Dr. Harolda Fraga de Campos Velho-March 13

Data assimilation by artificial neural network

Start

March 13, 2018 - 10:30 am

End

March 13, 2018 - 11:30 am

Address

120 David L. Boren Blvd. Room 4140, Norman, OK 73072   View map

Data assimilation by artificial neural network

Data assimilation is an essential process in operational prediction centers based on the time
integration of partial differential equations. The numerical weather prediction is a very relevant
example. The process of data assimilation is an inverse problem of initial condition identification
– called ”analysis” –, combining observations with data from the forecasting mathematical model.
Historically, the mathematical estimation process was started by least square approach. In deed,
all data assimilation schemes in some sense are based on least square estimator. We are going
to show the time evolution from the least square up to Kalman filter, and beyond – the particle
filter, and one step ahead. The variational method is also a generalized least square formulation.
In the talk, results with several data assimilation methods emulated by artificial neural networks
will be shown. The algorithmic complexity is reduced employing neural networks. This easy
realized for a 3D atmospheric models. Two atmospheric general circulation models, SPEED
and COAPS-FSU, are used to demonstrated the data assimilation speed-up with neural networks
emulating Local Ensemble Transform Kalman Filter (LETKF), and producing similar analysis

MORE DETAIL

Phone

405-325-6561

Email

chomeyer@ou.edu