We're kicking off this Volcano Risk Lunch with a discussion that follows on from Tom Sheldrake's expert elicitation session.

Discussion Topic

How to use Bayesian Belief Networks for forecasting volcanic activity on Montserrat: constructing and constraining a BBN model (Including preliminary results from the recent Volcano 'Dopplegangers' Expert Elicitation session)

Presenters

Tom Sheldrake, Henry Odbert

Future scheduling and topics

Having distilled some of the ideas for Volcano Risk Lunch, we've decided to attempt the following format:

• One or two people per session select a topic - a paper of interest, some recent results, a science problem, new software etc.
• They prepare some material that will lead a discussion. This does not necessarily have to be in the form of a powerpoint presentation, but might involve extracting some figures/ data/ software, etc, to discuss.
• We will meet about once per month, on a Tuesday at 12 pm for max. 1 hr. We can allow scheduling to evolve, but it's assumed that not all of us will be available to attend every session. Times to be announced via this list.
• Presenters/ topics will be voluntary but we'll assume this duty will rotate around most of the group. To volunteer a topic, contact Henry Odbert or Susanna Jenkins.

Note: This event is for internal staff/students only.