You are here

Probabilities for Rare Events (E.g., Large Explosive Volcanic Eruptions)

Jonathan Rougier, University of Durham
12-1pm  5th Feb 2020

Abstract

The Exceedance Probability (EP) of a hazard shows the probability of an event exceeding a specified size, in a specified future interval.  For example, if Merapi's 1-year EP for erupting 1Gt of mass were 0.001, then there would be a 0.001 probability of Merapi having an eruption exceeding 1Gt of mass in the next year. EP curves are crucial in hazard and risk assessment.  Unfortunately, though, there is no reason to think that they are 'smooth'.  For example, the eruptive behaviour of a volcano is full of non-linearities and phase-changes.  How, then, can we map a sparse historical dataset into an EP curve?  And how do we quantify our confidence in such an assessment?  There are no canonical answers to these questions, but I show how probability and statistics can help.

Short Bio

Jonathan Rougier is an applied statistician who works across the full range of theoretical, applied, and computational statistics.  His original training was in Economics (up to PhD level). After graduating from the University of Durham with a 1st class BSc in Economics (1988), he worked for two years as an economist for a large fund-management company.  He then returned to Durham as a Lecturer in Economics (1990), completing a PhD in 1996.

In 1997 he transferred to Mathematics at Durham to work as a Postdoctoral Researcher with Prof Michael Goldstein, in the area known as ‘Design and Analysis of Computer Experiments’.  In 2007 he took up a Lectureship in the School of Mathematics at the University of Bristol, being promoted to Reader in 2012 and to Professor in 2016 (Professor of Statistical Science).  Since coming to Bristol he has collaborated extensively across the Faculties of Science and Engineering, including in volcanology, glaciology and palaeoclimate reconstruction, flooding, and in the more general area of uncertainty and risk assessment.  He has worked with several UK Government departments and agencies, and also with non-profit organizations and SMEs.

Venue

Large Conference Room, O'Reilly Institute