|
ECMWF has a wide-ranging programme of research
and development directed at improving the quality and variety of
forecast products for the medium range and beyond. Approximately
70 staff and consultants work in the Research Department on all
aspects of the forecast and data assimilation systems. The result
of this research is one of the most comprehensive earth-system
models available anywhere.
The benefits of the research are seen in the steadily improving
performance of the medium-range forecast and the expanding range
of operational products which maintain the status of ECMWF as one
of the world's leading centres for operational forecasting.
Data assimilation and forecast model
ECMWF has played a leading role in the development
of forecast and data assimilation systems. In recent years ECMWF
has pioneered the development of variational data assimilation methods.
The current '4D-Var' method provides a physically consistent best
fit to observations over a 12-hour period, taking account of the
time each observation was made, and is particularly useful for the
increasingly important assimilation of satellite data. When it was
introduced in 1997, this was the first such system to be used in
operational weather forecasting, representing the culmination of
years of research and development.
The forecast model is an integral part of the data
assimilation process. Model developments have contributed to improvements
in forecast skill both directly and through helping improve the
initial analysis. Recent model changes designed to improve data
assimilation include the extension of the top of the model atmosphere
from 30 km to 65km, and the introduction of a significantly more
sophisticated representation of the interaction between the atmosphere
and the underlying land surface (an important step towards the better
use of satellite data over land).
The increasing detail and sophistication of the forecast and data
assimilation system makes huge demands on computational resources.
The current operational system could not have been implemented without
the major increases in computational efficiency provided by the
continual development of the numerical algorithms used to solve the
forecast equations.
Predictability research
Despite the increasing accuracy of the analyses
and forecasts, there is still an element of uncertainty in the predictions.
Since 1992, ECMWF's pioneering ensemble prediction system has provided
a growing range of new products to help forecasters deal scientifically
and quantitatively with the day to day variations in the predictability
of the atmosphere. There is now a substantial research effort devoted
to the assessment of predictability on monthly and seasonal
timescales.
Ocean wave forecasting
A range of products forecasting sea-state is provided
by an ocean wave forecast model. This model is directly coupled
to the atmospheric model: the ocean waves are driven by low-level
atmospheric winds, and in turn the roughness of the sea determines
how much the air is slowed down as it flows over the ocean surface.
This coupled wave model forms part of the ensemble prediction system,
providing valuable probabilistic information on future sea states.
|
