Seems counter-intuitive and definitely warrants research to understand the mechanism. The most unintuitive part being local plants not returning.

I feel like this has been studied before, and lots of times: nature is uncontrolled. Which really leads to constant (like every 10 years or so) disasters of one kind or another (usually an invasive species), destroying plant growth and biodiversity.

Humans built systems to prevent all this from happening, 1000s of little custom ones in 1000s of different little places with 1000s of different functions and kept those systems running. That's why there is the continuous nature across Europe. It's entirely artificial and it's due to farms being so incredibly widespread.

The "natural" Europe is really mostly steppe dotted with swamps and forests, usually next to each other, with vast open spaces, with very few animals, in between. That should be the situation one would expect to return.

I suspect this is the opposite in areas of the USA that have seen population declines. There is virtually nowhere in the USA in the 2020s that looks like, say, Ogimachi Village. And even in, say, the 1920s, so much forest had been cut for agriculture, logging, mining, and soon thereafter for the first suburbs, that I can't imagine human development had any positive effects whatsoever on wildlife or biodiversity.

Isn't this because humans killed and took the role of large grazing animals and large predators? After some time this can result in a new equilibrium, and when humans move out it's chaos again for a while. Question is not if nature will deal with it, but rather what we humans want — probably not some ecological desert for millenia.

If something is dependent on human intervention for survival in a specific environment, was it ever even wild?

I'm skeptical. Having been to some areas of rural Japan, I'd suggest that the macro cycle established by humans was irrigation related and the removal of the same causes deep challenges to former human-presence-biome members. A novel and restored biome, more suited to the natural hydrosphere of the area, will likely take decades to a century to re-emerge assuming its species survive and its preconditions haven't been ruined by concrete.

Note also in my limited experience there is typically a severe and sudden difference between flat land and mountains, the former being human managed with extreme prejudice and the latter being typically degraded/unnatural/reforested pine forest. This is one of the most striking features of the Japanese landscape relative to other nearby countries. I would assume the current study focused only on flatland, as the forests are not going to empty out when proximal humans desist: quite the opposite.

In your face depopulators!

Maybe they weren't wildlife but roomates just like the pigeons and the rats.

I can't help wondering if their consideration of climate change as a covariate factor has been rigorous enough.

Climate change is happening everywhere but it is quite noticeable in Japan with summers getting hotter and winters getting milder.

I wonder if this is somewhat specific to Japan or has been observed elsewhere

Give it 20 more years, it takes time for ecosystems to rebuild.