[CIG-ALL] 2018 AGU session DI23: Towards success in geodynamic modeling - Learning from and improving on model iterations
rene.gassmoeller at mailbox.org
Tue Jul 17 13:18:57 PDT 2018
You are invited to submit an abstract to DI session “Towards success in
geodynamic modeling - Learning from and improving on model iterations”
at the upcoming AGU Fall Meeting taking place in Washington DC on
December 10-14, 2018. We feel that members of the CIG community
have the breadth of work that would allow for an exciting
discussion on the topic.
Please note that in order to facilitate extensive exchange and
discussion this is an eLightning session and opposed to previous years
there are no additional fees for eLightning abstracts.
Confirmed invited presenters:
Taras Gerya (ETH)
Marc Spiegelman (Columbia University/LDEO)
Session full description and submission:
Note that the abstract submission deadline is August 1st, 2018.
Shi Joyce Sim (Carnegie/DTM)
Rene Gassmoller (UC Davis)
Adina Pusok (Scripps Institution of Oceanography)
DI023: Towards success in geodynamic modeling - Learning from and
improving on model iterations
Numerical and analog models form a crucial part of our theoretical
understanding of the Earth and they typically require careful
implementation to be successful. Published model results rarely mention
the attempts that were unsuccessful, therefore, burying the important
conclusions that could be drawn from these 'failed' models, or lessons
that could be learned about how to 'fail' best. This is particularly
true for geodynamic modeling, where the combination of assumptions,
complex methods, and complicated results makes the creation of an
informative model a tedious task.
We invite contributions (pertaining but not limited to core, mantle,
lithosphere and planetary dynamics) of unexpected, controversial, failed
and/or negative model results. In particular, we encourage (1)
conclusions from unexpected results, (2) the critical mindset when
evaluating models, (3) best practices to ensure reliability and
reproducibility, (4) the inherent limitations of models, and (5)
strategies to cope with repeated disappointments of unsuccessful models.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the CIG-ALL