Sunset in Tucson at Feb 9 was at 18:04 [0]. The incident occurred at 22:30, so it was pretty dark. I could not find any weather reports for that day, but since it disappeared in the undercast it must have been cloudy. Because it had been dark for some time for some time, the air must have cooled of, and it is not unreasonable to assume that there were also some smaller clouds below 14,000 feet.
It reportedly had a green flashing light, and because of the darkness it seems unlikely that they could could really see other details. The object could not be observed with night vision goggles which seems surprising since they are supposed to be more sensitive than the human eye.
When you look at a bright colored light source you will experience afterimages [1]. When you are looking at a point source you will see these as tracks (with a complementary color) around the source. These tracks are caused by your eye movements. Under low-light circumstances your vision mainly depends on rods[2]. A green light with a wave length of 523 nm is pretty close to the optimal rod sensitivity that lies at 498 nm so your rods are pretty sensitive for this wavelength.
Because the outputs of rods are interpreted as light vs dark, the afterimages of the rods will be interpreted as dark.
> While this makes rods more sensitive to smaller amounts of light,
> it also means that their ability to sense temporal changes, such as quickly changing images, is less accurate than that of cones.[2]
This also implies that afterimages from rods last longer than those of cones. Afterimages follow the movement of your eyes. So if you focus on a new point, the afterimage will follow to the new point, so it can appear to move at a tremendous speed.
In principle rods can detect a single photon[2]. Night vision goggles generally have optimal sensitivity in the near infra-red spectrum [3] so it is not very surprising that they failed to detect green light.
The central part of our eyes contains few rods [2]. The effect of this is that when you try to focus on a faint light source it will seem to disappear. This is why you should always try to focus on a point close to a star but not on the star itself. If you forget to do this, you will observe that the light seems to flicker even when it is not.
Humans are not very good at estimating the size of objects in the sky. The reason is pretty simple, if you do not know the distance between you and the object you cannot say anything about its size. In the air it is difficult to estimate the distances because there are no real points of reference. Also remember that it was dark.
It's pretty simple to buy a green laser pointer [4]. If you point it at some cloud it will appear as a green object on that cloud. Due to afterimages, it will appear to be a dark object with a green light on it. When you point at a different cloud, it will look like the object moved from its first position to its second position. If you believe that both positions are far apart then you will also believe that the object moved at a very high speed. It is also very simple to make the object disappear by turning of your pointer.
This seems to be a much simpler explanation than speculating about extremely powerful drones (or aliens).
[0] https://www.timeanddate.com/sun/usa/tucson?month=2
[1] https://en.wikipedia.org/wiki/Afterimage
[2] https://en.wikipedia.org/wiki/Rod_cell
[3] https://en.wikipedia.org/wiki/Infrared#Night_vision
[4] please use your favorite search engine or supplier
P.S. I am a bit disappointed in most of the HN responses that assumed it was some kind of physical object without considering other possible causes.