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

A Confusion regarding conformal transformations

  1. Jun 18, 2016 #1
    I am confused about conformal transformations on Riemannian manifolds. Here's what I have so far.

    1. Under a conformal transformation the metric changes by:

    g' -> Ω2g

    2. Under a Weyl transformation the metric changes by:

    g' -> exp(-2f)g

    3. Any 2D Riemann manifold is locally conformally flat and the metric can be defined in terms of isothermal coordinates.

    g = exp(f)(du2 + dv2 )

    Where u and v are Euclidean.

    How are these all tied together? There appears to be a commonality in form but the multiplying functions are different. Where does the exponential come from?
  2. jcsd
  3. Jun 23, 2016 #2
    Thanks for the post! This is an automated courtesy bump. Sorry you aren't generating responses at the moment. Do you have any further information, come to any new conclusions or is it possible to reword the post?
  4. Jul 14, 2016 #3
    OK, I have modified my original questions. I think (hope) I now have a better understanding. Perhaps somebody could critique this. However, I am still a little confused about the relationship between λ and Ω and the difference in exponents. Thanks.

    Conformal Manifolds:

    A manifold, M, is said to be conformal if two Riemannian metrics, g and g', on the smooth manifold M are related by the equation:

    g'μν = Ω2gμν

    Where Ω is a real-valued smooth function defined on the manifold. Metrics obeying this equation are said to be conformally equivalent.

    This equation represents a general coordinate transformation, x -> x', such that x' = f(x) has the following effect on the metric:

    g'μν = (∂xρ/∂x'μ)(∂xσ/∂x'ν)gρσ

    Under these conditions both the metric and the Weyl tensor are invariant meaning:

    gμν(x)dxμdxν = g'μν(x)dx'μdx'ν or ds2 = ds'2.


    Cabcd = C'abcd

    Conformal Transformations:

    A conformal transformation can be defined as a subset or special case of the above general coordinate transformation which acts as a rescaling of the metric x -> x by a scale factor, λ . This rescaling of the metric is often referred to as WEYL transformation and the 2 terms are often used interchangeably, although they are in fact different things. Therefore:

    gμν = λgμν

    From above:

    gμν = (∂x'ρ/∂xμ)(∂'xσ/∂xν)g'ρσ

    So we can write:

    (∂x'ρ/∂xμ)(∂'xσ/∂xν)g'ρσ = λgμν


    gμν = λ-1{(∂x'ρ/∂xμ)(∂'xσ/∂xν)g'ρσ}

    In other words any coordinate transformation such that g'μν = λgμν can be 'undone' by a Weyl transformation. The Weyl transformation takes us to a coordinate system where the metric has the same form as the one we started with, but the points have all been moved around and pushed closer together or farther apart depending on the scale factor. Therefore, under a Weyl transformation:

    ds2 -> λds2

    The Weyl transformation changes the proper distances at each point by a scalar factor λ; and, therefore, has no directional meaning. It is a local change of scale which preserves the angles between all lines. It is not a coordinate transformation on the space or spacetime.

    Conformal Flatness:

    A necessary condition for a Riemannian manifold to be conformally flat is that the WEYL TENSOR vanish. The interpretation of this is that if a manifold (M,g) has a neighborhood (U,h) at each point such that g'μν = η'μν = Ω2ημν it is locally conformally flat. Under these circumstances the Riemannian metric is conformal to the Euclidean metric and can be described by a local coordinate system in which the metric ds2 satisfies:

    ds2 = Ω2(dx12 + dx22 + ... dxn2)

    These are referred to as ISOTHERMAL COORDINATES.
  5. Jul 15, 2016 #4


    User Avatar
    Science Advisor

    The multiplying functions can be defined as you want and are conventional. In 2d, a metric has 3 components, so gaugefixing your 2 gct's leaves you with one function for the metric. A conformal metric can the be used to write the metric in the form you give.

    Hope this helps.
  6. Jul 17, 2016 #5
    Thanks. I think I am almost there. By convention Ω = exp(2f). Both the CT and the WT are applicable to d ≥ 2 but for d = 2, f = f(z) and is holomorphic. The scale factor becomes |df/dz|2. Correct?
  7. Jul 18, 2016 #6


    User Avatar
    Science Advisor

    You may need to look at this post
    Conformal Invariance Klein Gordon Action
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted