This article doesn’t take account of a very significant effect on a decaying orbit, due to the oblateness of the earth.
In Tiangong 1’s case, its orbit is tilted at about 43 degrees to the equator, and due to the equatorial bulge it comes closer to the atmosphere by about 10 km, twice each orbit as it crosses the equator.
I have seen very little mention of this, but the basic fact can be verified.
I have been using a tracking site, in the name of n2yo ( a radio call sign, ) which evidently DOES take this into account, in addition to the eccentricity of the orbit, which has only one high and one low point per orbit, of course.
It seems to me that this would be important in predicting the time of the final plunge, since I would think it would occur on one of these “dips” when the orbital nadir aligns with the equator, because a dip in surface altitude of an extra 10 km, say, could “snag” it.
Well, that’s all beyond my, ah, “resources”, so I’m contenting myself with obsessively watching the track.
It looks like the “end game” here. ( I’m just waiting for a few minutes to see what our new minimum altitude will be ... was about 172 km, so just waiting here ...
... well, hit 171.50, not so much lower but 160 is “death”, so we have to see how it goes, orbit by orbit.
There’s a lot of variation in altitude that is not accounted for by the models we see, so I wonder if the “variable atmosphere” isn’t something of an excuse for laziness, as I see it.
Anyway, we’re close.