I'd be surprised if any 'secret ecosystem projects' don't suffer the same fate as the Biospheres. Namely, catastrophic systemic collapse within a relativity short timescale brought on by an unmitigated pathogen.

The main reason those projects failed was quite simple: they were too small. Make it large enough, limit occupancy and stand back and nature will figure out a way to make it work, it doesn't need engineering, it needs scale and patience. After all, a planet is exactly that: a closed cycle biosphere that somehow balances itself if we don't mess too much with it.

I agree that those past projects were too small, and from what I understand they were really pedantic about what was coming in and out of the space. Rather than being so pure about it, I think it would be better to develop a system that was _mostly_ independent, then as we understand and get better at running it start closing it off.

I actually don't agree that a system of any size will just balance itself. I think maintaining a closed system that can maintain human life, as we know it now, indefinitely, would take a lot of careful balancing. It would need constant monitoring and intervention. Something as simple as an algal bloom could wipe out everything else living in your water. If your bees die of some disease, you lose pollination, and all your fruit.

Can you imagine how much tech and effort it would be to manage every bee in your colony to make sure they all don't just die one season. And what do you do if your bee population crashes, you have to have some kind of bee breeding program to get the numbers back up.

The earth doesn't somehow just balance itself, the earth could become one vast desert in just a few generations. Pockets of live here and there for thousands of years no doubt, but enough to sustain human life?

well, that's why we need to do the research and develop tech to help us measure, manage, and balance an ecosystem.

I can't disagree with that, but the problem with the notion that there are secret biosphere projects is that it's hard to keep secrets. The more people know, the easier it is for knowledge to leak. Conversely, research is hard, and the number of researchers required grows with the complexity of the research. Biospheres are incredibly complex, ergo private secret biosphere projects are almost certainly doomed to fail.