CVP Brown Bag Seminar - "Adaptive Covariance Inflation in the EnKF by Gaussian Scale Mixtures"

03/23/2018 - 12:00pm
ML Chapman Room
"Adaptive Covariance Inflation in the EnKF by Gaussian Scale Mixtures"
 
Patrick N. Raanes, Postdoctoral Fellow at Nansen Environmental and Remote Sensing Center
 
We study inflation: the complementary scaling of the state covariance in the ensemble Kalman filter (EnKF). Firstly, error sources in the EnKF are catalogued and discussed in relation to inflation; nonlinearity is given particular attention. Then, the “finite-size” refinement known as the EnKF-N is shown to be a Gaussian scale mixture, again demonstrating its connection to inflation. Existing methods for adaptive inflation estimation are reviewed, including the one used in DART. One such method is selected to complement the EnKF-N to make a hybrid that is suitable for contexts with model error. Benchmarks are obtained from experiments with the two-scale Lorenz model where only the slow scale is resolved. The proposed hybrid EnKF-N method of adaptive inflation is found to yield systematic accuracy improvements in comparison with the existing methods, albeit to a moderate degree.
 
Speaker: Patrick N. Raanes, Postdoctoral Fellow at Nansen Environmental and Remote Sensing Center; Co-authors: Marc Bocquet, Alberto Carrassi
Reference: arxiv.org/abs/1801.08474v1 (in review at QJRMS)
 
Biography:
Patrick is just finishing a 2-year postdoc at the Nansen center (Bergen, Norway), where he has been working on the theory of the EnKF and related methods. This was also the topic of his PhD, submitted at the maths institute in Oxford, where he also did his masters degree. His bachelors degree is in engineering, from Toulouse. His work has focused on smoothing (for reanalysis and iteration purposes) and square root updating schemes. In his next job, he will look at the control problem under the uncertainty parameterised by ensembles.
 
Friday, March 23, 2018
12:00 p.m. - 1:00 p.m.
Mesa Lab, Chapman Room
Bring your lunch!