, horizontal extent, and vertical thickness, identifying optimal conditions for mitigation.
• Showed that expanding the seeding area to capture upstream convective cells was more effective than increasing vertical thickness.
• Established a valuable foundation for understanding the potential of cloud overseeding and optimal conditions for mitigating heavy rainfall in convective systems.

Funding

• JST Moonshot R&D Program (JPMJMS2389-5-3)
• Joint Usage/Research Center for Interdisciplinary Large-scale Information Infrastructures (JHPCN) (Project ID: jh240013)
• High Performance Computing Infrastructure (HPCI) in Japan (Project ID: jh240013)

Citation

@article{Hiraga2026Numerical,
  author = {Hiraga, Yusuke and Mbugua, Jacqueline Muthoni and Kotsuki, Shunji and Suzuki, Yoshiharu and Chen, Shu-Hua and Hamada, Atsushi and Yasunaga, Kazuaki and Funatomi, Takuya},
  title = {Numerical experiments of cloud seeding for mitigating localization of heavy rainfall: a case study of Mesoscale Convective System in Japan},
  journal = {Natural hazards and earth system sciences},
  year = {2026},
  doi = {10.5194/nhess-26-1287-2026},
  url = {https://doi.org/10.5194/nhess-26-1287-2026}
}