Numerical weather prediction

Numerical weather prediction forecasts atmospheric conditions by using computers to solve mathematical equations that describe the atmosphere and its evolution.

Numerical weather prediction model: A computer model that simulates atmospheric conditions by solving equations governing the atmosphere. It turns atmospheric equations into a system that can be integrated to produce forecasts.
Primitive equations: A set of equations describing large-scale atmospheric motion, including momentum, mass, energy, and moisture. They provide the core dynamical framework used by many global forecast models.
Weather forecasting: The prediction of future atmospheric conditions at specified places and times. Numerical prediction is a major computational approach within the broader practice of forecasting.
Statistical weather forecasting: Weather prediction based on statistical relationships between past observations and future conditions. It predicts from learned or historical patterns rather than directly integrating atmospheric equations.
Predictability: The extent to which a system's future state can be inferred from its present state and governing dynamics. Atmospheric chaos places a finite limit on how far ahead detailed forecasts remain useful.
Data assimilation: Methods that combine observations with a model's previous estimate to create an initial state. It constructs the starting atmospheric state from incomplete, unevenly distributed measurements.
Navier–Stokes equations: Differential equations describing the motion of viscous fluids. Atmospheric dynamics are fluid motion governed by equations in this broader family.
Numerical weather prediction guidance: Model-generated forecast fields used by meteorologists to assess likely future weather. Guidance translates model output into information used to prepare operational forecasts.
Analog forecasting: A forecasting method that uses past weather situations resembling the current one to infer future conditions. It relies on historical resemblance, not a numerical simulation of atmospheric evolution.
Lorenz system: A three-dimensional nonlinear system of differential equations that exhibits deterministic chaos. Its behavior illustrates how tiny initial errors can grow into major forecast differences.

Linked from 56 pages