Appeal to authority : a practical example

I don't think I ever fully understood the "appeal to authority" fallacy until a few hours ago.

Scientist A writes a paper on a possible way of solving a long-standing problem in cosmology. Scientist B, who is older and more experienced, says, "my trustworthy collaborator Scientist C says this paper is wrong, here are his arguments." I go away and read the paper and C's arguments and I find in favour of A. I report back to scientist B with a detailed report of why I think C is in error.

Scientist B responds with, "Scientist C and myself are experts in this field, we know it can't be right, therefore how can it be true ?" and nothing else.

By itself, "I know they're wrong because I'm an expert" is a complete non-argument. This should not be confused with, "here are my expert, detailed opinions as to why they're wrong."

I'm still holding out hope that in this particular case B and/or C will eventually follow-up with proper arguments, though I wouldn't be surprised if they didn't.

Keenan's Ring

"Keenan's Ring" is the name we're giving to a giant starless cloud of hydrogen discovered near the Triangulum galaxy, M33. This is the largest, most massive hydrogen cloud discovered in the region since Wright's Cloud in 1979. In angular size it's about as large as the hydrogen disc of M33 (five times larger than the full Moon) and it's just been sitting there all this time, but it's so faint no-one had noticed it before. Well, not quite, but no-one realised just how large it was or that it was a ring.

Why does this matter ? Well, we don't have a good explanation for this object. The nearby Wright's Cloud is thought to be part of the much larger Magellanic Stream, but there's no obvious reason why there should be two large clouds at the end but offset at right-angles to the stream. Nor is it obvious why this one should be a ring - there's no particular reason to expect the gas to be missing in the centre of the structure. It's not likely to be a dark galaxy either (an object made of dark matter, gas, but without stars), because the velocity width is much smaller than would be expected. On the other hand, it does have a small velocity gradient, suggesting that it is a single coherent structure and not a chance alignment of lots of smaller clouds. It really is a mystery.

The figure shows the 3D data cube obtained with five years of Arecibo observations. The third axis is velocity, not distance (see link for details). Colours are chosen just to highlight different structures : blue for the Milky Way, red for everything else. The data looks noisy at one end but this is just because of how the data was processed. Keenan's Ring can be seen in this noisy red part of the data, but there are better images in the linked post.

Placeholder post intended to be replaced with a better summary.

Comparing galaxy environments

The density of our Local Group (left) compared to the Virgo galaxy cluster (right). The view spans about 2 Mpc (6 million light years) in each case; the size of the galaxies has been exaggerated by a factor of 20. Smaller galaxies in the Local Group (which are just too small to show up unless I make everything ridiculously large) are shown as faint transparent fuzzy patches. You can see Andromeda fly past if you watch closely, but Triangulum is still too small.

M33 in HI at different sensitivity levels

With acceptance of the publication now imminent, I suppose one little gif can't hurt...

This is M33 (Triangulum) as seen in the optical Sloan Digital Sky Survey and with atomic hydrogen data from AGES (the Arecibo Galaxy Environment Survey) at different sensitivity levels. It stars with a purely optical image. Next the hydrogen is added at a very low sensitivity so that only the brightest gas (blue) is shown, which follows the stellar disc pretty nicely. Then we ramp up the sensitivity a bit and the fainter hydrogen is seen to extend quite a bit further, especially in the upper-right where a "warp" in the disc is visible. Finally we go to the best sensitivity level we can (any more and you'd just see noise), showing how the full extent of the hydrogen is much greater than what was previously known.

The sensitivity levels are chosen entirely arbitrarily just to show how processing the data in different ways can reveal very different features from the same data set.

And yes, it looks very nice indeed in 3D, but you're going to have to wait for that one.