To go into a little bit more detail how a mathematical solution for a black hole is used to correct satellite clocks the story goes back to 1959.
An explanation of a metric can be found in this
post.
The Schwarzschild metric for a non rotating black hole is found to be;
ds² = c²(1- 2GM/c²r) dt² - dr²/(1- 2GM/ c²r) - r²(dθ² +sinθdφ²)
The metric is also a good approximation for any slow rotating body such as the Earth or Sun.
In 1959 two smart dudes Pound and Rebka who went on to win the Nobel Prize in physics proposed an experiment for two stationary clocks in a gravitational field.
For a stationary clock dr = dθ = dφ = 0 and the Schwarzschild metric reduces to:
ds² = c²(1- 2GM/c²r) dt²
If the clock is an atom or molecule which emits or absorbs radiation at a given frequency f which is unaffected by gravitational forces ds = cdτ or dτ = ds/c.
Substituting this into the reduced Schwarzschild equation gives;
dτ = √(1- 2GM/c²r)dt
For two stationary clocks in a gravitational field, the ratio of the clock time intervals between a given pair of events is defined as;
dτ₁/dτ₂ = √[(1- 2GM/c²r₁)/(1- 2GM/c²r₂)] measured by observers at radii r₁ and r₂.
If 2MG << c²r then √(1- 2GM/c²r) ≈ 1-GM/ c²r using the binomial theorem approximation and then dividing the equation gives;
dτ₂ = dτ₁[1-G/c²(M/r₂-M/r₁)]
Note how this equation looks very similar to the equation for the time dilation of a satellite clock due to gravity.
Δt = dτ[1-G/c²(M/r₂-M/r₁)]
In the experiment since both clocks, one of which emits radiation, the other absorbs radiation, are stationary the coordinate time Δt is the same as the proper time dτ.
Since the proper time is simply the inverse of the frequency f, the equation;
dτ₂ = dτ₁[1-G/c²(M/r₂-M/r₁)]
can be rewritten as
fₑ = fₐ[1-G/c²(M/r₂-M/r₁)]
where fₑ is the emitted frequency and fₐ is the absorbed frequency.
What Pound and Rebka wanted to show the frequency of the clocks changed according to where the clocks were located in the gravitational field.
They set up an experiment where they placed radioactive ⁵⁷Fe which emits gamma rays inside a loudspeaker cone at the top of the university building and ⁵⁷Fe in the basement which was the absorber.
The vertical distance between the absorber and emitter was 22.6 meters.
⁵⁷Fe should emit and absorb gamma ray photons at the same frequency; since emitter and absorber were at different levels in the gravitational field this did not occur since photons reaching the absorber were gravitationally blueshifted.
By adjusting the vibrations of the loudspeaker for the emitter they induced a Doppler shift which cancelled out the gravitational blueshift so that absorption did occur.
Mathematically the condition is.
fₑ/fₐ = [1-G/c²(M/r₂-M/r₁)] = √[(c-v)/(c+v)]
The last term is the formula for Doppler blueshift and v is the approach velocity of emitter inside the vibrating loudspeaker which is relative to the absorber when ⁵⁷Fe emits and absorbs photons at the same frequency.
By Einstein’s equivalence principle, absorption occurred when the upward acceleration of the loudspeaker was equal to –g, the acceleration due to gravity but in the opposite direction.
If the distance between the emitter and detector is d and the time of flight of a photon travelling from the emitter to absorber is d/c, then approach velocity v of the emitter is v = gd/c.
fₐ/fₑ = √[(c+v)/(c-v)] = √[(c+gd/c)/(c-gd/c)] ≈ 1 + gd/c².
fₐ/fₑ = 1 + gd/c²
Δf/fₑ= gd/c²
The theoretical or predicted fractional change in shift is;
Δf/fₑ = (9.8 x 22.6)/(3 x 10⁸)² ≈ 2.5 x 10⁻¹⁵ which was about 5% out from the experimentally determined value.