Mohamed Jardak;
N.E. Bowler;
A.M. Clayton;
G.W. Inverarity;
A.C. Lorenc;
M.A. Wlasak
Met Office
The 8th EnKF Data Assimilation Workshop
Recent developments in the Met Office Ensemble Prediction System MOGREPS-G
Talk:
Jardak_presentation.pptx
A key aspect for a hybrid data assimilation system is to improve the quality of the flow-dependent
covariance information it receives from an ensemble.
For many years the Met Office has run a short-range ensemble prediction system (EPS) using an ensemble
transform Kalman filter (ETKF) to calculate the perturbations to the initial condition.
We have developed an improvement to the EPS based on an ensemble
of 4-dimensional ensemble-variational assimilations (En-4DEnVar).
This change increases the link between the data assimilation and the EPS, in the sense
that the ensemble perturbations are more representative of the forecast errors seen by the assimilation.
The quality of the ensemble prediction system is strongly affected by the method used to inflate
the ensemble spread.
To simulate the effects of model error we have supplemented the existing stochastic physics schemes with
a scheme known as additive inflation- where an archive of increments is randomly sampled and added
to model trajectory each time-step. This has proved very successful.
Although the ensemble of data assimilations naturally generates its own mean analysis it may be better
to recentre the ensemble on the analysis generated by the high-resolution
A key aspect for a hybrid data assimilation system is to improve the quality of the flow-dependent
covariance information it receives from an ensemble.
For many years the Met Office has run a short-range ensemble prediction system (EPS) using an ensemble
transform Kalman filter (ETKF) to calculate the perturbations to the initial condition.
We have developed an improvement to the EPS based on an ensemble
of 4-dimensional ensemble-variational assimilations (En-4DEnVar).
This change increases the link between the data assimilation and the EPS, in the sense
that the ensemble perturbations are more representative of the forecast errors seen by the assimilation.
The quality of the ensemble prediction system is strongly affected by the method used to inflate
the ensemble spread.
To simulate the effects of model error we have supplemented the existing stochastic physics schemes with
a scheme known as additive inflation- where an archive of increments is randomly sampled and added
to model trajectory each time-step. This has proved very successful.
Although the ensemble of data assimilations naturally generates its own mean analysis it may be better
to recentre the ensemble on the analysis generated by the high-resolution
Thanks to McGill's Atmospheric and Oceanic Sciences department for hosting this web-site.