Improving Prediction of Fronts

Once a weather
model has been initialised from our best estimate of the current state
of the atmosphere, it will diverge from the real state of the
atmosphere over time. The period of time during which a single
deterministic weather model provides useful information is known as
the "useful forecast range'', which provides a measure of the
predictive skill of the model. For example, the ECMWF useful forecast
range is now typically between 7 and 10 days. Extending the useful forecast range of weather
forecasts (and improving accuracy over that range) would have clear
benefits to society and the economy; benefits include: energy management, protection of
infrastructure, coordinating disaster response, agricultural planning,
planning in travel and leisure industries, transportation management,
targeting of weather-sensitive health conditions, water resource
management, improved decision making in commodity markets,
environmental decision making and natural resource management. In this
project, we will develop a means of understanding of some current
limitations to predictive skill and the useful forecast range, and
will investigate practical methodologies for removing them. This is of
course a huge topic; we shall restrict our investigation to the dry
dynamical core (the part of the model that predicts winds, temperatures and pressures),
and concentrate on one important phenomena, namely the
evolution of fronts and, in particular, the resulting feedback
on the large scale atmospheric circulation. Fronts play a key role in the delivery of
precipitation, and many severe weather events, which have an important
role in all of the socioeconomic areas discussed above. The challenge associated
with fronts is that although the distance from one side of a weather front
to another is very small compared to the distance between two grid points in
an operational forecast model, fronts have a significant impact on the wind speed
and direction as well as the pressure over a large region around the front. The aim of this project
is to find improvements in dynamical core design that can improve the prediction
of the winds and pressures in the vicinity of fronts.

Beneficiaries of this research include:
* Operational forecasting centres such as the UK Met Office
* Customers of operational forecasting centres such as businesses and policy makers.

Extending the useful forecast range of weather
forecasts (and improving accuracy over that range) would have clear
benefits to society and the economy. Areas that would benefit include improvements in:
energy management, protection of infrastructure, coordinating disaster response, agricultural planning,
planning in travel and leisure industries, transportation management,
targeting of weather-sensitive health conditions, water resource
management, improved decision making in commodity markets,
environmental decision making and natural resource management.

This is blue skies research which will seek to identify strategies to improve
predictability in dynamical cores for numerical weather prediction. Any successful
strategies would be likely to be included on the timescale of the UK "GungHo" Dynamical
Core project which is aiming for operational use by 2020.

Dynamical core developers from operational forecasting centres will be encouraged to engage
through the test case workshop: as well as key participation from the UK Met Office and
the associated "GungHo" dynamical core project, invitees will include representatives of ECMWF,
the DWD/Hamburg/Rosstock ICON team, the NCAR/LANL MPAS team, the Japanese NICAM team,
developers of the various CAM dynamical cores such as CAM-SE, etc.

Policymakers will be engaged through the Grantham Institute by means of a Grantham Discussion Paper
on the topic of predictability and model error.