模式:

ECMWF: Global weather forecast model from the "European Centre for Medium-Range Weather Forecasts". ECMWF is now running its own Artificial Intelligence/Integrated Forecasting System (AIFS) as part of its experiment suite. These machine-learning-based models are very fast, and they produce a 10-day forecast with 6-hourly time steps in approximately one minute.

描述:

The method of Ivens (1987) is used by the forecasters at KNMI to predict the maximum wind velocity associated with heavy showers or thunderstorms. The method of Ivens is based on two multiple regression equations that were derived using about 120 summertime cases (April to September) between 1980 and 1983. The upper-air data were derived from the soundings at De Bilt, and observations of thunder by synop stations were used as an indicator of the presence of convection. The regression equations for the maximum wind velocity (w_max ) in m/s according to Ivens (1987) are:

if T_x - θ_w850 < 9°C

w_max = 7.66 + 0.653⋅(θ_w850 - θ_w500 ) + 0.976⋅U₈₅₀
if T_x - θ_w850 ≥ 9° C

w_max = 8.17 + 0.473⋅(θ_w850 - θ_w500 ) + (0.174⋅U₈₅₀ + 0.057⋅U₂₅₀)⋅√(T_x - θ_w850)

where

T_x is the maximum day-time temperature at 2 m in K
θ_wxxx the potential wet-bulb temperature at xxx hPa in K
U_xxx the wind velocity at xxx hPa in m/s.

The amount of negative buoyancy, which is estimated in these equations by the difference of the potential wet-bulb temperature at 850 and at 500 hPa, and horizontal wind velocities at one or two fixed altitudes are used to estimate the maximum wind velocity. The effect of precipitation loading is not taken into account by the method of Ivens. (Source: KNMI)

NWP:

Numerical weather prediction uses current weather conditions as input into mathematical models of the atmosphere to predict the weather. Although the first efforts to accomplish this were done in the 1920s, it wasn't until the advent of the computer and computer simulation that it was feasible to do in real-time. Manipulating the huge datasets and performing the complex calculations necessary to do this on a resolution fine enough to make the results useful requires the use of some of the most powerful supercomputers in the world. A number of forecast models, both global and regional in scale, are run to help create forecasts for nations worldwide. Use of model ensemble forecasts helps to define the forecast uncertainty and extend weather forecasting farther into the future than would otherwise be possible.

Wikipedia, Numerical weather prediction, http://zh.wikipedia.org/wiki/數值天氣預報(as of Feb. 9, 2010, 20:50 UTC).