Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Reeb Fields and Contact Fields.

  1. Oct 24, 2013 #1

    WWGD

    User Avatar
    Science Advisor
    Gold Member

    Hi again:

    I'm curious as to someone understands well the difference between a Reeb Vector Field and a general Contact field; a Contact field is one whose flow (even when only defined locally, i.e., in non-complete manifolds) preserves the contact structure, but not the form, i.e., if C is a Contact field with local flow[itex]\Phi[/itex] , and w is the contact form, then:

    [itex]\Phi[/itex]*w=gw , where g is a smooth nowhere-zero function ; here gw and w are equivalent, in that the kernel of gw is the same as that of w . Similarly, we have that [itex]\Phi[/itex]* takes contact planes to contact planes, i.e., the basis tangent vectors for the contact plane at p are pushed forward to the tangent plane at (p+t).

    OTOH, we have, for a Reeb field Rw for w, that Rw preserves not just the contact structure, but the contact form itself, i.e., if the flow of Rw is given by [itex]\Psi[/itex], then [itex]\Psi[/itex]*(w)=w .

    So every Reeb field is a contact field, but not otherwise. I guess in the case of the Reeb field Rw, its flow [itex]\Psi[/itex] is a sort of path of contact-form-preserving maps, i.e., for each teach [itex]\Psi[/itex]t preserves w. Since Rw is also a contact field, I guess in the case of the Reeb field we can somehow normalize the function g (since the flow a Contact field C takes w to gw ), so that g==1.

    Does anyone understand well what is going on here?
     
  2. jcsd
  3. Oct 24, 2013 #2

    Ben Niehoff

    User Avatar
    Science Advisor
    Gold Member

    Everything you've written down makes sense. What's your actual question? Do you understand flows and pushforwards?
     
  4. Oct 24, 2013 #3

    WWGD

    User Avatar
    Science Advisor
    Gold Member

    Yes, to a pretty good extent, I think. I would just like to know in some sense how/why a Reeb field is "stronger" than a non-Reeb contact field. Still, let me think things thru again see if I can clarify my question some more. Thanks, Ben.
     
  5. Oct 25, 2013 #4

    WWGD

    User Avatar
    Science Advisor
    Gold Member

    Actually, this is a more specific question: there is a result that a contact field Vthat is transverse to the contact pages is a Reeb field. I'm trying to show this is so; from other results, it follows that if the field V is transverse, then it will preserve not just the contact structure, but the contact form itself. I'm trying to show this.

    Please let me ramble-on a bit, see if my overall knowledge of Contact Structures is accurate; I will be giving a talk a few months from now and I want to see if/where I have gaps(more likely where than if ;) ), and I'm trying to put all these terms together. Please give me some time until I learn this version of Latex.

    Re my understanding, it is mostly formal, but not very geometric (given this is low-dimensional topology/geometry).

    I understand the flow associated to a vector field V in a manifold M to be a curve C(t); C:(-e,e)-->M with

    C'(t)=V(C(t)), i.e., the derivative of C at t coincides with the value of the vector field at that point,

    and that flows are guaranteed to exist (at least) locally by , I think, one of Picard's theorems.

    More rigorously, the flow ψ of V maps a point (t,p) in ℝxM-->M by sending (p,t) to the curve C(t)

    with:

    C(0)=p ; C'(t)=V(C(t))

    The pushforward ψ* associated with a map, say the flow , maps tangent vectors at

    TpM to TpM , mapping a deriva

    I'm also having trouble understanding or seeing a more geometric interpretation of the result
    LRw ω=0 , where L is the Lie derivative of ω. This means that the form ω is constant along the flow, but, what does this mean? I have some idea of what a constant vector field is, but I'm having trouble digesting what a constant tensor field is, or even what dω(Rw,.)=0 means geometrically.

    So, to be more specific in my questions:

    i)How do we show that if V is a contact field for ω , and V is transverse to the contact planes, then V is a Reeb field for ω .

    ii)How does one interpret the result L ω =0 ? This says, AFAIK, that ω is constant along the flow. BUT: what is the meaning of a form being constant, or a tensor field being constant?

    Thanks.
     
  6. Oct 28, 2013 #5

    WWGD

    User Avatar
    Science Advisor
    Gold Member

    Sorry if I rambled-on too much. My basic question is this:

    How to we show that a contact vector field (i.e., a v.field whose flow preserves the contact structure--but not necessarily the contact form, other than up to a multiple of the form by a nonzero smooth function) that is transverse to the contact planes is a Reeb field?

    A contact field is one whose flow preserves the contact form, meaning , for w the contact form , $$V_*$$ the flow of V, we have $$ V_*(w)=gw$$ , but the flow $$R_w* $$ of a Reeb field $$ R_w$$ preserves the form itself, i.e., $$R_w (w)* =w$$ , and the Reeb field satisfies:

    $$ w(R_w) $$ =1 , which basically says $$w(R_w) $$ is never 0.

    Now, I can see why $$ w(R_w)$$ is not zero if $$ R_w$$ is transverse, since this means $$R_w$$ is never in the contact planes, and the contact planes are, by definition, the kernel of $$w$$

    Now, I have no clue of the other par:, why the transversality of $$R_w$$ means that the flow of $$R_w$$ preserves the form. Any ideas?
    Thanks.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook