BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Computing (generalised) Lagrangian mean without tracking particles
  - Hossein Kafiabad\, Durham University
DTSTART:20230313T130000Z
DTEND:20230313T140000Z
UID:TALK196066@talks.cam.ac.uk
CONTACT:Prof. John R. Taylor
DESCRIPTION:Lagrangian averaging plays an important role in the analysis o
 f wave--mean-flow interactions and other multiscale fluid phenomena. Compa
 ring to its Eulerian counterpart Lagrangian averaging has several superior
 ities. For instance\, it removes the doppler shift of wave frequency by st
 rong background flow\, which eclipses the separation of time scale between
  them. Another advantage is that the Lagrangian mean fields usually inheri
 t the material conservation laws (such as the conservation of PV\, circula
 tion\, and magnetic field) that hold for instantaneous fields. The numeric
 al computation of Lagrangian means\, e.g. from simulation data\, is howeve
 r challenging. Typical implementations require tracking a large number of 
 particles to construct Lagrangian time series which are then averaged. Thi
 s has drawbacks that include large memory demands\, particle clustering an
 d complications of parallelisation. We develop a novel approach in which t
 he Lagrangian means of various fields (including particle positions) are c
 omputed without tracking particles in time. This approach leads to a set o
 f PDEs that is integrated over successive averaging time intervals. The PD
 Es can be discretised in a variety of ways\, e.g. using the same discretis
 ation as that employed for the governing dynamical equations\, and solved 
 on-the-fly to minimise the memory footprint.
LOCATION:MR5\, CMS
END:VEVENT
END:VCALENDAR