The recorded video is uploaded to YouTube. The talk from ECMWF is not included due to the issue of permission.
The video is splitted into two parts. Part 1 is available from this
Link; and Part 2 from this
link.
The available slides can be found below:
- 09:00-09:45 Multiscale alignment ensemble filtering technique and its application in geoscience. Link
- 09:45-10:15 Integrating measurement representativeness and release temporal variability to improve the Fukushima-Daiichi 137Cs source reconstruction. Link
- 11:30-12:00 Combining machine learning and data assimilation to improve the hydrodynamic forecasting in a tidal estuary. Link
- 13:30-14:00 Benefit of vertical localisation in sea surface temperature assimilation: identical twin experiments. Link
- 14:00-14:30 A Gaussian high-order sampling hybrid filter for biogeochemical data assimilation: application to chlorophyll satellite data. Link
- 16:30-17:00 Using Deep Learning to increase the ensemble size in an EnKF with the recentering technique: experiments with Lorenz 1996 model. Link
- 17:00-17:30 A new look at the ensemble Kalman Filter: Duality and non-asymptotic analysis. Link
- 17:30-18:00 How does the regression step in the two-step EnKF connect to Bayesian estimation? Link