BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:On the quantification of parametric uncertainty in Antarctic ice s
 heet model projections\, including the role of bedrock topography. - Andre
 as Wernecke (The Open University)
DTSTART:20191004T100000Z
DTEND:20191004T110000Z
UID:TALK131062@talks.cam.ac.uk
CONTACT:Dr Maria Vittoria Guarino
DESCRIPTION:Century scale predictions of the Antarctic mass balance have l
 arge uncertainties. Those uncertainties can be separated into systematic m
 odel uncertainty and parametric uncertainty\, the latter being caused by p
 oorly observed model parameters which cause a spread in predictions depend
 ing on the chosen value. While model sensitivities to uncertain parameters
  have been the subject of a number of studies\, there is very limited lite
 rature on quantifying predictive ice sheet uncertainties. For such uncerta
 inty quantifications\, sensitivities have to be combined with probabilisti
 c parameter interference so that estimates of how the ice sheet would deve
 lop for certain parameter values is joint by estimates of how likely those
  parameter values are to be 'correct'. \n\nI will present different Antarc
 tic ice sheet model projections with focus on uncertainty quantification. 
 First we use statistical modelling to emulate the DeConto and Pollard (201
 6) ensemble. By doing so we are able to better understand the model depend
 encies\, update the calibration and are able to derive new probability dis
 tributions. We will further adopt a probabilistic spatial calibration appr
 oach to high resolution BISICLES data from the Amundsen Sea Embayment (ASE
 ). This allows us to explore the potentials\, and statistical challenges o
 f exploiting the two-dimensional spatial pattern in ice thickness change t
 o improve sea level rise predictions. After that we will address the quest
 ion of how well the bedrock topography at Pine Island Glacier is known for
  the use in ice sheet models and are happy to discuss whether this could b
 e developed into a tool for future mission planning.\n\n
LOCATION:British Antarctic Survey\, Seminar Room 2
END:VEVENT
END:VCALENDAR
