Anosov flow Created 2026-10-14 Updated 2026-11-05
An Anosov flow is a nonsingular smooth flow on a compact smooth manifold with a continuous derivative-invariant splitting
Here is the generating vector field, the flow direction is one-dimensional, and in the usual nontrivial convention both and have positive dimension. Constants must work uniformly over all points, all and the corresponding tangent vectors:
The stable bundle of an Anosov flow contracts forward; the unstable bundle of an Anosov flow contracts backward. The neutral direction along is essential: changing the starting point slightly along an orbit mainly shifts its clock, so that direction cannot be included in either uniform contraction estimate.
A suspension flow of the cat map gives a three-dimensional example. A positive time change of a smooth flow preserves the Anosov property on a compact manifold, although its stable and unstable bundles can tilt along the flow direction. These are uniform structural conditions, stronger than a sign observed in one finite-time Lyapunov exponent calculation.
Cat map Created 2026-10-31 Updated 2026-11-05
The cat map is the area-preserving map of the real torus given by
Its integer matrix has determinant one and inverse , so reducing coordinates modulo one retains an invertible transformation of the torus.
Geometrically, stretches a unit square into a parallelogram. Cutting that parallelogram along the integer grid and translating its pieces by integer vectors brings it back into a unit square. Edges are identified on the torus, which is why cutting the drawing does not make the torus map discontinuous.
The eigenvalues and give one expanding and one contracting eigenvector. Their product one agrees with area preservation, while the absence of eigenvalues on the unit circle makes this a hyperbolic toral automorphism. The Fibonacci powers of the cat-map matrix give its iterates explicitly. Thus an integer matrix that is easy to multiply provides a concrete model of simultaneous uniform stretching, contraction and global wrapping.
Fibonacci powers of the cat-map matrix Created 2026-10-31 Updated 2026-11-05
Use the standard Fibonacci number convention , , . For the cat map matrix,
At , the right side is . To prove the induction step, multiply it by . The top-left entry becomes , the top-right becomes , and the lower row follows from the same recurrence. This advances all indices by two.
For example, , so two iterations send to modulo one. The formula turns repeated matrix multiplication into an explicit integer recurrence; the exponential growth of the Fibonacci numbers agrees with the expanding eigenvalue of the cat map.
Hyperbolic invariant set Created 2026-10-13 Updated 2026-11-05
Let be a smooth diffeomorphism and a compact invariant set. It is uniformly hyperbolic if its tangent spaces have a continuous splitting preserved by the derivative, with constants and such that
The same constants must work for every . Thus the stable directions contract forward and the unstable directions contract backward, uniformly over the set. A sign observed in a finite-time numerical Lyapunov exponent calculation does not establish these uniform estimates.
For the cat map on a two-dimensional torus, the constant matrix has eigenvalues . Its two eigenvectors give fixed stable and unstable directions; the smaller eigenvalue is the reciprocal of the larger. With the flat Euclidean metric, and give the required estimates everywhere. The whole torus is therefore hyperbolic.
A Smale horseshoe supplies a different example: its hyperbolic invariant set contains infinitely many saddle periodic orbits, yet need not attract a neighborhood. Hyperbolicity describes stretching and contraction, while being an attractor concerns nearby trajectories. For a nonsingular flow map, the analogous splitting also has the neutral direction along the vector field; it should not be omitted by copying the map definition unchanged.
Hyperbolic toral automorphism Created 2026-10-14 Updated 2026-11-05
A hyperbolic toral automorphism is a toral automorphism whose matrix has no eigenvalue on the unit circle. Here , so its determinant is . Invertibility merely over the real numbers would not suffice: multiplication by two on a circle is a many-to-one endomorphism, not an automorphism.
The real stable and unstable subspaces come from the generalized eigenspaces with eigenvalue modulus less than and greater than one. They give the same splitting at every point of the real torus. A Jordan block can add a polynomial factor to the powers of its eigenvalue; choosing a slightly weaker exponential rate absorbs that factor into a constant. Thus contraction forward on the stable subspace and backward on the unstable subspace is uniform over the whole torus, proving that the map is an Anosov diffeomorphism.
For the cat map, has eigenvalues . One is inside and the other outside the unit circle, and their product is one. The toral automorphism mixing criterion applies as well, since a root of unity would lie on that circle.
Suspension flow Created 2026-10-14 Updated 2026-11-05
A suspension flow turns a discrete-time map into continuous motion by moving along an extra time coordinate. For a diffeomorphism and constant roof one, form the mapping torus
The flow moves vertically at unit speed; each crossing of the top applies and re-enters at height zero. If has dimension , the suspension has dimension .
Suspending a hyperbolic toral automorphism gives an Anosov flow. After roof crossings, its transverse derivative is , so the base map's stable and unstable exponential bounds become the transverse flow bounds. A remaining fractional crossing has a uniformly bounded effect on a compact mapping torus. The vertical direction supplies the neutral flow direction. In particular, the two-dimensional cat map gives a three-dimensional suspension.
A constant roof also shows why this flow need not be mixing. The well-defined height function obeys . Under the product invariant probability measure its mean is zero, but its autocorrelation has modulus one for all . Thus even a mixing base map can produce a suspension with persistent clock correlations.
Toral automorphism Created 2026-10-14 Updated 2026-11-05
An integer matrix with determinant defines a toral automorphism
on the real torus . Integer entries make the map independent of the chosen representative. The inverse matrix also has integer entries, so the transformation is invertible. It preserves normalized Haar measure.
The map is a mixing measure-preserving transformation exactly when no eigenvalue of is a root of unity. This can be proved with characters of a real torus , since . Orthogonality gives
If no eigenvalue is a root of unity, a nonzero integer frequency cannot repeat: a repetition would give for some . It therefore eventually leaves every finite set of frequencies. The displayed correlations prove mixing first for trigonometric polynomials, then for square-integrable functions by approximation.
Conversely, a root-of-unity eigenvalue makes singular for some . As an integer matrix it has a nonzero rational kernel vector, which can be scaled to a nonzero integer vector . The nonconstant character is then invariant under , obstructing mixing. The cat map is a mixing example, while the identity matrix gives an invertible measure-preserving map that is not mixing.