Smooth quotient bundle
$begingroup$
Let $E rightarrow M$ be a smooth vector bundle.
This link gives a construction of quotient bundle for a subbundle. $E' subseteq E$.
We define an equivalence relation $sim$ on $E$ by $v_x sim v_y$ if and only if $x=y$ and $v_x-v_y in E'_x$ and define $E/sim rightarrow M$ to be our quotient bundle.
My question is, how does this construction also give smooth bundle?
My attempt, the problem is thus to prove we can give a smooth structure to $E/sim$.
Transition function formulation: This is equivalent to finding an open cover with local trivializations
$$ {phi_U:E/sim |_U rightarrow U times Bbb R^k }$$ such that the transition functions of any two on $U cap V$ are smooth map from
$$ U cap V rightarrow GL_k(Bbb R)$$
Using the transition functions of $E$, $E'$.
Let us choose transition functions, form $E'$ and $E$, on open sets
$$ phi': U cap V rightarrow GL_m(Bbb R)$$
$$ phi: U cap V rightarrow GL_n(Bbb R)$$
where $E$ is bundle of rank $n$, $E'$ is of rank $m$. But how does this induce a well defined smooth map on the quotient bundle?
differential-geometry algebraic-topology smooth-manifolds vector-bundles
$endgroup$
add a comment |
$begingroup$
Let $E rightarrow M$ be a smooth vector bundle.
This link gives a construction of quotient bundle for a subbundle. $E' subseteq E$.
We define an equivalence relation $sim$ on $E$ by $v_x sim v_y$ if and only if $x=y$ and $v_x-v_y in E'_x$ and define $E/sim rightarrow M$ to be our quotient bundle.
My question is, how does this construction also give smooth bundle?
My attempt, the problem is thus to prove we can give a smooth structure to $E/sim$.
Transition function formulation: This is equivalent to finding an open cover with local trivializations
$$ {phi_U:E/sim |_U rightarrow U times Bbb R^k }$$ such that the transition functions of any two on $U cap V$ are smooth map from
$$ U cap V rightarrow GL_k(Bbb R)$$
Using the transition functions of $E$, $E'$.
Let us choose transition functions, form $E'$ and $E$, on open sets
$$ phi': U cap V rightarrow GL_m(Bbb R)$$
$$ phi: U cap V rightarrow GL_n(Bbb R)$$
where $E$ is bundle of rank $n$, $E'$ is of rank $m$. But how does this induce a well defined smooth map on the quotient bundle?
differential-geometry algebraic-topology smooth-manifolds vector-bundles
$endgroup$
$begingroup$
If you choose a Riemannian metric on your bundle this is easy: it's just the orthogonal complement.
$endgroup$
– user98602
Jan 2 at 16:12
add a comment |
$begingroup$
Let $E rightarrow M$ be a smooth vector bundle.
This link gives a construction of quotient bundle for a subbundle. $E' subseteq E$.
We define an equivalence relation $sim$ on $E$ by $v_x sim v_y$ if and only if $x=y$ and $v_x-v_y in E'_x$ and define $E/sim rightarrow M$ to be our quotient bundle.
My question is, how does this construction also give smooth bundle?
My attempt, the problem is thus to prove we can give a smooth structure to $E/sim$.
Transition function formulation: This is equivalent to finding an open cover with local trivializations
$$ {phi_U:E/sim |_U rightarrow U times Bbb R^k }$$ such that the transition functions of any two on $U cap V$ are smooth map from
$$ U cap V rightarrow GL_k(Bbb R)$$
Using the transition functions of $E$, $E'$.
Let us choose transition functions, form $E'$ and $E$, on open sets
$$ phi': U cap V rightarrow GL_m(Bbb R)$$
$$ phi: U cap V rightarrow GL_n(Bbb R)$$
where $E$ is bundle of rank $n$, $E'$ is of rank $m$. But how does this induce a well defined smooth map on the quotient bundle?
differential-geometry algebraic-topology smooth-manifolds vector-bundles
$endgroup$
Let $E rightarrow M$ be a smooth vector bundle.
This link gives a construction of quotient bundle for a subbundle. $E' subseteq E$.
We define an equivalence relation $sim$ on $E$ by $v_x sim v_y$ if and only if $x=y$ and $v_x-v_y in E'_x$ and define $E/sim rightarrow M$ to be our quotient bundle.
My question is, how does this construction also give smooth bundle?
My attempt, the problem is thus to prove we can give a smooth structure to $E/sim$.
Transition function formulation: This is equivalent to finding an open cover with local trivializations
$$ {phi_U:E/sim |_U rightarrow U times Bbb R^k }$$ such that the transition functions of any two on $U cap V$ are smooth map from
$$ U cap V rightarrow GL_k(Bbb R)$$
Using the transition functions of $E$, $E'$.
Let us choose transition functions, form $E'$ and $E$, on open sets
$$ phi': U cap V rightarrow GL_m(Bbb R)$$
$$ phi: U cap V rightarrow GL_n(Bbb R)$$
where $E$ is bundle of rank $n$, $E'$ is of rank $m$. But how does this induce a well defined smooth map on the quotient bundle?
differential-geometry algebraic-topology smooth-manifolds vector-bundles
differential-geometry algebraic-topology smooth-manifolds vector-bundles
asked Jan 2 at 9:56
CL.CL.
2,4163925
2,4163925
$begingroup$
If you choose a Riemannian metric on your bundle this is easy: it's just the orthogonal complement.
$endgroup$
– user98602
Jan 2 at 16:12
add a comment |
$begingroup$
If you choose a Riemannian metric on your bundle this is easy: it's just the orthogonal complement.
$endgroup$
– user98602
Jan 2 at 16:12
$begingroup$
If you choose a Riemannian metric on your bundle this is easy: it's just the orthogonal complement.
$endgroup$
– user98602
Jan 2 at 16:12
$begingroup$
If you choose a Riemannian metric on your bundle this is easy: it's just the orthogonal complement.
$endgroup$
– user98602
Jan 2 at 16:12
add a comment |
2 Answers
2
active
oldest
votes
$begingroup$
We will appeal to Godemont's theorem, which states that if $X$ is a smooth manifold and $R$ is an equivalence relation on $X$, then $X/R$ has the structure of a smooth manifold such that $Xrightarrow X/R$ becomes a smooth submersion if and only if $Rsubseteq Xtimes X$ is a smoooth submanifold, and $pr_1:Rrightarrow X$ is a submersion. I can't find the exact reference I was looking for at the moment for this theorem, but it is a standard result. I know Bourbaki has a statement (its in French). A good general reference is section 4 of Lee's book Introduction to Smooth Manifolds.
To this end we consider the fibred product $Etimes_ME={(e_1,e_2)in Etimes Emid pi(e_1)=pi(e_2)}$, which is the pullback of $pi$ along itself. This space is a smooth submanifold of the product $Etimes E$. Likewise we consider $Etimes_ME'={(e,e')in Etimes E'mid pi(e)=pi'(e')}$, which, under the assumption that $E'subseteq E$ is a smooth submanifold, is itself a smooth submanifold $Etimes_ME'subseteq Etimes_ME$.
We however, will consider a non-standard embedding of this manifold. Since $E$ is a vector bundle it has a fibrewise sum $mu:Etimes_MErightarrow E$, which restricts to the standard vector space addition $(e_1,e_2)mapsto e_1+e_2$ on each fibre $E_ptimes E_p$, and we use this map to define
$$Theta:Etimes_ME'rightarrow Etimes_ME,qquad Theta(e,e')=(e,e+e').$$
It is clear that this map is a diffeomorphism onto its image, and that its image $Theta(Etimes_ME'):=mathcal{R}$ is precisely the graph of the relation $e_1sim e_2$ iff $pi(e_1)=pi(e_2)$ and $e_2-e_1in E'_{pi(e_1)}$. Then $mathcal{R}subseteq Etimes_MEsubseteq Etimes E$ is a smooth submanifold, the projection $pr_1:mathcal{R}rightarrow E$ is smooth, and we can appeal to Godemont's critereon to get a smooth structure on the quotient $E/E'$ such that the canonical map $q:Erightarrow E/E'$ becomes a smooth submersion. This then induces from the map $pi$ a submersive projection $rho:E/E'rightarrow M$ satisfying $pi=rhocirc q$. Submersions always have local sections, so $E/E'$ is a locally trivial fibre bundle over $M$. In fact, if $s$ is any local section of $pi$, then $qcirc s$ is a local section of $rho$.
Now from here it is not difficult to see that the fibrewise linear structure on $E$ transfers to $E/E'$ in such a way as to make $rho$ into a topological vector bundle, although ee have not yet shown that the induced operations of fiberwise addition and scalar multiplication are smooth. However this is not hard to do using the fact that $q$ is a submersion. In fact it is not difficult to see that $qtimes_Mq:Etimes_MErightarrow E'times_ME'$ is a submersion, and so induces the smooth fibrewise addition $mu':(E/E')times_M(E/E')rightarrow E/E'$ uniquely from $mu$. Similarly, the smoothness of the fibrewise scalar multiplicaiton $mathbb{K}times E/E'rightarrow E/E'$ follows.
To construct charts on $E/E'$ we use the previous remark on local sections to construct fibrewise linear local trivialisations $(U,varphi)$, $Usubseteq M$, of $rho$ in such a way that the restriction $q|_U$ becomes the quotient $Utimesmathbb{K}^srightarrow Utimesmathbb{K}^{s-r}$ locally modelling the vector space quotients $E_xrightarrow E_x/E'_x$.
$endgroup$
$begingroup$
I checked out lee's text but couldn't find it, you know which page it is?
$endgroup$
– CL.
Jan 4 at 6:44
$begingroup$
As I said, the theorem is not actually discussed by Lee, although the material in his book is a good general reference for submersions in general. Check out Roger Godement's book Introduction to the Theory of Lie Groups, section 4.9, Theorem 5, page 133 (Springer-Verlag 2017 English Translation) for the exact statement.
$endgroup$
– Tyrone
Jan 4 at 10:06
add a comment |
$begingroup$
You have to construct compatible trivializations of the bundles $E'$ and $E$, which is slightly complicated to express in terms of transition functions. The easilest way to describe this in my opinion is in terms of local frames. Choose a local frame for $E'$ defined on $U$. Shrinking $U$, you can extend this to a local frame for $E$. (In a point $xin U$, extend the values of your frame which form a basis for $E'_x$ to a basis for $E_x$ and then extend the additional elements to local smooth sections of $E$.) Then you get a local trivialization $E|_Uto Utimesmathbb R^n$ which restricts to a trivialzation $E'|_Uto Utimesmathbb R^k$. From this, you get a trivialization from the quotient bundle to $mathbb R^n/mathbb R^kcongmathbb R^{n-k}$.
In terms of transition functions, you can use the above construction to show that you can arrange things in such a way that the transtition functions for $E$ have values in block matrices of the form $begin{pmatrix} A & B \ 0 & Cend{pmatrix}$ such that the $A$-component gives you transition functions for $E'$. In this setting, the $C$-block gives you the transition functions for $E/E'$.
$endgroup$
add a comment |
Your Answer
StackExchange.ready(function() {
var channelOptions = {
tags: "".split(" "),
id: "69"
};
initTagRenderer("".split(" "), "".split(" "), channelOptions);
StackExchange.using("externalEditor", function() {
// Have to fire editor after snippets, if snippets enabled
if (StackExchange.settings.snippets.snippetsEnabled) {
StackExchange.using("snippets", function() {
createEditor();
});
}
else {
createEditor();
}
});
function createEditor() {
StackExchange.prepareEditor({
heartbeatType: 'answer',
autoActivateHeartbeat: false,
convertImagesToLinks: true,
noModals: true,
showLowRepImageUploadWarning: true,
reputationToPostImages: 10,
bindNavPrevention: true,
postfix: "",
imageUploader: {
brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
allowUrls: true
},
noCode: true, onDemand: true,
discardSelector: ".discard-answer"
,immediatelyShowMarkdownHelp:true
});
}
});
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f3059300%2fsmooth-quotient-bundle%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
2 Answers
2
active
oldest
votes
2 Answers
2
active
oldest
votes
active
oldest
votes
active
oldest
votes
$begingroup$
We will appeal to Godemont's theorem, which states that if $X$ is a smooth manifold and $R$ is an equivalence relation on $X$, then $X/R$ has the structure of a smooth manifold such that $Xrightarrow X/R$ becomes a smooth submersion if and only if $Rsubseteq Xtimes X$ is a smoooth submanifold, and $pr_1:Rrightarrow X$ is a submersion. I can't find the exact reference I was looking for at the moment for this theorem, but it is a standard result. I know Bourbaki has a statement (its in French). A good general reference is section 4 of Lee's book Introduction to Smooth Manifolds.
To this end we consider the fibred product $Etimes_ME={(e_1,e_2)in Etimes Emid pi(e_1)=pi(e_2)}$, which is the pullback of $pi$ along itself. This space is a smooth submanifold of the product $Etimes E$. Likewise we consider $Etimes_ME'={(e,e')in Etimes E'mid pi(e)=pi'(e')}$, which, under the assumption that $E'subseteq E$ is a smooth submanifold, is itself a smooth submanifold $Etimes_ME'subseteq Etimes_ME$.
We however, will consider a non-standard embedding of this manifold. Since $E$ is a vector bundle it has a fibrewise sum $mu:Etimes_MErightarrow E$, which restricts to the standard vector space addition $(e_1,e_2)mapsto e_1+e_2$ on each fibre $E_ptimes E_p$, and we use this map to define
$$Theta:Etimes_ME'rightarrow Etimes_ME,qquad Theta(e,e')=(e,e+e').$$
It is clear that this map is a diffeomorphism onto its image, and that its image $Theta(Etimes_ME'):=mathcal{R}$ is precisely the graph of the relation $e_1sim e_2$ iff $pi(e_1)=pi(e_2)$ and $e_2-e_1in E'_{pi(e_1)}$. Then $mathcal{R}subseteq Etimes_MEsubseteq Etimes E$ is a smooth submanifold, the projection $pr_1:mathcal{R}rightarrow E$ is smooth, and we can appeal to Godemont's critereon to get a smooth structure on the quotient $E/E'$ such that the canonical map $q:Erightarrow E/E'$ becomes a smooth submersion. This then induces from the map $pi$ a submersive projection $rho:E/E'rightarrow M$ satisfying $pi=rhocirc q$. Submersions always have local sections, so $E/E'$ is a locally trivial fibre bundle over $M$. In fact, if $s$ is any local section of $pi$, then $qcirc s$ is a local section of $rho$.
Now from here it is not difficult to see that the fibrewise linear structure on $E$ transfers to $E/E'$ in such a way as to make $rho$ into a topological vector bundle, although ee have not yet shown that the induced operations of fiberwise addition and scalar multiplication are smooth. However this is not hard to do using the fact that $q$ is a submersion. In fact it is not difficult to see that $qtimes_Mq:Etimes_MErightarrow E'times_ME'$ is a submersion, and so induces the smooth fibrewise addition $mu':(E/E')times_M(E/E')rightarrow E/E'$ uniquely from $mu$. Similarly, the smoothness of the fibrewise scalar multiplicaiton $mathbb{K}times E/E'rightarrow E/E'$ follows.
To construct charts on $E/E'$ we use the previous remark on local sections to construct fibrewise linear local trivialisations $(U,varphi)$, $Usubseteq M$, of $rho$ in such a way that the restriction $q|_U$ becomes the quotient $Utimesmathbb{K}^srightarrow Utimesmathbb{K}^{s-r}$ locally modelling the vector space quotients $E_xrightarrow E_x/E'_x$.
$endgroup$
$begingroup$
I checked out lee's text but couldn't find it, you know which page it is?
$endgroup$
– CL.
Jan 4 at 6:44
$begingroup$
As I said, the theorem is not actually discussed by Lee, although the material in his book is a good general reference for submersions in general. Check out Roger Godement's book Introduction to the Theory of Lie Groups, section 4.9, Theorem 5, page 133 (Springer-Verlag 2017 English Translation) for the exact statement.
$endgroup$
– Tyrone
Jan 4 at 10:06
add a comment |
$begingroup$
We will appeal to Godemont's theorem, which states that if $X$ is a smooth manifold and $R$ is an equivalence relation on $X$, then $X/R$ has the structure of a smooth manifold such that $Xrightarrow X/R$ becomes a smooth submersion if and only if $Rsubseteq Xtimes X$ is a smoooth submanifold, and $pr_1:Rrightarrow X$ is a submersion. I can't find the exact reference I was looking for at the moment for this theorem, but it is a standard result. I know Bourbaki has a statement (its in French). A good general reference is section 4 of Lee's book Introduction to Smooth Manifolds.
To this end we consider the fibred product $Etimes_ME={(e_1,e_2)in Etimes Emid pi(e_1)=pi(e_2)}$, which is the pullback of $pi$ along itself. This space is a smooth submanifold of the product $Etimes E$. Likewise we consider $Etimes_ME'={(e,e')in Etimes E'mid pi(e)=pi'(e')}$, which, under the assumption that $E'subseteq E$ is a smooth submanifold, is itself a smooth submanifold $Etimes_ME'subseteq Etimes_ME$.
We however, will consider a non-standard embedding of this manifold. Since $E$ is a vector bundle it has a fibrewise sum $mu:Etimes_MErightarrow E$, which restricts to the standard vector space addition $(e_1,e_2)mapsto e_1+e_2$ on each fibre $E_ptimes E_p$, and we use this map to define
$$Theta:Etimes_ME'rightarrow Etimes_ME,qquad Theta(e,e')=(e,e+e').$$
It is clear that this map is a diffeomorphism onto its image, and that its image $Theta(Etimes_ME'):=mathcal{R}$ is precisely the graph of the relation $e_1sim e_2$ iff $pi(e_1)=pi(e_2)$ and $e_2-e_1in E'_{pi(e_1)}$. Then $mathcal{R}subseteq Etimes_MEsubseteq Etimes E$ is a smooth submanifold, the projection $pr_1:mathcal{R}rightarrow E$ is smooth, and we can appeal to Godemont's critereon to get a smooth structure on the quotient $E/E'$ such that the canonical map $q:Erightarrow E/E'$ becomes a smooth submersion. This then induces from the map $pi$ a submersive projection $rho:E/E'rightarrow M$ satisfying $pi=rhocirc q$. Submersions always have local sections, so $E/E'$ is a locally trivial fibre bundle over $M$. In fact, if $s$ is any local section of $pi$, then $qcirc s$ is a local section of $rho$.
Now from here it is not difficult to see that the fibrewise linear structure on $E$ transfers to $E/E'$ in such a way as to make $rho$ into a topological vector bundle, although ee have not yet shown that the induced operations of fiberwise addition and scalar multiplication are smooth. However this is not hard to do using the fact that $q$ is a submersion. In fact it is not difficult to see that $qtimes_Mq:Etimes_MErightarrow E'times_ME'$ is a submersion, and so induces the smooth fibrewise addition $mu':(E/E')times_M(E/E')rightarrow E/E'$ uniquely from $mu$. Similarly, the smoothness of the fibrewise scalar multiplicaiton $mathbb{K}times E/E'rightarrow E/E'$ follows.
To construct charts on $E/E'$ we use the previous remark on local sections to construct fibrewise linear local trivialisations $(U,varphi)$, $Usubseteq M$, of $rho$ in such a way that the restriction $q|_U$ becomes the quotient $Utimesmathbb{K}^srightarrow Utimesmathbb{K}^{s-r}$ locally modelling the vector space quotients $E_xrightarrow E_x/E'_x$.
$endgroup$
$begingroup$
I checked out lee's text but couldn't find it, you know which page it is?
$endgroup$
– CL.
Jan 4 at 6:44
$begingroup$
As I said, the theorem is not actually discussed by Lee, although the material in his book is a good general reference for submersions in general. Check out Roger Godement's book Introduction to the Theory of Lie Groups, section 4.9, Theorem 5, page 133 (Springer-Verlag 2017 English Translation) for the exact statement.
$endgroup$
– Tyrone
Jan 4 at 10:06
add a comment |
$begingroup$
We will appeal to Godemont's theorem, which states that if $X$ is a smooth manifold and $R$ is an equivalence relation on $X$, then $X/R$ has the structure of a smooth manifold such that $Xrightarrow X/R$ becomes a smooth submersion if and only if $Rsubseteq Xtimes X$ is a smoooth submanifold, and $pr_1:Rrightarrow X$ is a submersion. I can't find the exact reference I was looking for at the moment for this theorem, but it is a standard result. I know Bourbaki has a statement (its in French). A good general reference is section 4 of Lee's book Introduction to Smooth Manifolds.
To this end we consider the fibred product $Etimes_ME={(e_1,e_2)in Etimes Emid pi(e_1)=pi(e_2)}$, which is the pullback of $pi$ along itself. This space is a smooth submanifold of the product $Etimes E$. Likewise we consider $Etimes_ME'={(e,e')in Etimes E'mid pi(e)=pi'(e')}$, which, under the assumption that $E'subseteq E$ is a smooth submanifold, is itself a smooth submanifold $Etimes_ME'subseteq Etimes_ME$.
We however, will consider a non-standard embedding of this manifold. Since $E$ is a vector bundle it has a fibrewise sum $mu:Etimes_MErightarrow E$, which restricts to the standard vector space addition $(e_1,e_2)mapsto e_1+e_2$ on each fibre $E_ptimes E_p$, and we use this map to define
$$Theta:Etimes_ME'rightarrow Etimes_ME,qquad Theta(e,e')=(e,e+e').$$
It is clear that this map is a diffeomorphism onto its image, and that its image $Theta(Etimes_ME'):=mathcal{R}$ is precisely the graph of the relation $e_1sim e_2$ iff $pi(e_1)=pi(e_2)$ and $e_2-e_1in E'_{pi(e_1)}$. Then $mathcal{R}subseteq Etimes_MEsubseteq Etimes E$ is a smooth submanifold, the projection $pr_1:mathcal{R}rightarrow E$ is smooth, and we can appeal to Godemont's critereon to get a smooth structure on the quotient $E/E'$ such that the canonical map $q:Erightarrow E/E'$ becomes a smooth submersion. This then induces from the map $pi$ a submersive projection $rho:E/E'rightarrow M$ satisfying $pi=rhocirc q$. Submersions always have local sections, so $E/E'$ is a locally trivial fibre bundle over $M$. In fact, if $s$ is any local section of $pi$, then $qcirc s$ is a local section of $rho$.
Now from here it is not difficult to see that the fibrewise linear structure on $E$ transfers to $E/E'$ in such a way as to make $rho$ into a topological vector bundle, although ee have not yet shown that the induced operations of fiberwise addition and scalar multiplication are smooth. However this is not hard to do using the fact that $q$ is a submersion. In fact it is not difficult to see that $qtimes_Mq:Etimes_MErightarrow E'times_ME'$ is a submersion, and so induces the smooth fibrewise addition $mu':(E/E')times_M(E/E')rightarrow E/E'$ uniquely from $mu$. Similarly, the smoothness of the fibrewise scalar multiplicaiton $mathbb{K}times E/E'rightarrow E/E'$ follows.
To construct charts on $E/E'$ we use the previous remark on local sections to construct fibrewise linear local trivialisations $(U,varphi)$, $Usubseteq M$, of $rho$ in such a way that the restriction $q|_U$ becomes the quotient $Utimesmathbb{K}^srightarrow Utimesmathbb{K}^{s-r}$ locally modelling the vector space quotients $E_xrightarrow E_x/E'_x$.
$endgroup$
We will appeal to Godemont's theorem, which states that if $X$ is a smooth manifold and $R$ is an equivalence relation on $X$, then $X/R$ has the structure of a smooth manifold such that $Xrightarrow X/R$ becomes a smooth submersion if and only if $Rsubseteq Xtimes X$ is a smoooth submanifold, and $pr_1:Rrightarrow X$ is a submersion. I can't find the exact reference I was looking for at the moment for this theorem, but it is a standard result. I know Bourbaki has a statement (its in French). A good general reference is section 4 of Lee's book Introduction to Smooth Manifolds.
To this end we consider the fibred product $Etimes_ME={(e_1,e_2)in Etimes Emid pi(e_1)=pi(e_2)}$, which is the pullback of $pi$ along itself. This space is a smooth submanifold of the product $Etimes E$. Likewise we consider $Etimes_ME'={(e,e')in Etimes E'mid pi(e)=pi'(e')}$, which, under the assumption that $E'subseteq E$ is a smooth submanifold, is itself a smooth submanifold $Etimes_ME'subseteq Etimes_ME$.
We however, will consider a non-standard embedding of this manifold. Since $E$ is a vector bundle it has a fibrewise sum $mu:Etimes_MErightarrow E$, which restricts to the standard vector space addition $(e_1,e_2)mapsto e_1+e_2$ on each fibre $E_ptimes E_p$, and we use this map to define
$$Theta:Etimes_ME'rightarrow Etimes_ME,qquad Theta(e,e')=(e,e+e').$$
It is clear that this map is a diffeomorphism onto its image, and that its image $Theta(Etimes_ME'):=mathcal{R}$ is precisely the graph of the relation $e_1sim e_2$ iff $pi(e_1)=pi(e_2)$ and $e_2-e_1in E'_{pi(e_1)}$. Then $mathcal{R}subseteq Etimes_MEsubseteq Etimes E$ is a smooth submanifold, the projection $pr_1:mathcal{R}rightarrow E$ is smooth, and we can appeal to Godemont's critereon to get a smooth structure on the quotient $E/E'$ such that the canonical map $q:Erightarrow E/E'$ becomes a smooth submersion. This then induces from the map $pi$ a submersive projection $rho:E/E'rightarrow M$ satisfying $pi=rhocirc q$. Submersions always have local sections, so $E/E'$ is a locally trivial fibre bundle over $M$. In fact, if $s$ is any local section of $pi$, then $qcirc s$ is a local section of $rho$.
Now from here it is not difficult to see that the fibrewise linear structure on $E$ transfers to $E/E'$ in such a way as to make $rho$ into a topological vector bundle, although ee have not yet shown that the induced operations of fiberwise addition and scalar multiplication are smooth. However this is not hard to do using the fact that $q$ is a submersion. In fact it is not difficult to see that $qtimes_Mq:Etimes_MErightarrow E'times_ME'$ is a submersion, and so induces the smooth fibrewise addition $mu':(E/E')times_M(E/E')rightarrow E/E'$ uniquely from $mu$. Similarly, the smoothness of the fibrewise scalar multiplicaiton $mathbb{K}times E/E'rightarrow E/E'$ follows.
To construct charts on $E/E'$ we use the previous remark on local sections to construct fibrewise linear local trivialisations $(U,varphi)$, $Usubseteq M$, of $rho$ in such a way that the restriction $q|_U$ becomes the quotient $Utimesmathbb{K}^srightarrow Utimesmathbb{K}^{s-r}$ locally modelling the vector space quotients $E_xrightarrow E_x/E'_x$.
answered Jan 2 at 12:17
TyroneTyrone
5,47711228
5,47711228
$begingroup$
I checked out lee's text but couldn't find it, you know which page it is?
$endgroup$
– CL.
Jan 4 at 6:44
$begingroup$
As I said, the theorem is not actually discussed by Lee, although the material in his book is a good general reference for submersions in general. Check out Roger Godement's book Introduction to the Theory of Lie Groups, section 4.9, Theorem 5, page 133 (Springer-Verlag 2017 English Translation) for the exact statement.
$endgroup$
– Tyrone
Jan 4 at 10:06
add a comment |
$begingroup$
I checked out lee's text but couldn't find it, you know which page it is?
$endgroup$
– CL.
Jan 4 at 6:44
$begingroup$
As I said, the theorem is not actually discussed by Lee, although the material in his book is a good general reference for submersions in general. Check out Roger Godement's book Introduction to the Theory of Lie Groups, section 4.9, Theorem 5, page 133 (Springer-Verlag 2017 English Translation) for the exact statement.
$endgroup$
– Tyrone
Jan 4 at 10:06
$begingroup$
I checked out lee's text but couldn't find it, you know which page it is?
$endgroup$
– CL.
Jan 4 at 6:44
$begingroup$
I checked out lee's text but couldn't find it, you know which page it is?
$endgroup$
– CL.
Jan 4 at 6:44
$begingroup$
As I said, the theorem is not actually discussed by Lee, although the material in his book is a good general reference for submersions in general. Check out Roger Godement's book Introduction to the Theory of Lie Groups, section 4.9, Theorem 5, page 133 (Springer-Verlag 2017 English Translation) for the exact statement.
$endgroup$
– Tyrone
Jan 4 at 10:06
$begingroup$
As I said, the theorem is not actually discussed by Lee, although the material in his book is a good general reference for submersions in general. Check out Roger Godement's book Introduction to the Theory of Lie Groups, section 4.9, Theorem 5, page 133 (Springer-Verlag 2017 English Translation) for the exact statement.
$endgroup$
– Tyrone
Jan 4 at 10:06
add a comment |
$begingroup$
You have to construct compatible trivializations of the bundles $E'$ and $E$, which is slightly complicated to express in terms of transition functions. The easilest way to describe this in my opinion is in terms of local frames. Choose a local frame for $E'$ defined on $U$. Shrinking $U$, you can extend this to a local frame for $E$. (In a point $xin U$, extend the values of your frame which form a basis for $E'_x$ to a basis for $E_x$ and then extend the additional elements to local smooth sections of $E$.) Then you get a local trivialization $E|_Uto Utimesmathbb R^n$ which restricts to a trivialzation $E'|_Uto Utimesmathbb R^k$. From this, you get a trivialization from the quotient bundle to $mathbb R^n/mathbb R^kcongmathbb R^{n-k}$.
In terms of transition functions, you can use the above construction to show that you can arrange things in such a way that the transtition functions for $E$ have values in block matrices of the form $begin{pmatrix} A & B \ 0 & Cend{pmatrix}$ such that the $A$-component gives you transition functions for $E'$. In this setting, the $C$-block gives you the transition functions for $E/E'$.
$endgroup$
add a comment |
$begingroup$
You have to construct compatible trivializations of the bundles $E'$ and $E$, which is slightly complicated to express in terms of transition functions. The easilest way to describe this in my opinion is in terms of local frames. Choose a local frame for $E'$ defined on $U$. Shrinking $U$, you can extend this to a local frame for $E$. (In a point $xin U$, extend the values of your frame which form a basis for $E'_x$ to a basis for $E_x$ and then extend the additional elements to local smooth sections of $E$.) Then you get a local trivialization $E|_Uto Utimesmathbb R^n$ which restricts to a trivialzation $E'|_Uto Utimesmathbb R^k$. From this, you get a trivialization from the quotient bundle to $mathbb R^n/mathbb R^kcongmathbb R^{n-k}$.
In terms of transition functions, you can use the above construction to show that you can arrange things in such a way that the transtition functions for $E$ have values in block matrices of the form $begin{pmatrix} A & B \ 0 & Cend{pmatrix}$ such that the $A$-component gives you transition functions for $E'$. In this setting, the $C$-block gives you the transition functions for $E/E'$.
$endgroup$
add a comment |
$begingroup$
You have to construct compatible trivializations of the bundles $E'$ and $E$, which is slightly complicated to express in terms of transition functions. The easilest way to describe this in my opinion is in terms of local frames. Choose a local frame for $E'$ defined on $U$. Shrinking $U$, you can extend this to a local frame for $E$. (In a point $xin U$, extend the values of your frame which form a basis for $E'_x$ to a basis for $E_x$ and then extend the additional elements to local smooth sections of $E$.) Then you get a local trivialization $E|_Uto Utimesmathbb R^n$ which restricts to a trivialzation $E'|_Uto Utimesmathbb R^k$. From this, you get a trivialization from the quotient bundle to $mathbb R^n/mathbb R^kcongmathbb R^{n-k}$.
In terms of transition functions, you can use the above construction to show that you can arrange things in such a way that the transtition functions for $E$ have values in block matrices of the form $begin{pmatrix} A & B \ 0 & Cend{pmatrix}$ such that the $A$-component gives you transition functions for $E'$. In this setting, the $C$-block gives you the transition functions for $E/E'$.
$endgroup$
You have to construct compatible trivializations of the bundles $E'$ and $E$, which is slightly complicated to express in terms of transition functions. The easilest way to describe this in my opinion is in terms of local frames. Choose a local frame for $E'$ defined on $U$. Shrinking $U$, you can extend this to a local frame for $E$. (In a point $xin U$, extend the values of your frame which form a basis for $E'_x$ to a basis for $E_x$ and then extend the additional elements to local smooth sections of $E$.) Then you get a local trivialization $E|_Uto Utimesmathbb R^n$ which restricts to a trivialzation $E'|_Uto Utimesmathbb R^k$. From this, you get a trivialization from the quotient bundle to $mathbb R^n/mathbb R^kcongmathbb R^{n-k}$.
In terms of transition functions, you can use the above construction to show that you can arrange things in such a way that the transtition functions for $E$ have values in block matrices of the form $begin{pmatrix} A & B \ 0 & Cend{pmatrix}$ such that the $A$-component gives you transition functions for $E'$. In this setting, the $C$-block gives you the transition functions for $E/E'$.
answered Jan 2 at 11:45
Andreas CapAndreas Cap
11.5k923
11.5k923
add a comment |
add a comment |
Thanks for contributing an answer to Mathematics Stack Exchange!
- Please be sure to answer the question. Provide details and share your research!
But avoid …
- Asking for help, clarification, or responding to other answers.
- Making statements based on opinion; back them up with references or personal experience.
Use MathJax to format equations. MathJax reference.
To learn more, see our tips on writing great answers.
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f3059300%2fsmooth-quotient-bundle%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
$begingroup$
If you choose a Riemannian metric on your bundle this is easy: it's just the orthogonal complement.
$endgroup$
– user98602
Jan 2 at 16:12