=============================================================================
Royal Astronomical Society Discussion Meeting:
GEOLOGICAL PRIOR INFORMATION
9:30am, May 9th, 2003
Burlington House, London
Announcement and call for Speakers and Posters
------------------------------------------------------------------------
If you would be interested to present at this meeting, please respond to the addresses given below. Please include a title and an abstract (< 1
page), state whether a talk or a poster would be preferred, and state whether or not you are a fellow of the RAS.
MEETING TOPIC
Geoscientists infer models of the Earth from physical, chemical or biological measurements. Such inferences are inevitably non-unique, and
geological or intuitive 'reasonableness' is often used to narrow the range of possible models. However, this latter operation is, however, usually
non-quantitative and is often overly optimistic since it is difficult to quantify the degree of constraint offered.
The power of computation now available allows the parameterisation of geological knowledge intelligently, quantitatively and objectively prior to
conducting more specific geoscientific investigations. The range of Earth models considered in geoscientific investigations may thus be limited a
priori to those likely to occur, providing realistic, quantitative constraints independent of any new data being considered. Uncertainty on the final
results may also be quantified within the limitations of the basic assumptions made, and if identical, quantitative prior information is used in
different studies then the results of those studies may more readily be compared.
The aim of this meeting will be to investigate both the range of geoscientific fields within which geological prior information can be quantified and
used to pre-condition geoscientific investigations, and techniques required to achieve this.
Interesting topics within this area include:
o Capture and parameterisation of geological prior information
- how do we achieve this quantitatively?
o Herding
- many researchers will have followed leaders who may in turn
be wrong; how do we assess the quality of prior information?
(c.f. bubble economies).
o Model uncertainty reduction
- to what extent might model estimates benefit from this approach?
o Common prior information
- Is it possible that in the future different studies will all
use similar (or identical) prior information? If so, how can we
facilitate this?
Requests to participate should be sent to:
ANDREW CURTIS
[log in to unmask] or [log in to unmask]
Department of Geology and Geophysics, Edinburgh University,
Grant Institute, West Mains Rd., Edinburgh EH9 3JW
or Schlumberger Cambridge Research, High Cross, Madingley Road,
Cambridge CB3 0EL
and
RACHEL WOOD
[log in to unmask] or [log in to unmask]
Department of Earth Sciences, Cambridge University, Downing Street,
Cambridge CB2 3EQ
or Schlumberger Cambridge Research, High Cross, Madingley Road,
Cambridge CB3 0EL
==========================================================================
|