HR: 0830h
AN: NG31A-0608 [PDF]
TI: Signatures of Self-Organized Criticality as System Forcing Changes
AU: * Woodard, R
EM: ryan@timehaven.org
AF: University of Alaska Fairbanks, PO Box 755920, Fairbanks, AK 99775-5920 United States
AU: Newman, D
EM: ffden@uaf.edu
AF: University of Alaska Fairbanks, PO Box 755920, Fairbanks, AK 99775-5920 United States
AU: Sanchez, R
EM: rsanchez@fis.uc3m.es
AF: Universidad Carlos III de Madrid, C/ Madrid 126
Getafe, Madrid, 28903
Spain
AU: Carreras, B
EM: carreras@FED.ORNL.GOV
AF: Oak Ridge National Lab, P.O. Box 2008, Oak Ridge, TN 37831 United States
AB:
Because of the broad claims of applicability made for self-organized criticality (SOC), it is important to define measurable
characteristics which can validate or invalidate a statement that a system or model is SOC. We examine two such measures,
spectral and R/S analysis, on a SOC sandpile model with varying driving rate and very long time records and find that: 1) the
length of the data record can be crucial in distinguishing SOC from random dynamics, 2) R/S analysis gives a more consistent
indication of the dynamics (whether correlated or not) in a SOC system than does the power spectrum, 3) a 1/f region in the
SOC power spectrum does not arise from just a random superposition of events (pulses) from a power law distribution and 4)
the 1/f spectrum can be substantially altered by keeping the same events and simply changing the spacing between them. The
first result implies that SOC should not necessarily be discounted as a model for a geophysical process whose data do not
match SOC signatures, since many such processes have long time scales that preclude the collection of relevantly long time
series. The second result is an argument for including R/S analysis in one's standard toolbox when characterizing the
dynamics of any physical system. The final two results imply a test to see if a process with a 1/f signal is possibly due to
SOC.
DE: 3220 Nonlinear dynamics
DE: 3250 Fractals and multifractals
DE: 7260 Theory and modeling
DE: 7839 Nonlinear phenomena
SC: Nonlinear Geophysics [NG]
MN: 2003 Fall Meeting