Madison Chaos and Complex Systems Seminar

Spring 2005 Seminars

All seminars are Tuesday at 12:05 pm in 4274 Chamberlin except as noted.

Short List

Join us for lunch during the summer on the Union Terrace at noon each Tuesday, starting May 10th!


January 18, 2005

Self-organization - Nature's intelligent design

Clint Sprott, UW Department of Physics

Complex patterns are common throughout nature, from the distribution of the galaxies in the Universe to the organization of neurons in the human brain. It is generally assumed that such complex structure must have a complex cause, but it may be that simple mathematical and computer models can produce such patterns. This talk will provide examples of self-organization in nature and will describe six simple models that can replicate the features of these patterns. The models typically produce fractal spatial structure and chaotic temporal dynamics characterized by power laws and unpredictability, even when the models are simple and purely deterministic. The work has application to fields as diverse as physics, ecology, political science, economics, sociology, and art.

This talk is available as a PowerPoint Presentation and in HTML format.

January 25, 2005

Magnetic self-organization in laboratory and astrophysical plasmas

Jim Reardon, UW Department of Physics

A plasma is an ionized gas that contains a very large number of charged particles.  The law that governs the movement of an individual charged particle is easy to write down, and has been known for more than 100 years.  However, experiments have consistently shown that plasma behavior on all length and time scales is dominated by collective effects, few of which were predicted in advance.  Recently, it has become clear that the behaviors of certain plasmas in the laboratory and in space are dominated by magnetic self-organization, a type of collective behavior in which the movements of particles are influenced by the fluctuations in magnetic fields, and vice-versa.  Examples include astrophysical jets and the MST device in Chamberlin Hall.

February 1, 2005

Climate Variability in the Upper Atmosphere

Susan Nossal, UW Department of Physics

 Understanding the influence of sources of natural variability is necessary for characterizing properties of the Earth's atmosphere and for isolating signatures of natural variability from those due to human caused change. Scientists are investigating the Earth's climate by studying patterns of variation both geographically and vertically in the atmosphere. This talk will focus on studies of higher altitude regions of the atmosphere. For example, there is predicted to be major greenhouse cooling in the stratosphere and higher altitudes, accompanying warming in the troposphere as a result of increased atmospheric concentrations of carbon dioxide. There are also expected to be changes in the chemical composition of the atmosphere at higher altitudes. This talk will address in more detail ongoing research of sources of natural variability in upper atmospheric hydrogen and predictions for change in its composition. Upper atmospheric hydrogen is a byproduct of chemistry in the middle and lower regions of the atmosphere involving hydrogen-containing species such as methane and water vapor, two of the most influential greenhouse gases.

February 8, 2005

A dynamics-based approach to sound source identification

Robert Lutfi, UW Department of Communicative Disorders

Everyday experience suggests that the ear is rather good at inferring basic attributes of objects and events from sound. The 'clunk' of the table struck with your knuckle indicates that the table is hollow and made of wood; the 'woosh' of the passing vehicle suggests a high speed chase. Such simple examples represent in principle how we use sound to gain information about our surroundings, but even in regard to these simple examples we understand very little of the process. In this talk I'll describe a method for measuring precisely how listeners judge the physical attributes objects and events from sound. The method involves estimating partial correlations of the listener's judgments with small, experimentally-introduced perturbations in acoustic parameters. Importantly, the perturbations are introduced in such a way so as not to violate the dynamic relations governing sound emission. One notable conclusion of these studies is that basic limits in auditory sensitivity exert a far greater influence on listener judgments than does knowledge of the intrinsic acoustic relations that would serve to disambiguate sources.

February 15, 2005


Ann Althouse, UW Law School

UW law professor (and prolific blogger) Ann Althouse will discuss the burgeoning phenomenon of time-stamped, frequently updated, on-line journals.

See a report on this talk in the Madison Capitol Times, and visit the speaker's blog.

February 22, 2005

Numerical simulations of the Madison Dynamo Experiment

Adam Bayliss, UW Department of Physics

Saturation of a mean field magnetic dynamo via backreaction with flow is described for increasing fluid  turbulence. Our dynamically and  inductively self-consistent incompressible MHD code utilizes a propellor model for a mechanically driven flow which qualitatively compares well to LDV measurements in a purely hydrodynamic analogue of the liquid sodium experiment in Madison. We find that for a laminar flow the exponentially growing magnetic field is saturated when the Lorentz force alters the shape and strength of the flow until a steady state is reached where the growth rate of the large scale saturated magnetic field is zero or producing damped  mild oscillations about a steady state magnetic energy; both strength and character of the nonlinear saturation vary with magnetic Reynolds number. As the fluid Reynolds number, thus the mean flow, is increased via  more power into the simulated propellors, inertial terms create increasingly nonlinear flows leading to strong magnetic fields which re-laminarize the fluid flow. Recent simulations examine the role of turbulence in the anisotropic, inhomogeneous spherical system. We observe a small-scale dynamo which may inhibit the growth of the large-scale magnetic field, despite being kinematically predicted (using a fixed flow averaged over temporally correlated fluctuations). Results of the simulations will numerically investigate the growth, saturation, and turbulent evolution of the liquid sodium dynamo experiment in Madison.

March 1, 2005

Continuous measurements of crustal deformation through the earthquake cycle, western Mexico: Evidence for non-linear crustal response following megathrusting subduction earthquakes

Chuck DeMets, UW Department of Geology and Geophysics

The classical notion of faults as features that exhibit simple stick-slip behavior, represented by earthquakes and intervening aseismic periods of strain accumulation, is discussed in view of mounting evidence that other non-seismic phenomena exert important influences on the seismic cycle of active faults. Evidence for such phenomena will be presented from continuous GPS measurements in western Mexico and other places.

March 8, 2005

Effects of biodiversity on the functioning of ecosystems ... when, why, and who cares?

Brad Cardinale, UW Department of Zoology

Abstract: Loss of biological complexity -- due to species extinction as well as homogenization by the spread of exotic taxa -- ranks among the most striking changes to the global environment.  It has been argued that ecosystems characterized by lower biocomplexity may be less efficient at performing fundamental biological processes - processes that may ultimately prove important to society.  Using 3 case studies as examples, I will address several conditions under which diversity does vs. does not influence key fluxes of energy and matter.  These will focus on the impacts of (i) freshwater invertebrates on fluid dynamics, (ii) marine benthic invertebrates on sediment bioturbation, and (iii) terrestrial invertebrate predators on pest control.

March 15, 2005

The dynamics of host-parasite coevolution and the evolution of recombination

Andy Peters, UW Department of Zoology

Antagonistic coevolution (also known as the "Red Queen process"), in which biological enemies such as hosts and parasites are under constant selection to evade each other's recent adaptations, has long been known to provide a potential advantage to the shuffling of genomes that results from genetic recombination.  I present the results of computer simulations showing that the dynamics of this process make it qualitatively different from any other model of the evolution of recombination.  I also consider the feedback between the increase in recombination and the process of coevolution, and show that the mechanisms by which host populations respond to parasite selection change fundamentally as recombination becomes more common.

March 29, 2005

Dynamics of protein molecules

George Phillips, UW Departments of Biochemistry and Computer Sciences

Proteins molecules comprise thousands of atoms that self assemble into working machines. The basic physical forces that drive them to a folded structure are generally understood, but the energy landscape of their possible configurations is rich and complex. Aspects of protein folding, such as metastable states and prions will be discussed as well as attempts to visualize these dynamic systems using simulations and experiments alike.

April 5, 2005

Sleep apnea: from obscurity to epidemic in just a few decades

Terry Young, UW Department of Population Health Sciences

Until the early 1990’s, little was known about sleep apnea, a condition of repeated breathing pauses during sleep. We initiated the early epidemiological studies of this disorder by recruiting a random population sample of men and women to undergo overnight sleep studies every four years: the Wisconsin Sleep Cohort (WSC). The studies based on the WSC, now in its 17th year, uncovered a rather shockingly high prevalence of undiagnosed sleep apnea across a wide severity spectrum in both men and women. Most importantly, there is a dose-response (but nonlinear!) association between sleep apnea and significant behavioral and cardiovascular morbidity. The dilemma lies in the lack of clinical resources to deal with diagnosis and treatment of the huge number of people with unrecognized sleep apnea. Where, on the severity spectrum should the “clinical significance” cut-point be? Current findings from the WSC relevant to this question will be presented.

April 12, 2005

Influence of habitat type and field shape on margin weed diversity

Ed Luschei, UW Department of Agronomy 

Despite the tremendous economic costs associated with weeds, surprising little is known about their population dynamics. The primary research focus of applied weed scientists has been on fine tuning our ability to decrease weed seedling survivorship through cultural practices, mechanical disturbance and chemical application. While this focus makes sense as a means to minimize crop yield lost to resource competition, it clearly must be considered tending the proverbial symptom rather than the disease. I will briefly discuss the traditional stage-structured formulation of weed population dynamics and why it has largely failed to helpful in an applied sense. In short, I argue that stochasticity and scale conspire to make most experiments relatively uninformative. As an alternative to the “bottom up” construction of dynamics, I will describe several experiments my lab has conducted that more explicitly address the stochastic nature of processes of potential importance in weed dynamics.

April 19, 2005

Complex dynamics in spatially-dependent Lotka-Volterra ring systems

 Joe Wildenberg, UW Department of Physics

Mathematical models often need to incorporate spatial dependence to accurately model real-world systems.  We consider competitive Lotka-Volterra models modified to include this spatial dependence through organization of the competing species into a one-dimensional ring by an appropriate choice of the interaction matrix.  We show that these systems can exhibit complex dynamics, spatiotemporal chaos, and spontaneous symmetry breaking. A high-dimensional, spatially homogeneous, nearest-neighbor example with interaction strengths decreasing with distance is characterized including an analysis of how the dynamics of the system vary with dimension.  We also show the existence of Lyapunov functions that arise from this method of including spatial dependence and how they prohibit complex dynamics for certain regions of the parameter space.

This talk is available as a PowerPoint presentation.

April 26, 2005

Entropy, symbolic dynamics, and dynamical systems

Jeff Noel, UW Department of Physics

We all know entropy as that mysterious physical necessity that hot objects always become cool and cool objects never spontaneously become hot. Increasing entropy, or more noticeably increasing disorder, is the Second Law of Thermodynamics. Boltzmann showed that the Second Law was really only a statistical likelihood, that manifested as a law for macroscopic systems. Claude Shannon used the same mathematical formulation as Boltzmann to define the entropy of information and communication. This formulation allows for a quantification of the disorder of information and gives several important results including the value of the maximum possible compression of data. This same technique when applied to dynamical systems and their associated symbolic dynamics allows for investigation of the disorder of the systems and gives alternative methods to calculate quantities such as the Lyapunov exponent and the predicability time. In this talk, entropy, information theory, and the symbolic dynamics of a low dimensional chaotic Lotka-Volterra system are discussed.

May 3, 2005

Physics of running

Jim Reardon, UW Department of Physics

I would like to demonstrate that running (the sort that humans do, for example when they are late to the Chaos and Complex Systems Seminar) is both very much more simple and very much more complex than is commonly thought.  "Simple" in that it is possible to build anthropomorphic robots, with gaits quite like human gaits, that have almost no control systems but nonetheless have gross dynamic stability, and "complex" in that humans achieve fine dynamic stability by means of a much more sophisticated control system which is almost always "user-transparent".  Possible implications of all this for lowering your 5k time will be discussed.