Dienstag, 31. Mai 2022, 16:30 - 18:00 iCal

Vortrag von Rüdiger BRECHT

(Memorial University of Newfoundland, St. John's, Canada)


"Improving Lagrangian particle models using deep learning"


Vortrag im Rahmen des Meteorologisch-Geophysikalischen Kolloquiums.

Room 2F513, (Exner-Raum)
Josef-Holaubek-Platz 2, UZA 2, 1090 Wien


Abstract: The resolution (spatial or temporal) of meteorological fields (e.g. wind, temperature, geopotential) provided by forecasting centres is not sufficient for general particle transport models. Currently, the intermediate fields are obtained through linear interpolation. Here, we demonstrate different neural network architectures, to generate higher resolutions of meteorological fields and outperform the linear interpolation. Preliminary results are presented.


Zur Webseite der Veranstaltung


Institut für Meteorologie und Geophysik


Andreas Plach/Herta Gassner
Institut für Meteorologie und Geophysik