This is actually how archeologists estimate their ranges for human habitation in an area, just with math instead of guesses. This paper isn't about that (try timing papers like [1] instead), it's about the actual radioisotope boundary dates for an actual site.
Part of the problem these LGM dates keep running into is that it's not obvious how people got here in the first place. After about 48kya, the ice free corridors close up and the ice sheets encompass most of the coastal islands as well until about 22kya when humans can get to Alaska again and 18kya when humans can easily get down the coastal routes. What would have been possible in the middle is very much still in the ??? grey area.
I assume they built boats. It isn't exactly surprising after 20k years that we can't find any proof of these boats, but how else would then have come across? Fly?
We know that humans had boats by that point due to their presence in Australia and the current dominance of the coastal migration hypothesis. The problem is that our current understanding of the climate is that the coast was too ice-locked by glaciers for coastal foraging. If there's something this old, either they were doing some very impressive and unexpectedly long distance nautical journeys, or there are gaps in the details of our paleoclimate models.
It's not that either is impossible or even improbable, it's just that it forces us to revisit everything again to try and work out the routes if these (and other similarly early dates that have been proposed in the last couple years) hold up under review.
Coastal foraging? Why can't they just fish? There are lots of fish in those northern reaches today, I assume there would have been even more back then. And fish doesn't necessarily have to be cooked, so no requirement to stop to make fires.
Note that the phrase "coastal foraging" typically implies fishing and the exploitation of other littoral/marine resources.
Anyway, the current understanding of the paleoclimate is that entire region between approximately Valdez and Vancouver Island was entirely covered by glaciers out to the edge of continental shelf until ~18kya. Lesnek et al has some good diagrams [1]. Living exclusively off deep-sea marine resources in an iceberge minefield without fire for over a thousand miles in one of the coldest, most dangerous oceans in the world without landing suggests an unprecedented level of both nautical technology and experience. Where did that come from? We have no good answers right now.
What is the reasoning for thinking they came from the top? You’ve said they had already reached Australia via boat. Couldn’t they have reached some other part of the Americas by boat, maybe South America via Africa or Easter island?
Mainly because both Americas where inhabited way before the pacific islands. For example Easter Island is estimated to have been inhabited around 300 to 1200 CE. This is also confirmed by genetics, with an origin of the Amerindians located around central Siberia.
We've definitively established that Clovis populations and all modern Native Americans are genetically related to Beringian populations out of Eurasia. It's reasonable to suspect pre-clovis populations probably came from much the same area, especially given that suggested pre-clovis lithics so closely resemble clovis traditions.
Additionally, the farthest east that we've found evidence of Australasian populations is the Solomon islands, some 7,000+ mi from South America. There's no evidence of habitation on the intervening islands until Austronesian peoples show up much later. As for Africa, we simply have no evidence for it whatsoever.
The authors' speculation that this was due to both populations originating in east Asia eventually turned out to be true once further work was done on sequencing Tianyuan man [1]. It's not indicative of Pleistocene trans-pacific contacts.
One point in favor of a water route is you can carry a lot more stuff that way. Most of the migrations into the Americas were before animals other than perhaps dogs were domesticated and before the wheel was invented.
So anything you wanted to bring with you on a land route had to be carried by humans, dogs, or pulled on sleds by humans or dogs.
On a water route you could tow another boat or raft behind full of your stuff. (Rope was invented long before any migration to the Americas).
But now they are saying that people were here before the land bridge was usable or existed, right? So, maybe they came some way other than the land bridge. Or, earlier peoples came via some other route (hence the current footprints) and then a later migration came via the land bridge.
And possibly more raw intelligence as we have mainly been selecting for "smarter than grass" since beginning this experiment in agriculture 10k - 20k years ago.
Part of the problem these LGM dates keep running into is that it's not obvious how people got here in the first place. After about 48kya, the ice free corridors close up and the ice sheets encompass most of the coastal islands as well until about 22kya when humans can get to Alaska again and 18kya when humans can easily get down the coastal routes. What would have been possible in the middle is very much still in the ??? grey area.
[1] https://doi.org/10.1038/s41586-020-2491-6