Source author record

Spencer A. Smith

Spencer A. Smith appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record

nlin.CD physics.flu-dyn math-ph math.DS math.MP physics.comp-ph

Catalog footprint

What is connected

4works

6topics

4close collaborators

Actions

Connect this record

Open graph Browse works

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2016arXiv

Reynolds number effects on mixing due to topological chaos

Topological chaos has emerged as a powerful tool to investigate fluid mixing. While this theory can guarantee a lower bound on the stretching rate of certain material lines, it does not indicate what fraction of the fluid actually participates in this minimally mandated mixing. Indeed, the area in which effective mixing takes place depends on physical parameters such as the Reynolds number. To help clarify this dependency, we numerically simulate the effects of a batch stirring device on a 2D incompressible Newtonian fluid in the laminar regime. In particular, we calculate the finite time Lyapunov exponent (FTLE) field for three different stirring protocols, one topologically complex (pseudo-Anosov) and two simple (finite-order), over a range of viscosities. After extracting appropriate measures indicative of both the amount of mixing and the area of effective mixing from the FTLE field, we see a clearly defined Reynolds number range in which the relative efficacy of the pseudo-Anosov protocol over the finite-order protocols justifies the application of topological chaos. More unexpectedly, we see that while the measures of effective mixing area increase with increasing Reynolds number for the finite-order protocols, they actually exhibit non-monotonic behavior for the pseudo-Anosov protocol.

preprint2016arXiv

Using invariant manifolds to construct symbolic dynamics for three-dimensional volume-preserving maps

Topological techniques are powerful tools for characterizing the complexity of many dynamical systems, including the commonly studied area-preserving maps of the plane. However, the extension of many topological techniques to higher dimensions is filled with roadblocks preventing their application. This article shows how to extend the homotopic lobe dynamics (HLD) technique, previously developed for 2D maps, to volume-preserving maps of a three-dimensional phase space. Such maps are physically relevant to particle transport by incompressible fluid flows or by magnetic field lines. Specifically, this manuscript shows how to utilize two-dimensional stable and unstable invariant manifolds, intersecting in a heteroclinic tangle, to construct a symbolic representation of the topological dynamics of the map. This symbolic representation can be used to classify system trajectories and to compute topological entropy. We illustrate the salient ideas through a series of examples with increasing complexity. These examples highlight new features of the HLD technique in 3D. Ultimately, in the final example, our technique detects a difference between the 2D stretching rate of surfaces and the 1D stretching rate of curves, illustrating the truly 3D nature of our approach.

preprint2015arXiv

Point Vortices: Finding Periodic Orbits and their Topological Classification

The motion of point vortices constitutes an especially simple class of solutions to Euler's equation for two dimensional, inviscid, incompressible, and irrotational fluids. In addition to their intrinsic mathematical importance, these solutions are also physically relevant. Rotating superfluid helium can support rectilinear quantized line vortices, which in certain regimes are accurately modeled by point vortices. Depending on the number of vortices, it is possible to have either regular integrable motion or chaotic motion. Thus, the point vortex model is one of the simplest and most tractable fluid models which exhibits some of the attributes of weak turbulence. The primary aim of this work is to find and classify periodic orbits, a special class of solutions to the point vortex problem. To achieve this goal, we introduce a number of algorithms: Lie transforms which ensure that the equations of motion are accurately solved; constrained optimization which reduces close return orbits to true periodic orbits; object-oriented representations of the braid group which allow for the topological comparison of periodic orbits. By applying these ideas, we accumulate a large data set of periodic orbits and their associated attributes. To render this set tractable, we introduce a topological classification scheme based on a natural decomposition of mapping classes. Finally, we consider some of the intriguing patterns which emerge in the distribution of periodic orbits in phase space. Perhaps the most enduring theme which arises from this investigation is the interplay between topology and geometry. The topological properties of a periodic orbit will often force it to have certain geometric properties.

preprint2010arXiv

A Robust Numerical Method for Integration of Point-Vortex Trajectories in Two Dimensions

The venerable 2D point-vortex model plays an important role as a simplified version of many disparate physical systems, including superfluids, Bose-Einstein condensates, certain plasma configurations, and inviscid turbulence. This system is also a veritable mathematical playground, touching upon many different disciplines from topology to dynamic systems theory. Point-vortex dynamics are described by a relatively simple system of nonlinear ODEs which can easily be integrated numerically using an appropriate adaptive time stepping method. As the separation between a pair of vortices relative to all other inter-vortex length scales decreases, however, the computational time required diverges. Accuracy is usually the most discouraging casualty when trying to account for such vortex motion, though the varying energy of this ostensibly Hamiltonian system is a potentially more serious problem. We solve these problems by a series of coordinate transformations: We first transform to action-angle coordinates, which, to lowest order, treat the close pair as a single vortex amongst all others with an internal degree of freedom. We next, and most importantly, apply Lie transform perturbation theory to remove the higher-order correction terms in succession. The overall transformation drastically increases the numerical efficiency and ensures that the total energy remains constant to high accuracy.