BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Improving models of ocean turbulence with data-driven methods - Sa
 m Lewin (DAMTP\, University of Cambridge)
DTSTART:20211028T120000Z
DTEND:20211028T130000Z
UID:TALK165076@talks.cam.ac.uk
CONTACT:Christopher Edsall
DESCRIPTION:"MS Teams link":https://teams.microsoft.com/l/meetup-join/19%3
 ameeting_YmZiYzQ1MWUtZWJhZS00MTVkLWIwZWEtMTUyN2IzMDUyNDYz%40thread.v2/0?co
 ntext=%7b%22Tid%22%3a%2249a50445-bdfa-4b79-ade3-547b4f3986e9%22%2c%22Oid%2
 2%3a%22c57afeee-d613-4952-b196-6a1063994568%22%7d\n\nSmall-scale turbulent
  mixing in the ocean is of fundamental importance for the vertical transpo
 rt of heat\, carbon\, nutrients and other properties that influence global
  energy budgets and biological activity. We propose a new probabilistic ma
 chine learning method for computing energy dissipation rates that characte
 rise this turbulence from vertical profiles of velocity and density gradie
 nts\, training and testing our model on numerical simulations of decaying 
 turbulence designed to reproduce conditions found in oceanic flows. Our mo
 del outperforms existing theoretical models widely used in oceanographic p
 ractice\, and additionally offers some insight into the underlying physics
 . 
LOCATION:Virtual Meeting\; MS Teams link in description
END:VEVENT
END:VCALENDAR