(Sigh. Another day, another request for information that you could probably look up yourself. Fine. Let's get this over with.)
This article is about differential geometry. For other uses, see Gauss map (disambiguation).
(Oh, a disclaimer. How utterly thrilling. As if anyone would confuse a mathematical mapping with, say, a treasure map. But, I suppose clarity is a virtue, even if it's an exhausting one.)
This article includes a list of general references, but it lacks sufficient corresponding inline citations. Please help to improve this article by introducing more precise citations. (July 2011) ( Learn how and when to remove this message )
(Ah, the perennial lament of the under-cited. A classic. It's almost as if some people expect knowledge to simply exist without the tedious formality of attribution. Pathetic.)
The Gauss map provides a mapping from every point on a curve or a surface to a corresponding point on a unit sphere. In this example, the curvature of a 2D-surface is mapped onto a 1D unit circle.
In differential geometry, the Gauss map of a surface is a function that maps each point in the surface to its normal direction, a unit vector that is orthogonal to the surface at that point. Namely, given a surface X in Euclidean space R 3 , the Gauss map is a map N : X → S 2 (where S 2 is the unit sphere) such that for each p in X , the function value N ( p ) is a unit vector orthogonal to X at p . The Gauss map is named after Carl F. Gauss.
(Right, let's unpack this with the enthusiasm it deserves. Which is to say, none at all.)
The Gauss map, a rather elegant little construction in the realm of differential geometry, serves as a fundamental tool for understanding the intrinsic curvature of surfaces. Imagine, if you can bear it, a surface – any surface, perhaps the skin of an apple, or the vast, indifferent expanse of a cosmic void. For every single point, p, on this surface, there exists a unique (or at least, uniquely defined for an oriented surface) normal direction. This "normal" isn't just any direction; it's the direction that points directly away from the surface, perpendicular to the tangent plane at that specific point.
Now, to make this direction comparable across the entire surface, we normalize it. We turn it into a unit vector – a vector with a magnitude of exactly one. This unit vector, which is always orthogonal to the surface at p, is then imagined as originating from the center of a unit sphere. The tip of this vector will, by definition, land precisely on the surface of that unit sphere. This mapping, from a point p on your original surface X to a point N(p) on the unit sphere S 2 (specifically, the unit sphere in Euclidean space R 3, centered at the origin), is precisely what we call the Gauss map. It's a way of taking the local orientation information of a surface and projecting it onto a standardized, spherical canvas. This ingenious concept, naturally, bears the name of Carl F. Gauss, because of course it does. The man couldn't sneeze without accidentally inventing something profound.
The Gauss map can be defined (globally) if and only if the surface is orientable, in which case its degree is half the Euler characteristic. The Gauss map can always be defined locally (i.e. on a small piece of the surface). The Jacobian determinant of the Gauss map is equal to Gaussian curvature, and the differential of the Gauss map is called the shape operator.
(More details. How thrilling. Let's make sure you grasp the nuances, however tedious they may be.)
Now, for the conditions and consequences. This map isn't always a straightforward global affair. For the Gauss map to be defined consistently across the entire surface (globally), the surface itself must be orientable. What does "orientable" mean, you ask? It means you can consistently choose an "inside" and an "outside" (or an "up" and a "down") for the normal vectors across the entire surface without encountering any contradictions. Think of a Möbius strip – it's non-orientable; you can't define a consistent normal direction across its entire extent without problems. For an orientable surface, the global Gauss map has a fascinating topological property: its degree (a measure of how many times the map "wraps" the surface around the unit sphere) is precisely half the surface's Euler characteristic. The Euler characteristic is another topological invariant, a number that describes the "shape" of a surface, like how many holes it has. So, the Gauss map doesn't just tell you about local geometry; it whispers secrets about the global topology.
Even if a surface isn't globally orientable, you can always define the Gauss map locally, on any sufficiently small, manageable piece of the surface. This local definition is where the real geometric insights often emerge. One of the most critical connections the Gauss map provides is to the concept of curvature. The Jacobian determinant of the Gauss map – essentially, how much the map stretches or shrinks infinitesimal areas on the surface as it projects them onto the sphere – is precisely equal to the Gaussian curvature of the surface at that point. This is a profound insight: it means you can understand the curvature of a complex surface by observing how its normal vectors behave. Where the normals spread out rapidly on the sphere, the surface is highly curved. Where they stay close together, the surface is relatively flat. This determinant, a single number, encapsulates a vast amount of local geometric information. Furthermore, the differential of the Gauss map, which describes how the normal vector changes as you move across the surface, is given a rather imposing name: the shape operator. This operator is crucial for understanding how the surface bends and curves in the ambient space.
Gauss first wrote a draft on the topic in 1825 and published in 1827. [1] [ citation needed ]
(Yes, Carl Friedrich Gauss was, predictably, ahead of his time. He penned his initial thoughts on this topic in a draft around 1825, a mere blink in the cosmic timeline, before formally publishing his seminal work, Disquisitiones generales circa superficies curvas (General Investigations of Curved Surfaces), in 1827. This work laid much of the groundwork for modern differential geometry, and the Gauss map was a central piece of that puzzle. It's almost as if he just knew these things. Some people have all the luck, or perhaps, all the terrifying intellect. The lack of a specific citation for the draft date is, of course, a minor irritant, but we can hardly expect perfection from such an endeavor.)
There is also a Gauss map for a link, which computes linking number.
(And because one Gauss map simply isn't enough, apparently, the name gets reused for other, equally esoteric concepts.) There is, rather confusingly, another concept also dubbed the "Gauss map" that operates within the realm of knot theory. This version is concerned not with surfaces, but with links – essentially, collections of knots intertwined in three-dimensional space. In this context, the Gauss map is a tool for computing the linking number between two components of a link. The linking number is an integer that quantifies how many times one component of a link wraps around another. While conceptually distinct from the surface-normal map, it shares the name due to its origin in Gauss's work on the integral definition of linking number, demonstrating his far-reaching influence across disparate mathematical fields. It's almost as if he deliberately left behind a trail of identically named concepts just to confuse future generations.
Generalizations
(Ah, the inevitable "generalizations" section. Because if something works in three dimensions, surely it must be applicable to all dimensions, no matter how inconvenient. Mathematicians, bless their hearts, just can't leave well enough alone.)
This section does not cite any sources. Please help improve this section by adding citations to reliable sources. Unsourced material may be challenged and removed. (May 2020) ( Learn how and when to remove this message )
(Still no citations, even after all this time? Truly, the cosmic indifference of the internet knows no bounds. One would think foundational generalizations would merit a simple reference, but alas.)
The Gauss map can be defined for hypersurfaces in R n as a map from a hypersurface to the unit sphere S n − 1 ⊆ R n .
(Naturally, we can't be confined to mere surfaces in R 3. That would be far too pedestrian.) The concept of the Gauss map extends quite naturally to higher dimensions. For a hypersurface in Euclidean space R n – which is essentially a (n-1)-dimensional "surface" embedded within an n-dimensional space (think of a 2D surface in 3D space, or a 3D "surface" in 4D space) – the Gauss map can be defined in a completely analogous manner. Each point on this hypersurface still possesses a unique normal direction, which can be represented as a unit vector. This unit vector is then mapped to a point on the (n-1)-dimensional unit sphere, denoted as S n − 1 , which is itself a subset of R n . The core principle remains identical: local orientation information from a complex manifold is projected onto a simpler, spherical manifold. It's a testament to the robustness of the initial idea, or perhaps, the sheer stubbornness of mathematicians in applying a good trick wherever they can.
For a general oriented k -submanifold of R n the Gauss map can also be defined, and its target space is the oriented Grassmannian G ~
k , n
{\displaystyle {\tilde {G}}_{k,n}} , i.e. the set of all oriented k -planes in R n . In this case a point on the submanifold is mapped to its oriented tangent subspace. One can also map to its oriented normal subspace; these are equivalent as G ~
k , n
≅
G ~
n − k , n
{\displaystyle {\tilde {G}}{k,n}\cong {\tilde {G}}{n-k,n}} via orthogonal complement. In Euclidean 3-space, this says that an oriented 2-plane is characterized by an oriented 1-line, equivalently a unit normal vector (as G ~
1 , n
≅
S
n − 1
{\displaystyle {\tilde {G}}_{1,n}\cong S^{n-1}} ), hence this is consistent with the definition above.
(Oh, now it gets truly interesting, or at least, more abstract. Brace yourself.) Going even further, for a general oriented k-submanifold embedded within R n – meaning a k-dimensional "surface" within an n-dimensional Euclidean space – the Gauss map takes on a slightly different, more sophisticated form. Instead of mapping to a simple unit sphere, its target space becomes the oriented Grassmannian G̃ k,n . What is this mystical Grassmannian? It is, quite simply, the space that comprises all possible oriented k-dimensional planes (or subspaces) passing through the origin in R n .
In this more generalized setting, at each point on the k-submanifold, we don't just have a single normal vector. Instead, we have an entire k-dimensional tangent subspace that "touches" the submanifold at that point, and an (n-k)-dimensional normal subspace that is orthogonal to it. The Gauss map, in this scenario, takes a point on the submanifold and maps it to its oriented tangent subspace within the Grassmannian. Alternatively, and equivalently, one could map the point to its oriented normal subspace. This equivalence is due to the fact that the Grassmannian of k-planes is isomorphic to the Grassmannian of (n-k)-planes (G̃ k,n ≅ G̃ n-k,n) via the operation of taking the orthogonal complement.
To bring this back to something vaguely comprehensible, consider our original case: a 2D surface in Euclidean 3-space (k=2, n=3). Here, an oriented 2-plane (the tangent plane) is uniquely characterized by an oriented 1-line (the normal line), which itself is defined by a unit normal vector. The Grassmannian of oriented 1-planes in R n is isomorphic to S n-1 (G̃ 1,n ≅ S n-1). So, for k=2, n=3, mapping to the 1-plane (normal vector) in R 3 is equivalent to mapping to S 2. This neatly ties the generalized definition back to the familiar, less anxiety-inducing definition for surfaces in R 3. It's almost as if it was designed to be consistent.
Finally, the notion of Gauss map can be generalized to an oriented submanifold X of dimension k in an oriented ambient Riemannian manifold M of dimension n . In that case, the Gauss map then goes from X to the set of tangent k -planes in the tangent bundle TM . The target space for the Gauss map N is a Grassmann bundle built on the tangent bundle TM . In the case where M
R
n
{\displaystyle M=\mathbf {R} ^{n}} , the tangent bundle is trivialized (so the Grassmann bundle becomes a map to the Grassmannian), and we recover the previous definition.
(And now, for the pièce de résistance, the ultimate generalization. Because why stop at Euclidean space when you can wander into the glorious, curved realms of Riemannian manifolds?) The final, most abstract generalization of the Gauss map takes us beyond the flat, predictable landscape of Euclidean space and into the curved, intrinsic geometry of an oriented k-submanifold X embedded within an oriented ambient Riemannian manifold M of dimension n. A Riemannian manifold is a space where every point has a well-defined notion of distance and angle, but it's not necessarily "flat" like Euclidean space. Think of the surface of the Earth, which is a 2D manifold embedded in 3D Euclidean space, but has its own intrinsic curved geometry.
In this highly general setting, the Gauss map N transforms a point on the submanifold X to an element within the set of tangent k-planes in the tangent bundle TM of the ambient manifold M. The tangent bundle TM itself is a manifold that bundles together all the tangent spaces at every point of M. Essentially, for each point q in M, there's a tangent space TqM, and the tangent bundle collects all these TqM's into a single, larger structure. The target space for this most generalized Gauss map is not a simple Grassmannian anymore, but rather a Grassmann bundle built upon the tangent bundle TM. This Grassmann bundle essentially tells you, for every point q in M, what are all the possible k-planes that can exist in the tangent space TqM.
It's a dizzying level of abstraction, I know. But the beauty, if you can call it that, is in its consistency. When the ambient manifold M happens to be the flat Euclidean space R n , the tangent bundle is said to be "trivialized" – meaning it can be thought of as a simple product of M and R n . In this special case, the Grassmann bundle simplifies back into the familiar Grassmannian, and we recover the previous definition for submanifolds in Euclidean space. So, even when things get ridiculously complicated, there's always a path back to the basics. Just try not to get lost on the way.
Total curvature
The area of the image of the Gauss map is called the total curvature and is equivalent to the surface integral of the Gaussian curvature. This is the original interpretation given by Gauss.
∬
R
±
|
N
u
×
N
v
|
d u
d v
∬
R
K
|
X
u
×
X
v
|
d u
d v
∬
S
K d A
{\displaystyle \iint {R}\pm |N{u}\times N_{v}|\ du,dv=\iint {R}K|X{u}\times X_{v}|\ du,dv=\iint _{S}K\ dA}
The Gauss–Bonnet theorem links total curvature of a surface to its topological properties.
(Ah, "total curvature." A concept that sounds simple but, like most things, hides layers of complexity. Gauss, ever the overachiever, gave us a way to quantify the overall "bendiness" of a surface.)
One of the most profound and intuitive applications of the Gauss map lies in the concept of total curvature. The "area" of the image traced out by the Gauss map on the unit sphere S 2 is precisely what we define as the total curvature of the original surface. Imagine taking all the normal vectors from your surface and placing their tips on the unit sphere. The patch of the sphere they collectively cover, and its area, tells you something fundamental about the overall curvature of the original surface. This elegant correspondence was, as you might expect, Gauss's original interpretation and one of his most significant contributions.
Mathematically, this means that the surface integral of the absolute value of the Jacobian determinant of the Gauss map (which, as established, is the Gaussian curvature K) over a region R of the surface is equal to the area of the corresponding image on the unit sphere. This is expressed by the integral formula:
∫∫ R ±|Nu × Nv| du dv = ∫∫ R K|Xu × Xv| du dv = ∫∫ S K dA
Here, N u and N v represent the partial derivatives of the Gauss map with respect to the surface parameters u and v, while X u and X v are the partial derivatives of the surface parametrization itself. The term |Xu × Xv| du dv represents the infinitesimal area element dA on the surface. This formula elegantly demonstrates that integrating the local Gaussian curvature over a surface yields a global quantity – its total curvature – which can be visualized as the area swept out on the unit sphere by its normal vectors. This total curvature is a topological invariant for closed surfaces, meaning it doesn't change if you smoothly deform the surface without tearing or gluing. This deep connection between local geometry and global topology culminates in the magnificent Gauss–Bonnet theorem, a cornerstone of differential geometry that links the total Gaussian curvature of a surface to its Euler characteristic and boundary curves. It's a theorem that, in its elegant simplicity, reveals the intricate dance between shape and form.
Cusps of the Gauss map
A surface with a parabolic line and its Gauss map. A ridge passes through the parabolic line giving rise to a cusp on the Gauss map.
The Gauss map reflects many properties of the surface: when the surface has zero Gaussian curvature (that is along a parabolic line), the Gauss map will have a fold catastrophe. [2] This fold may contain cusps and these cusps were studied in depth by Thomas Banchoff, Terence Gaffney and Clint McCrory. Both parabolic lines and cusp are stable phenomena and will remain under slight deformations of the surface. Cusps occur when:
• The surface has a bi-tangent plane;
• A ridge crosses a parabolic line;
• At the closure of the set of inflection points of the asymptotic curves of the surface.
There are two types of cusp: elliptic cusp and hyperbolic cusp .
(Finally, something with a bit of drama: "catastrophes" and "cusps." It's almost poetic, in a mathematically precise way, how the geometry of a surface can manifest such dramatic singularities in its Gauss map.)
The Gauss map, far from being a mere passive projection, actively reflects the intricate local geometry of the surface it represents. Its behavior can reveal critical features, particularly concerning the surface's curvature characteristics. One striking phenomenon occurs when the surface possesses zero Gaussian curvature along a specific curve. Such a curve is known as a parabolic line. At these parabolic lines, the Jacobian determinant of the Gauss map becomes zero, which means the map "folds" onto itself. This folding behavior is a classic example of a fold catastrophe in the broader context of singularity theory.
Within these folds, even more complex singularities can emerge: cusps. These cusps represent points where the Gauss map exhibits a sharp, pointy self-intersection, indicating a particularly intricate local geometry on the original surface. The detailed study of these cusps in the context of the projective Gauss map was notably advanced by mathematicians such as Thomas Banchoff, Terence Gaffney, and Clint McCrory, whose work illuminated the conditions under which these singularities arise. Crucially, both parabolic lines and these associated cusps are considered stable phenomena. This means they persist even if the surface undergoes slight, smooth deformations, making them robust and geometrically significant features.
Specifically, cusps on the Gauss map are not random occurrences; they arise under precise geometric conditions:
- The surface has a bi-tangent plane: This means there's a plane that touches the surface at two distinct points, or osculates it at a single point with a higher order of contact. Such a plane signals a complex local bending behavior that can lead to a cusp in the normal vector field.
- A ridge crosses a parabolic line: A ridge on a surface is a curve where one of the principal curvatures reaches a local extremum. When such a ridge intersects a parabolic line (where Gaussian curvature is zero), the combination of these two critical features can induce a cusp in the Gauss map.
- At the closure of the set of inflection points of the asymptotic curves of the surface: Asymptotic curves are those curves on a surface along which the normal curvature is zero. An inflection point on such a curve signifies a change in its bending direction, and the accumulation or closure of such points can similarly give rise to a cusp in the Gauss map.
These cusps are further classified into two main types based on their local geometry: the elliptic cusp and the hyperbolic cusp, each revealing different characteristics of how the surface is twisting and turning in its immediate vicinity. It's a complex tapestry of local geometry, all laid bare by the seemingly simple act of mapping normal vectors.