How "good" are mesoscale forecasts? The answer depends on what you want to
know. High-resolution models are often used to explicitly predict discrete,
highly structured phenomena. Thus information regarding the ability of the
model to predict events as coherent entities is a useful statement of
performance. Observational constraints are a significant problem, though,
as the shape, size, and intensity of any given event are often only
partially known. Composite techniques offer an attractive approach because
it is not necessary to know all information about any one event. If enough
quasi-random observations of a distribution of events exist, bulk properties
of the distributions of the event forecasts and observations can be
estimated.
A technique has been developed whereby meteorological events are isolated
using a rules-based algorithm and composited on a relative grid centered on
each event. This technique is explained and demonstrated by comparing the
27 km COAMPS(tm) Mistral wind forecasts to the SSM/I observations for a
one-year period. Diagnostic information regarding the forecast reliability,
error type, and error spatial characteristics will be shown. The results
indicate the Mistral is remarkably predictable, with high pattern
correlations out to 66 hours. The spatial distributions indicate the
center of the predicted Mistrals tends to be shifted slightly southwest of
the observations. Otherwise, at almost all forecast times, if a Mistral is
predicted there is a high probability that high winds will be observed.
|
