BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Developing data-driven models to emulate GCM’s - Rachel Furner\,
  University of Cambridge/ British Antarctic Survey
DTSTART:20200121T120000Z
DTEND:20200121T130000Z
UID:TALK137254@talks.cam.ac.uk
CONTACT:Jonathan Rosser
DESCRIPTION:Chair: Scott Hosking\nAbstract: Climate models represent the b
 est tools we have to predict\, understand and potentially mitigate climate
  change\, however these process-based models are incredibly complex and re
 quire huge amounts of high-performance computing resources. Machine learni
 ng offers opportunities to greatly improve the computational efficiency of
  these models. Here we discuss preliminary work looking to develop analogo
 us data-driven models.\n\n\nThe talk will focus on two separate pieces of 
 work:\n\n\nWe begin by presenting a neural network capable of replicating 
 the behaviour of Lorenz model (an idealised dynamical system which include
 s chaos). This is used as a test-bed to assess the importance of the time-
 stepping method used in network based models of dynamical processes\, and 
 the loss functions used to train the model.\n\n\nSecondly we present very 
 preliminary results from work developing a linear regressor to emulate an 
 idealised general circulation model of the ocean. The regressor is trained
  using the outputs from an idealised sector configuration of general circu
 lation model (MITgcm). Our aim is to develop an algorithm which is able to
  predict the future state of the model to a similar level of accuracy. Som
 e results from investigations into the sensitivity of the regressor to var
 ious inputs (e.g. temperature on different spatial and temporal scales\, a
 nd meta-variables such as location information) will be presented.\n\n\nWo
 rk with supervisors: Peter Haynes (University of Cambridge)\, Dan Jones (B
 ritish Antarctic Survey)\, Dave Munday (British Antarctic Survey)\, Brooks
  Paige (UCL)\, Emily Shuckburgh (University of Cambridge)\n\n\n
LOCATION:Bullard Lab\, Seminar Room
END:VEVENT
END:VCALENDAR