BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Is sea ice modeling rocket science? - Bruno Tremblay (McGill Unive
 rsity)
DTSTART:20220921T100000Z
DTEND:20220921T103000Z
UID:TALK178265@talks.cam.ac.uk
DESCRIPTION:\nIn addition to simulating sea ice thermodynamic processes\, 
 a&nbsp\;sea ice dynamical model is necessary for a realistic&nbsp\;represe
 ntation&nbsp\;of the spatial distribution of ice thickness and concentrati
 on\, and (solid) freshwater transport from the Arctic to the northern Nort
 h Atlantic&nbsp\;in for instance Global Climate Models. This can be accomp
 lished&nbsp\;by&nbsp\;using any parameterizations of ice-ice interactions&
 nbsp\;in the dynamical model&nbsp\;that include some transition from low d
 eformation to large deformation (or fracture) when critical internal ice s
 tresses in shear and compression are reached. Of equal importance&nbsp\;is
  a realistic simulation of sea ice deformation (shear and divergence) alon
 g the fracture lines\, called Linear Kinematic Features (LKFs)\, where moi
 sture\, salt and heat between the surface ocean and the atmosphere are exc
 hanged. LKFs are both intermittent in time and localized in space and conj
 ugate pairs of fracture lines intersect over a specific range of angles. T
 he source of this intermittency and localization however remains unknown: 
 yet again\, any model that considers a transition from low deformation to 
 large deformation when critical internal stresses in shear and compression
  are reached reproduce these fractal properties (at the scale they are cur
 rently resolved). And while the angle between fracture lines follows theor
 y for rheological models with a normal flow rule (alignment&nbsp\;of stres
 s and strain invariants axes)\, they no longer follow theory for models wi
 th a non-normal flow rule (appropriate for real sea ice) in simple uniaxia
 l loading experiment nor in realistic pan-Arctic simulations irrespective 
 of the rheological&nbsp\;model&nbsp\;used. The sensitivity of the spatial 
 and temporal scaling to ad hoc parameterizations or poorly constrained par
 ameters such as damage or shear and compressive strength\, and the relativ
 e importance of ice-ice interactions compared to the first order balance o
 f surface wind stress and water drag also point to a lack of understanding
  of first order processes governing sea ice deformations. All in all\,&nbs
 p\;current sea-ice dynamical models are sensitive to poorly constrained pa
 rameters and a better understanding of the questions posed above will be n
 ecessary before we can propose a rheological model that&nbsp\;reproduces a
 ll key observed sea-ice features.&nbsp\;&nbsp\;&nbsp\;\n\n&nbsp\;
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR