Manifolds: extrinsic and intrinsic

  • Context: Graduate 
  • Thread starter Thread starter shereen1
  • Start date Start date
  • Tags Tags
    Extrinsic Manifolds
Click For Summary

Discussion Overview

The discussion revolves around the concepts of extrinsic and intrinsic definitions of manifolds, particularly focusing on how these definitions relate to the dimensionality of the manifold and its neighborhoods. Participants explore the implications of viewing a manifold as embedded in a higher-dimensional space versus defining it through its intrinsic properties.

Discussion Character

  • Technical explanation
  • Conceptual clarification
  • Debate/contested

Main Points Raised

  • Some participants assert that a manifold is globally non-Euclidean but locally Euclidean, with neighborhoods homeomorphic to open sets of Euclidean space of the same dimension as the manifold.
  • One participant clarifies the distinction between extrinsic and intrinsic views, noting that the extrinsic view defines a manifold as a subspace of a larger space, while the intrinsic view defines it via metrics that describe curvature.
  • Another participant emphasizes that for any extrinsic definition, there is only one matching intrinsic definition, but multiple extrinsic definitions can correspond to a single intrinsic definition.
  • A question is raised about the dimensionality of neighborhoods in the context of a 2-sphere embedded in 3-dimensional space, specifically whether these neighborhoods are homeomorphic to subsets of dimension 2 or 3.
  • Responses indicate that neighborhoods of points on a 2-sphere are homeomorphic to open subsets of Euclidean space of dimension 2, regardless of the higher-dimensional embedding.
  • One participant mentions the subspace topology of the Euclidean space and the ability to flatten a patch of the sphere, reinforcing the idea that local neighborhoods resemble flat 2-space.

Areas of Agreement / Disagreement

Participants generally agree that neighborhoods of a manifold's points correspond to open subsets of Euclidean space of the same dimension as the manifold itself. However, there is some debate regarding the implications of extrinsic versus intrinsic definitions and how they relate to dimensionality.

Contextual Notes

Some discussions involve assumptions about the nature of manifolds and their embeddings that may not be universally accepted, particularly regarding the definitions and implications of extrinsic and intrinsic properties.

shereen1
Messages
50
Reaction score
1
Dear all
We all agree that a manifold is globally non euclidean but locally it is. So we can find near each point a hemeomorphic to an open set of euclidean space of the same dimension as the manifold. This is a general definition for all manifold to follow. Then what is the difference between extrinsic way of constructing a manifold and intrinsic one. I know that the first is related to embedded surface but still i didnt understand. Thank you
 
Physics news on Phys.org
shereen1 said:
So we can find near each point a hemeomorphic to an open set of euclidean space of the same dimension as the manifold.
'Homeomorphic', not 'hemeomorphic', just in case that wasn't a typo.

shereen1 said:
Then what is the difference between extrinsic way of constructing a manifold and intrinsic one. I know that the first is related to embedded surface but still i didnt understand.
The extrinsic view of a manifold is a subspace of a larger space - usually a space with more dimensions. In that case the manifold can be defined by an equation that tells us what points in that larger space it occupies. For instance the unit sphere, viewed extrinsically as a subset of Euclidean 3-space ('embedded' in that space), is the set of points ##(x,y,z)## that satisfies the equation ##x^2+y^2+z^2=1##. The intrinsic properties of the manifold - the curvature - can be derived from the extrinsic definition.

The intrinsic view does not define the manifold as embedded in another space. Rather, it defines it via the metric or related information that tells us about curvature. The metric for the above unit sphere is

$$
\left( \begin{array}{cc}
1 & 0 \\
0 & sin^2\theta \end{array} \right)
$$

where the coordinates are ##\theta## and ##phi## with ranges ##[0,\pi)## and ##[0,2\pi)## respectively and correspond to the spherical coordinate angles in the extrinsic formulation.

For any extrinsic definition there is only one possible matching intrinsic definition. But for any intrinsic definition there will usually be many different possible compatible extrinsic definitions, which are different ways of embedding the manifold in different parts of different host spaces.
 
  • Like
Likes   Reactions: shereen1
Than
andrewkirk said:
'Homeomorphic', not 'hemeomorphic', just in case that wasn't a typo.The extrinsic view of a manifold is a subspace of a larger space - usually a space with more dimensions. In that case the manifold can be defined by an equation that tells us what points in that larger space it occupies. For instance the unit sphere, viewed extrinsically as a subset of Euclidean 3-space ('embedded' in that space), is the set of points ##(x,y,z)## that satisfies the equation ##x^2+y^2+z^2=1##. The intrinsic properties of the manifold - the curvature - can be derived from the extrinsic definition.

The intrinsic view does not define the manifold as embedded in another space. Rather, it defines it via the metric or related information that tells us about curvature. The metric for the above unit sphere is

$$
\left( \begin{array}{cc}
1 & 0 \\
0 & sin^2\theta \end{array} \right)
$$

where the coordinates are ##\theta## and ##phi## with ranges ##[0,\pi)## and ##[0,2\pi)## respectively and correspond to the spherical coordinate angles in the extrinsic formulation.

For any extrinsic definition there is only one possible matching intrinsic definition. But for any intrinsic definition there will usually be many different possible compatible extrinsic definitions, which are different ways of embedding the manifold in different parts of different host spaces.
Thank you for this reply yes for sure it is a typing error :).
I just need to check something. If the dimension of the manifold is n then we say that the neighborhood of each point in the manifold is Homeomorphic to an open subset of euclidean space of dimension n ( same dimension). My question is suppose i am looking to the manifold from an extrinsic point view let's say a 2- sphere embbeded in a 3 dimension euclidean space. Then do i say that the neighborhood of each point in the manifold( dimension 2 here) is Homeomorphic to an open subset of euclidean space of dimension 3 or 2??
Thank you for your reply
 
It's 2. The dimension of the space in which the manifold is embedded is not relevant.
 
you need to look at the 2-sphere in the subspace topology of the euclidean 3-space. you can remove the north pole from the sphere and lay it flat through projection because it is locally homeomorphic to flat 2-space.
 
  • Like
Likes   Reactions: shereen1
shereen1 said:
Than

Thank you for this reply yes for sure it is a typing error :).
I just need to check something. If the dimension of the manifold is n then we say that the neighborhood of each point in the manifold is Homeomorphic to an open subset of euclidean space of dimension n ( same dimension). My question is suppose i am looking to the manifold from an extrinsic point view let's say a 2- sphere embbeded in a 3 dimension euclidean space. Then do i say that the neighborhood of each point in the manifold( dimension 2 here) is Homeomorphic to an open subset of euclidean space of dimension 3 or 2??
Thank you for your reply

It would be 2, because the sphere is a 2-manifold. Notice a patch of the sphere "looks more" like a copy of ##\mathbb R^2 ## than of ## \mathbb R^3 ## -- you can flatten it to look like a square. And notice that every point has a neighborhood homeomorphic _to an open ball_ which is itself homeomorphic to ## \mathbb R^n ##, stating only an open set allows possibilities like an open annulus, etc.
 
  • Like
Likes   Reactions: shereen1

Similar threads

  • · Replies 37 ·
2
Replies
37
Views
4K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 40 ·
2
Replies
40
Views
9K
  • · Replies 4 ·
Replies
4
Views
4K
  • · Replies 8 ·
Replies
8
Views
5K
  • · Replies 16 ·
Replies
16
Views
2K
  • · Replies 7 ·
Replies
7
Views
4K
  • · Replies 20 ·
Replies
20
Views
6K
  • · Replies 156 ·
6
Replies
156
Views
26K
  • · Replies 9 ·
Replies
9
Views
7K