Numerical Weather Prediction (NWP) have become a significant resource for weather forecasting, decision support systems, and other various applications dependent on meteorological information.
In a nutshell, a NWP model is a sophisticated set of computer programs containing mathematical and physical equations/algorithms that describe the atmosphere and how it will change over time in order to produce meteorological forecasts.
The initial condition for an NWP forecast comes from the Analysis, which is constructed by assimilating various meteorological observations into the model. The quality of the analysis (and therefore the forecast) depends not only upon the methods used, but also on the quality of the observations. Observations covering land, ocean areas, and the upper atmosphere are all important.
Depending on the application and need, multiple types of NWP models are commonly used, from global scale models that produce forecasts as far out as 15+ days with 15-100km resolution to regional or mesoscale models that produce shorter scale forecasts with much higher spatial resolution and more frequent output.