HR: 0800h
AN: ED51A-0005    [Abstracts]
TI: Capturing Intuition Through Interactive Inverse Methods: Examples Drawn From Mechanical Non-Linearities in Structural Geology
AU: * Moresi, L
EM: louis.moresi@sci.monash.edu
AF: Monash Cluster Computing, School of Mathematical Sciences Building 28 Monash University, Clayton, Vic 3800 Australia
AU: May, D
EM: david.may@maths.monash.edu.au
AF: Monash Cluster Computing, School of Mathematical Sciences Building 28 Monash University, Clayton, Vic 3800 Australia
AU: Peachey, T
EM: Tom.Peachey@infotech.monash.edu.au
AF: Monash Cluster Computing, School of Mathematical Sciences Building 28 Monash University, Clayton, Vic 3800 Australia
AU: Enticott, C
EM: Colin.Enticott@csse.monash.edu.au
AF: Distributed Systems Technology Centre, Building C Monash University, Caulfield, VIC 3145 Australia
AU: Abramson, D
EM: davida@csse.monash.edu.au
AF: Monash Cluster Computing, School of Computer Science and Software Engineering Building 75 Monash University, Clayton, VIC 3800 Australia
AU: Robinson, T
EM: t.robinson3@pgrad.unimelb.edu.au
AF: Monash Cluster Computing, School of Mathematical Sciences Building 28 Monash University, Clayton, Vic 3800 Australia
AB: Can you teach intuition ? Obviously we think that this is possible (though it's still just a hunch). People undoubtedly develop intuition for non-linear systems through painstaking repetition of complex tasks until they have sufficient feedback to begin to "see" the emergent behaviour. The better the exploration of the system can be exposed, the quicker the potential for developing an intuitive understanding. We have spent some time considering how to incorporate the intuitive knowledge of field geologists into mechanical modeling of geological processes. Our solution has been to allow expert geologist to steer (via a GUI) a genetic algorithm inversion of a mechanical forward model towards "structures" or patterns which are plausible in nature. The expert knowledge is then captured by analysis of the individual model parameters which are constrained by the steering (and by analysis of those which are unconstrained). The same system can also be used in reverse to expose the influence of individual parameters to the non-expert who is trying to learn just what does make a good match between model and observation. The ``distance'' between models preferred by experts, and those by an individual can be shown graphically to provide feedback. The examples we choose are from numerical models of extensional basins. We will first try to give each person some background information on the scientific problem from the poster and then we will let them loose on the numerical modeling tools with specific tasks to achieve. This will be an experiment in progress - we will later analyse how people use the GUI and whether there is really any significant difference between so-called experts and self-styled novices.
UR: http://www.mcc.monash.edu/Codes/NimrodOI
DE: 8010 Fractures and faults
DE: 8020 Mechanics
DE: 0845 Instructional tools
DE: 0850 Geoscience education research
SC: Education and Human Resourcese [ED]
MN: 2004 AGU Fall Meeting