Numerical prediction is one of the modern techniques in the field of meteorology. It is based on a set of physical and dynamical equations that simulate the processes of the atmosphere and oceans to forecast future changes in weather, sea conditions, or climate.
These equations and physical processes are translated into programming code, and initial inputs to the model are prepared by data assimilation of meteorological observations. The calculations rely on advanced Numerical methods and require significant computational power to get results. Therefore, high-performance computing (HPC) systems are typically used to run these models.
Numerical and ocean models produce detailed 3D forecast maps for atmospheric and marine parameters extending up to 80 km in the atmosphere and several kilometres deep in the oceans, depending on the model used.
Examples of model outputs include:
- Air Temperature
- Relative humidity
- Precipitation amount
- Wind speed and direction
- Tropical cyclone tracks
- Sea wave height
- Storm surge level
- Ocean currents