# If I surrounded a black hole with gravitometers and observed the gravitational field as mass passed the event horizon, what would I observe? [duplicate]

By surrounding a black hole with gravitometers I would be able to get a 3-D map of the gravitational field. If I observed this field as an object approaches the event horizon, what would I see?

Would the gravitational field allow me to track the mass on its journey from the event horizon to the singularity? Would this allow me to observe what is going on inside the black hole?

To clarify: At some point in time there are two distinct gravitational fields, one for the approaching object and one for the singularity. At a later time there is just the gravitational field for the singularity. It is the nature of the transition in the gravitational field that I am specifically interested in.

## marked as duplicate by Ben Crowell, Jon Custer, Kyle Kanos, Aaron Stevens, glSOct 22 at 10:30

• to track what happens inside you also have to go inside, since no information can come out. in the frame of the stationary gravitometers the test mass never enters the horizon since gravitational time dilation makes them freeze above the horizon. – Yukterez Oct 11 at 22:48
• What Yuktetez said. But how are you stopping your gravitometers from falling into the black hole? – PM 2Ring Oct 11 at 23:32
• If no information can come out, then please explain what I would observe happening to the gravitational field. – Barnaby Golden Oct 12 at 7:57
• OK, I have modified my question so that is refers to the gravitational field as the object approaches the event horizon and not as it crosses it. I'm still not clear how the gravitational field appears in 3 dimensions as the object approaches. That is the fundamental question I am asking here. At some stage there are two distinct gravitational fields and at a later time there is just the field centred on the singularity. What does the transition look like? – Barnaby Golden Oct 12 at 16:54
• The inverse problem for classical gravitational fields is not well defined, since a small dense sphere has the same (distance) field as a larger sphere of the same mass. So you can't really "see". I don't know about time varying GR. – Keith McClary Oct 13 at 0:26

This problem can be (and has been) studied numerical. One can simulate the gravitational field of a small massive object dropping radially into a (much larger) Schwarzschild black hole. This was done, for example, in arXiv:1012.2028 by Mitsou. (This is simply the first one I found, there are more, and probably much earlier.)

The easiest way to represent the gravitational field is in the form of the Weyl scalar $$\psi_4$$, which contains (almost) all gauge invariant dynamical information about the gravitational field. Further more, it is convenient to write the field as a sum of (spin-weighted) spherical harmonics $$Y_{lm}$$. (You can think of this as the analog of a Fourier series on the sphere.) This conveniently captures the angular dependence of the gravitational field. Moreover, if one picks coordinates such that particle falls along the coordinate axis, the field is axisymmetric, meaning that all but the $$m=0$$ modes are zero.

With this setup one can plot the time dependence of the remaining $$l$$ modes. The plot below (from arXiv:1012.2028) plots the time dependence of the $$l=2$$ mode. As the particle approaches the black hole the field grows monotonically (mostly not shown in this plot) until particle reaches the horizon, after which the field "ringsdown" decaying exponentially while oscillating with a characteristic frequency known as a quasinormal mode (or QNM).

For higher $$l$$ modes the picture is qualitatively similar except that the amplitude is (much) smaller and the frequencies of the QNM ringdown higher. For example here is the $$l=6$$ mode from the same paper.

Note that the time scale for this to happen is quite short. In the plots, the units of time are scaled by the natural time scale for the black hole. For a 10 solar mass black hole one such unit is about 50 microseconds.

• I just want to be clear on this. Is it correct to say that the decaying oscillations can give no indication of the distribution of mass inside the event horizon? – Barnaby Golden Oct 14 at 8:23
• @BarnabyGolden. No, and they can't because inside the horizon is not in the causal past of any outside observer. If the object was actually a bomb set to detonate just after it crosses the horizon, the graphs would look exactly the same. – mmeent Oct 14 at 8:28
• Thank you for your answer. – Barnaby Golden Oct 14 at 10:56
• @safesphere There is no such thing as "an external" frame in general relativity. What does exist are "time-slicings", most of which used in time domain numerical simulations cross the horizon at a finite time. – mmeent Oct 18 at 6:05
• @safesphere There is no unique coordinate system associated to a single physical observer. The closest you can probably come in GR are Fermi-Walker coordinates for an observer. In these, equal time slices will cross the horizon for most observers outside the horizon. – mmeent Oct 18 at 7:54