Blog

3D Isolated Cumulonimbus SDM Simulation and Visualization

A high-resolution SDM cloud simulation visualized as CG by us and shared on the JST channel for the Moonshot program.

3D Isolated Cumulonimbus SDM Simulation and Visualization cover image

We conducted a high-resolution simulation of an isolated cumulonimbus cloud using the Super-Droplet Method (SDM), then converted the simulation output into CG imagery to visualize the cloud’s evolution and internal microphysical structure.

Simulation Setup

  • Model approach: Super-Droplet Method (particle-based cloud microphysics)
  • Horizontal grid spacing: 62.5 m
  • Domain size: 60 x 60 x 25 km
  • Output: physically based simulation rendered as CG for intuitive interpretation
Cross-section visualization of cumulonimbus cloud life cycle
Life cycle visualization of an isolated cumulonimbus cloud (cross section).
Color-coded visualization by cloud particle type
Color-coded simulation frame by cloud particle category (cloud water, rain, graupel/hail, snowflakes, and cloud ice).

Why This Matters

This work supports Moonshot Goal 8 research on severe-weather understanding and control-oriented forecasting. As discussed in Prof. Shima’s interview, several points directly connect to this simulation:

  • SDM represents massive particle populations with representative super-droplets, enabling detailed cloud-particle evolution at feasible computational cost.
  • Compared with traditional bulk approaches, SDM is better suited for resolving fine microphysical processes and aerosol-related effects.
  • These simulations are still computationally expensive (for example, a short cloud event can require roughly one week of compute), which is why algorithmic acceleration, supercomputing, and AI-assisted pathways are being developed for future operational use.

Newer Post Back to Blog Older Post