Disputas: Osvanny Ramos Rosales

Prøveforelesning

Bedømmelseskomité

Professor Eric Clement, École Supérieure de Physique et de Chimie Industrielles (ESPCI), Paris
Professor Kim Sneppen, Niels Bohr Instituttet, København
Professor Yuri Galperine, Universitetet i Oslo, Oslo

Leder av disputas:  Ørjan G. Martinsen

Veileder:  Knut Jørgen Måløy og Eirik Grude Flekkøy

Sammendrag

It is a common belief that a large earthquake is always the result of a small one that did not stop growing. This growing process can occur at any time, without warning, making no difference between small and large events and therefore dooming prediction. The same is believed to apply to other phenomena with scale invariant (i.e., power law distributed) avalanches. Nevertheless, recent experiments and numerical results demonstrate the possibility of predicting large events in phenomena ruled by power law distributed avalanches.

In a collaboration between the Complex group at the Physics Department of the University of Oslo and the Physics Faculty of University of Havana, O. Ramos (Complex) and collaborators have analyzed the avalanche dynamics in a simple pile of spherical beads. In spite of its simplicity, the pile's dynamics is rather complicated, showing avalanches with no correlation between the events, and no characteristic avalanche sizes (they follow a power law distribution). This resembles the behavior of many natural and social phenomena like earthquakes, solar flares, stock markets, etc; where the prediction of large events is extremely important. The experiment has shown characteristic times between large avalanches, suggesting the possibility of long-term prediction; and large avalanches are also preceded by continuous, detectable variations in the internal structure of the pile. So, by monitoring these variations it is possible, in principle, to predict when a large avalanche is going to happen.

The previous findings are generalized through numerical simulations in a classic earthquake model, where results similar to those in the experiment have been obtained. Other experiments in the "pile of grains" setup have verified some prediction made by the earthquake model, bridging the gap between those two - at first glance very different - systems.

Kontaktperson

For mer informasjon, kontakt Heidi Bruvoll.