Supercomputers & Radar Boost Storm Forecasting

     (CN) — Identifying sudden torrential rainstorms before they occur can be done more accurately when supercomputers and high-tech weather radar are used in conjunction, according to an upcoming study.
     Existing tools for predicting storms rely on supercomputer simulations using grids spaced at least one kilometer apart that incorporate new observational data every hour. The calculations are often imprecise, however, and can’t accurately predict sudden torrential rainstorms that can develop within minutes.
     An international team of researchers has incorporated the K computer and advanced radar observational data to predict the occurrence of torrential rains in specific areas. The key is “big data assimilation” using computational power to synchronize observational data and large-scale computer simulations.
     The team used the K computer — a supercomputer located at the RIKEN Advanced Institute for Computational Science campus in Kobe, Japan — to compute 100 parallel simulations of a convective weather system using the nonhydrostatic mesoscale model, the primary short-term weather forecast model for the United States.
     Instead of spacing the grids at about 2 kilometer to 5 kilometers apart, the team used 100-meter grid spacing, and assimilated data from a next-generation phased-array weather radar. This enabled the team to establish high-resolution three-dimensional distribution map of rain that refreshes every 30 seconds, or 120 times more rapidly than the typical hourly updated systems used at weather prediction centers around the world.
     To ensure the accuracy of the simulations, the team attempted to model a real case — a 2013 storm that occurred in Kyoto, Japan — close enough to Osaka that it was caught by the radar stations at Osaka University. The 100-meter grids enabled very accurate replication of the storm in comparison to actual observations.
     Takemasa Miyoshi, who led the study, mentioned several potential uses for enhanced weather forecasting.
     “Supercomputers are becoming more and more powerful, and are allowing us to incorporate ever more advanced data into simulations. Our study shows that in the future it will be possible to use weather forecasting to predict severe local weather phenomena such as torrential rains, a growing problem which can cause enormous damage and cost lives,” Miyoshi said in a statement.
     Their findings will be published later this month in the journal Bulletin of the American Meteorological Society.

%d bloggers like this: