## Saturday, 6 June 2020

### Question

Consider a comoving observer sitting at constant spatial coordinates $\left(r_\ast\ ,\theta_\ast\ ,\ \phi_\ast\right)$ around a Schwarzschild black hole of mass $M$. The observer drops a beacon onto the black hole (straight down along a radial trajectory). The beacon emits radiation at a constant wavelength $\lambda_{em}$ (in the beacon rest frame).

a) Calculate the coordinate speed of the beacon as a function of $r$.
b) Calculate the proper speed of the beacon. That is, imagine there is a comoving observer at fixed $r$, with a locally inertial coordinate system set up as the beacon passes by, and calculate the speed as measured by the comoving observer. What is it at $r=2GM$?
c) Calculate the wavelength $\lambda_{obs}$, measured by the observer at $r_\ast$, as a function of the radius $r_{em}$ at which the radiation was emitted.
d) Calculate the time $t_{obs}$ at which a beam emitted by the beacon at radius $r_{em}$ will be observed at $r_\ast$.
e) Show that at late times, the redshift grows exponentially: $\lambda_{obs}/\lambda_{em}\propto e^{t_{obs}/T}$. Give an expression for the time constant $T$ in terms of the black hole mass $M$.

This question is a fascinating can of worms. I needed the help of an unwitting mentor Jeriek Van den Abeele from the University of Oslo. The first crucial help was to use the timelike killing vector constant $$E=\left(1-\frac{2GM}{r}\right)\frac{dt}{d\tau}$$This came in useful not only for the beacon velocities but also for the  travel time of the photon back to the observer at $\ r_\ast$ in question d.

I also needed help from Physics Forums on comoving coordinates and I think I understand them now, although there was some dispute about how the (second)  comoving observer in question (b) would actually achieve that state. When people talk about comoving coordinates for the Universe, they are talking about something quite different from here: Coordinates comoving with the Universe are growing with the Universe.

Part of the worminess arises from confusion about what coordinates are referring to what. There was a profusion of subscripts: $r_{em},r_\ast,\ R_s,t_\gamma,t_{em},t_{obs},t_b$. The last of those was introduced gratuitously by my mentor who I refer to as Oslo. In my opinion $t_b\equiv t_{em},\ \ r_b\equiv r_{em}$. I try to avoid these subscripts as much as possible.

I got two answers to question (b). One was the proper speed calculated in the strange comoving inertial coordinate system, as Carroll asked, and the other was the proper speed in Schwarzschild coordinates. I called them $dr^\prime/d\tau^\prime$ and $dr/d\tau$. The first was what Carroll asked for and can be used (but is not essential) in question (c). Apart from that I am not sure how useful it is. $dr/d\tau$ is what would be experienced by an astronaut falling with the beacon and it does not reach the speed of light at the event horizon.

Interestingly I calculated the answer to (c) using Doppler redshift + gravitational redshift and Oslo did it in one leap - which was more complex and contained a small error which had no effect. The formulas in the two answers looked quite different but when plotted gave the same lines. Eventually I proved that the formulas were in fact the same.

As usual the actual geodesic equations are not used to find out about all these geodesics. Nevertheless I will have a try to see if I can do better with that beacon in section 5.6.

Read it all at Ex 5.5 Observer and beacon.pdf (11 pages not including other documents)

1. Here's my solution to part (b). From part (a), the time component of the 4-velocity of the beacon is $$v^0 = \left[\frac{1}{1-2GM/r}\frac{r(r_*-2GM)}{ r_*(r-2GM)} \right]^{1/2}.$$ The observer's 4-velocity is $$v'_\mu = ((1-2GM/r_{obs})^{1/2},0,0,0).$$ So we have $$v^\mu v'_\mu = \left[\frac{1-2GM/r_{obs}}{1-2GM/r}\frac{r(r_*-2GM)}{ r_*(r-2GM)} \right]^{1/2}.$$ Now in a locally inertial coordinate system, where the metric is Minkowski, this quantity is nothing other than $1/\sqrt{1-\vec{v}^2}$. (This is actually special relativity stuff.) With $r=r_{obs} = 2GM$, we have $v^\mu v'_\mu = \infty$ and therefore $\vec{v} = 1$, where $\vec{v}$ is the 3-velocity of the beacon observed by the observer who suddenly decided to fall freely as the beacon passes by (and making its own frame a locally inertial coordinate system).
1. Now in a locally inertial coordinate system, where the metric is Minkowski, this quantity is nothing other than $1/\sqrt{1-\vec{v}^2}$. (This is actually special relativity stuff.) With $r=r_{obs} = 2GM$, we have $v^\mu v'_\mu = \infty$ and therefore $\vec{v} = 1$, where $\vec{v}$ is the 3-velocity of the beacon observed by the observer who suddenly decided to fall freely as the beacon passes by (and making its own frame a locally inertial coordinate system).