Exploding Supernova Evidence
Here again is the Type Ia supernova data. Low redshifts increase linearly with apparent magnitude, which is how Edwin Hubble concluded that the Universe expands. Theory predicts that ci/co = sqrt(1 + Z), so when light of redshift Z was emitted, c was faster by sqrt(1 + Z). This leads to a precise solution of the "accelerating universe" problem.
An object of redshift Z = 1.0 recedes at 60% of today's speed of light. That is only 42% of c at time that light was emitted. The redshift we measure is z = 0.57. For an object of absolute redshift Z = 0.5, apparent z = 0.38 (horizontal arrows). Energy conversion E=mc^2 is also affected by c change. For an object of absolute Z = 0, that energy is doubled for a magnitude change of -0.75 (vertical arrow).
For low redshifts, sqrt(1 + Z) is negligible and the line is straight. Only at high redshifts does the line curve upward, making the universe appear to accelerate. The black prediction line is difficult to see because it is covered in data points. In the lower right it is blown up bigger. Within the error bars, the prediction still intercepts 95% of data points. This has been accomplished without dark energies, cosmic constants, or other epicycles. When a prediction fits the data this closely, what does it mean?
The sqrt(1 + Z) factor explains the "jerk parameter," why redshifts increase linearly at low Z and then curve upward. Corroborating data comes from the geological record compared to the "Faint Young Sun." If c had not changed in exactly the amounts predicted, Earth would have been frozen solid and we would not be around. People believed the accelerating universe because it came from two "independent" groups both on the Berkeley campus and in constant communication. Now there are two lines of evidence from truly different sources pointing to c change in physics.