WorksMathematics
A Smooth Nonnegatively Curved Sphere Whose Zero-Curvature Point Locus Has Hausdorff Dimension 1/2
Abstract
For every integer , we construct a smooth Riemannian metric on a manifold diffeomorphic to such that , every sectional curvature is strictly positive away from a closed nowhere-dense set, and the point locus on which some sectional curvature vanishes has Hausdorff dimension exactly . The metric is induced on the boundary of a smooth convex body in . The essential local model is a convex graph whose Hessian has a one-dimensional kernel precisely on a Cantor subset of one coordinate axis; a regularized maximum then inserts this graph into a round sphere without introducing additional degenerate points. The main content of this paper is generated by ChatGPT 5.6 and verified by the author.
Statement and convention
Part of the motivation of this construction comes from recent advances in the study of positively curved manifolds, e.g. [1, 2, 3].
Acknowledgement: the author thanks Xi’an Jiaotong-Liverpool University for providing financial and academic support. He also thanks Zipei Nie for providing access to ChatGPT and insights about AI usage. His gratitude also goes to his daughter and son who were born last year, and all of his family members who have always been supportive during his academic career.
For a Riemannian manifold with nonnegative sectional curvature, define the zero-curvature point locus
This is a subset of the manifold, rather than a subset of the Grassmann bundle of two-planes.
The main conclusion of this paper is the following theorem:
Theorem 1.1. For every integer , there is a smooth closed hypersurface , diffeomorphic to , such that its induced metric satisfies
everywhere;
every two-plane has strictly positive sectional curvature at every point of the open dense set ;
.
In fact, is bi-Lipschitz equivalent to a self-similar Cantor set of Hausdorff dimension .
We note that in the construction below, degeneracy is confined simultaneously in the transverse variables, so the exceptional point set itself is only a Cantor set.
2010 Mathematics Subject Classification. 53C21.
Key words and phrases. Sectional Curvature, Hausdorff Dimension.
Three elementary tools
A Cantor set of dimension 1/2
Let be the unique nonempty compact set satisfying
Equivalently, at each stage one retains the left and right subintervals of relative length . Put
Then is compact, nowhere dense, and symmetric under .
Lemma 2.1. The Hausdorff dimension of is .
Proof. Translation and nonzero scaling do not change Hausdorff dimension, so it is enough to treat . At level , the set is covered by intervals of length . Hence, for every ,
which gives .
For the reverse inequality, let be the natural probability measure assigning mass to every level- interval. Distinct level- intervals have length and are separated by gaps of length at least . Therefore an interval of length at most meets at most two level- intervals. If
then
Thus, for every countable interval cover by sufficiently short intervals,
It follows that the -dimensional Hausdorff measure of is positive, and hence . □
A smooth function with a prescribed zero set
Lemma 2.2. If is closed, then there is a function such that and . If , the function may be chosen even and bounded by .
Proof. Let . Choose nonnegative functions such that the open sets cover . Choose so rapidly decreasing that
For every fixed derivative order, the corresponding differentiated series converges uniformly. Hence
is smooth, nonnegative, and has zero set exactly . If , replace by and then by
The resulting function is smooth, even, takes values in , and has zero set . □
Apply Lemma 2.2 to the symmetric set in (2). Fix from now on an even function
A regularized maximum
Lemma 2.3. For every , there is a smooth function with the following properties:
(a) ;
(b) whenever ;
(c) if and are smooth convex functions on a convex domain, then is convex;
(d) at a point where , if either or is positive definite, then is positive definite.
Proof. Choose an even nonnegative mollifier supported in , positive for , and satisfying . Define
Then is smooth, even, and convex. Since is even, it has mean zero, and therefore for . Jensen’s inequality and the triangle inequality give
Moreover, for , and .
Define
The first two assertions follow immediately. For , put and
A direct differentiation gives
All three terms are positive semidefinite when and are convex. If , then , so a positive-definite Hessian among makes the right-hand side positive definite. □
A convex local model with Cantor degeneracy
Write a point of as
Let and fix
Define
Since is even, the function is even. Moreover,
Set
Let
Proposition 3.1. The function is convex on . Its Hessian is positive definite on , and at every point of its Hessian has rank exactly . Moreover,
Proof. Since the constant does not affect the Hessian, it is enough to consider . In the splitting ,
The lower-right block is positive definite. Its Schur complement is
Because , the coefficient is positive. Thus the Schur complement is nonnegative, and it vanishes exactly when and , namely at the points of . This proves convexity and positive definiteness off . At a point , the matrix in (10) is
which has rank .
Finally, (6) and give
The mixed term in (7) is the localization mechanism. The -direction Hessian may vanish when , but for the positive term in restores strict convexity. Hence the degeneracy does not extend to an -dimensional fiber.
Insertion into a round sphere
On , write the lower and upper unit hemispheres as graphs
The function is strictly convex, since
Fix
and define
Proposition 4.1. The function has the following properties.
is smooth and convex on .
is singular exactly on ; there it has rank , and it is positive definite on .
whenever .
for every .
Proof. Convexity and smoothness follow from Lemma 2.3, Proposition 3.1, and (13). Put
If , then and
Thus on a neighborhood of , so has rank there.
If away from , then and . If , then and . In the transition region , formula (4) contains a positive multiple of the positive-definite matrix , so . At the boundary value , the equality holds; such a point is not in by (15), and hence . The case reduces to . This proves (2).
If , then
Using (9),
Therefore on this outer collar, proving (3).
It remains to compare with . If , then
Both and , so Lemma 2.3 gives
If , then . This proves (4).
Define a closed hypersurface in by
Because on an outer collar, the first graph agrees there with the lower unit hemisphere. Hence the union in (16) is a smooth embedded hypersurface and is diffeomorphic to . Proposition 4.1(4) shows that the two graph pieces are disjoint away from the equator.
The hypersurface is also the boundary of the convex body
Indeed, is convex and is concave, so the set in (17) is convex. This global convexity is consistent with the local curvature calculation below, but the calculation itself will give the precise zero-curvature locus.
Sectional curvature of the hypersurface
Let be the metric induced on from the Euclidean metric. We first record the graph calculation. For a smooth function on a domain in , parameterize its graph by
For tangent vectors , the first and second fundamental forms, using the upward unit normal, are
The Euclidean Gauss equation therefore gives
for linearly independent .
If is positive semidefinite, the numerator in (19) is nonnegative by the Cauchy–Schwarz inequality for a positive-semidefinite bilinear form. If is positive definite, its restriction to every two-dimensional subspace is positive definite, so the numerator is strictly positive.
On the upper hemisphere, the induced metric is the standard round metric and has sectional curvature . On the lower graph, take . Proposition 4.1 shows that , and hence every sectional curvature is nonnegative. At every point outside the graph of , the Hessian is positive definite, so every sectional curvature is strictly positive.
At a point , the function agrees with , and (10) shows that
Taking and any nonzero in a -direction in (19) gives a zero-curvature two-plane. Consequently,
This equality also proves that no additional zero-curvature points are introduced by the global patching.
Hausdorff dimension and density
Let
For every and , the estimate used in (15) gives , not merely at points of . Hence
The Euclidean distance between and is at least . Conversely, the curve
lies in and has length at most
Since intrinsic distance is at least Euclidean distance and at most the length of this curve, is bi-Lipschitz from onto . Lemma 2.1 now yields
The set is compact and has Hausdorff dimension smaller than , so it has empty interior. Its complement is therefore open and dense. Every sectional curvature is strictly positive at every point of this complement, as proved in the preceding section. The induced metric is , and hence in particular is of class . This completes the proof of Theorem 1.1.
Remark 6.1. The same construction gives any prescribed dimension . Replace the ratio in (1) by . The resulting two-branch self-similar Cantor set has Hausdorff dimension , while all convexity and curvature arguments are unchanged.
References
References
- [1]L. Kennard. Torus actions and positive curvature. Virtual seminar on geometry with symmetries, May 6, 2020.
- [2]A. Krishnan and M. Wiemeler. 10-dimensional positively curved manifolds with T^3-symmetry. Commun. Contemp. Math., Vol. 27, No. 10, 2550018 (2025).
- [3]Y. Liu. On Closed Six-Manifolds Admitting Riemannian Metrics with Positive Sectional Curvature and Non-Abelian Symmetry. Acta Mathematica Sinica-English Series, 40(12), 2024, 3003–3026.