Better data selection for forecast models
STAR and the
National Centers for Environmental Prediction, of the National Weather
Service (NWS), are improving the prediction of weather from numerical
models, especially the Global Forecast System (GFS). The two centers have
tried and tested new techniques that exploit the power of the National
Aeronautics and Space Administration (NASA) Atmospheric Infrared Sounder
(AIRS) instrument. The GFS model cannot use every single data point, or
"field of view (FOV)," of the AIRS
instrument, so it uses only one of
every nine fields of view. This greatly speeds up the processing of data
and cuts the time required to make a forecast. The field of nine points
corresponds to a single field of view of another instrument (the Advanced
Microwave Sounding Unit). The new technique selects the one best field of
the nine that is "least obscured by clouds" in all channels of the
AIRS
instrument. Since the model cannot use all nine points, it can either take
the center point (blue bars in the figure), or it can use the "least
cloudy view" (red bars). The figure suggests that the latter choice leads
to better forecasts.
Day 5 Anomaly Correlations for Mid-Latitudes
Geopotential Heights Waves 1-20
|
