In general, the redshift of a photon emitted from one static observer at point $A$ to another static observer located at point $B$ is of the form

$$ z_{AB} = \frac{\omega_A}{\omega_B} - 1 $$

where $\omega_A$ is the frequency of the photon observed in observer A's frame of reference. I understand how, in the presence of a weak gravitational field, we may take

$$ \frac{\omega_A}{\omega_B} = \sqrt{\frac{g_{00}(R + L)}{g_{00}(R)}} \approx \sqrt{\frac{1 + 2\Phi(R + L)}{1 + 2\Phi(R)}} $$

where $g_{00}$ is the time-time component of the metric, $\Phi(r) = -\frac{GM}{r}$ is the classical Newtonian potential, $R$ is the radius of the Earth, and $L$ is the distance from Earth's surface to the static observer at point B. Furthermore, I know what the redshift should be:

$$ z_{AB} = \frac{gL}{c^2} $$

where $g = \frac{GM}{R^2}$ is the acceleration due to gravity and we've divided by $c^2$ to get the correct units. We may easily verify that

$$ z_{AB} \approx \Phi(R + L) - \Phi(R) $$

reproduces the correct result. However, the problem I am facing is in demonstrating that this follows from the expression I have above for $\frac{\omega_A}{\omega_B}$. Specifically, this is what my professor did:

\begin{eqnarray} \sqrt{\frac{1 + 2\Phi(R + L)}{1 + 2\Phi(R)}} &\approx \sqrt{1 + 2[\Phi(R + L) - \Phi(R)]}\\ &\approx 1 + [\Phi(R + L) - \Phi(R)]. \end{eqnarray}

The second approximation is not mysterious. We're simply using a Taylor expansion of the form

$$ \sqrt{1 + x} \approx 1 + \frac{1}{2}x + \cdots $$ but I have no idea where the first approximation comes from. If it's a Taylor expansion of some sort, I don't know how to formulate it, let alone verify it. I'd really appreciate it if someone could fill in this gap in my understanding of what is otherwise a very neat derivation. Thanks!


It's the same idea, except now the expansion is $$\frac{1}{1+x} \approx 1 - x$$

  • $\begingroup$ Yeah, I figured it out xD. You treat the term in the numerator as a constant and take the Taylor expansion of the denominator. I was confused because I was assuming that you had to expand using both expressions involving Phi. I was wrong. $\endgroup$ Nov 28 '21 at 20:03

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.