BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20210916T132453Z
LOCATION:Jean-Jacques Rousseau
DTSTART;TZID=Europe/Stockholm:20210708T170000
DTEND;TZID=Europe/Stockholm:20210708T173000
UID:submissions.pasc-conference.org_PASC21_sess196_msa326@linklings.com
SUMMARY:Data Assimilation for a 3D Hydrodynamic Model of Lake Geneva Combi
 ning Novel Particle Filtering and Machine Learning Algorithms
DESCRIPTION:Minisymposium\n\nData Assimilation for a 3D Hydrodynamic Model
  of Lake Geneva Combining Novel Particle Filtering and Machine Learning Al
 gorithms\n\nSafin, Bouffard, Sukys, Ozdemir, Runnalls\n\nAs part of the DA
 TALAKES project, we develop a comprehensive data assimilation framework fo
 r a 3D hydrodynamic model of Lake Geneva. We use the SPUX package, which e
 nables Baeysian inference and particle Markov Chain Monte Carlo methods fo
 r uncertainty quantification. For data assimilation, the filtering algorit
 hm uses trajectory resampling, resulting in physically consistent predicti
 ons. This approach allows highly frequent assimilation of the various data
  sources, and improves predictions by 10%-30%. We deploy a Bidirectional L
 ong Short-Term Memory (Bi-LSTM) machine learning algorithm to predict skin
  temperature from the bulk, which enables the assimilation and uncertainty
  quantificaiton of lake surface water temperature (LSWT). We show that our
  model is capable of reproducing observations well, and discuss some of th
 e advantages and limitations of our particle filtering approach.\n\nDomain
 : CS and Math, Emerging Applications, Climate and Weather
END:VEVENT
END:VCALENDAR