|
Back to Colloquia
Physics Colloquium Friday, Apr. 25th, 2008,
4:00 P.M.
E300 Math/Science
Center; Refreshments at 3:30 P.M. in
Room E200
Complexity Science Group
Department of Physics and Astronomy
University of Calgary
Scaling, universality and spatio-temporal clustering
in seismicity and rock fracture phenomena
Many striking features of natural processes
including seismicity can be portrayed as patterns or
clusters of localized events. A generic attribute in
all these cases is that one event can trigger or
somehow induce another one to occur - or possibly
numerous further events. Sometimes, an accounting of
causal connections between clustered events is
explicitly rationalized by the microscopic state and
rules of the dynamical system. More often than not,
though, the causal connections cannot be resolved
from the data at hand and remain ambiguous. Thus,
one is confronted with inferring a plausible causal
structure from clusters of localized events without
a detailed or "fundamental" knowledge of the true
microscopic dynamics. I will present a method to
search for such signs of causal structure in
spatio-temporal data making minimal a priori
assumptions about the underlying microscopic
dynamics. To this end, the elementary concept of
recurrence for a point process in time is
generalized to recurrent events in space and time.
For earthquakes, the method in particular allows to
recover the scaling of the rupture length with
magnitude.
Moreover, I will present a detailed statistical
analysis of acoustic emission time series from a
range of rock fracture experiments. In all
considered cases, the waiting time distribution can
be described by a unique scaling function indicating
its universality. This scaling function is even
indistinguishable from that for earthquakes
suggesting its general validity for fracture
processes independent of time, space and magnitude
scales.
|
|