There are three separate papers that have been accepted for publication in The Astrophysical Journal but haven't appeared yet. Let's see what they say.
The metaphysical equivalent of an Encyclical. Swapping one group of experts for a group of experts with different training doesn't get one anything without the premise that the new experts espouse the one true view.
This is the best science can do when experimentation is impossible: apply theories to measurements.
Most of our science is done without the benefit of a laboratory at all. Historical sciences (paleontology, geology, etc) and cosmology have to rely on other methods. It doesn't make them based on faith.
In other words, in regards to the full moon I am not required to place my trust in others based upon their ritualistic training or privileged access to an unseen world.
'"Our findings are statistically consistent with a head-on collision between the Andromeda galaxy and our Milky Way galaxy," adds Roeland van der Marel of the STScI.'
How the hell does one establish statistical consistency when predicting an event four billion years in the future across 2.5 million light years? The sample size which has been analyzed is zero.
"The Hubble Space Telescope team, led by van der Marel, conducted extraordinarily precise observations of the sideways motion of M31 that remove any doubt that it is destined to collide and merge with the Milky Way."
Here is the fundamental fallacy at heart of the matter - the very one which makes this certainty bad science: the degree of precision with which one makes their measurements has absolutely no bearing on the accuracy of those measurements. Instead it indicates that all the errors were repeatable.
Science does some wonderful things...when it uses its own original methods rather than relying on faith that its high purpose makes its conclusions true.