I am reading the paper on Gravitational Waves in General Relativity. VII. Waves from Axi-Symmetric Isolated Systems by H. Bondi, M. G. J. van der Burg, A. W. K. Metzner. (link) Here is a quote(s) from that paper

Throughout this paper, we shall suppose that the 4-space is axially symmetric and reflexion symmetrical....

...... From the axial symmetry, the azimuth angle is readily defined. Suppose we now put a source of light at a point $O$ on the axis of symmetry and surround it by a small sphere on which we can produce the azimuth co-ordinate $\phi$ together with a colatitude $\theta$ and a time co-ordinate $u$. We then define the $u$, $\theta$, $\phi$ co-ordinates of an arbitrary event $E$ to be the $u$, $\theta$, $\phi$ co-ordinates of the event at which the light ray $OE$ intersects the small sphere. In other words, along an outward radial light ray the three co-ordinates $u$, $\theta$, $\phi$ are constant. If we wish to write down the metric for such a system of co-ordinates (in which the part referring to the azimuth angle $\phi$ appears separately) then we know that since only the co-ordinate $r$ varies along a light ray, the term $g_{11}$ of the metric tensor must vanish, the four co-ordinates $u$, $r$, $\theta$, $\phi$ being denoted by 0, 1, 2, 3 in that order. Moreover, we must have $$\Gamma^0_{11} = \Gamma^2_{11}$$

QUESTION: Why do the above Christoffel symbols vanish?

EDIT: QUESTION 2: What does one mean by reflexion symmetry?


I finally figured it out. It's simple really. He requires that we have radial null geodesics right? This implies that the geodesic equation $$ \frac{d^2 x^\lambda}{d \tau^2} + \Gamma^\lambda_{\mu\nu} \frac{d x^\mu}{d \tau} \frac{d x^\nu}{d \tau} = 0 $$ should be solved by $(u,\theta,\phi)$ constant. Plugging this into the equation above, we immediately find the constraints $$ \Gamma^0_{11} = \Gamma^2_{11} = \Gamma^3_{11} = 0 $$ He doesn't mention $\Gamma^3_{11}$ because it is trivially satisfied for the metric at hand, since $$ \Gamma^3_{11} = \frac{1}{2} g^{3\rho} \left( 2 g_{\rho 1, 1} - g_{11,\rho} \right) = \frac{1}{2} g^{33} \left( 2 g_{3 1, 1} - g_{11,3} \right) = 0 $$ where we have used $g_{11} = 0$ ($(u,\theta,\phi)$ constant along null geodesics), $\partial_3 g_{\mu\nu} = 0$ (axial symmetry) and $g_{03}=g_{13}=g_{23} = 0$(reflexion symmetry)


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.