Aurora - a critique by Stephen Baxter and others

[Rynn]

a) I would suggest that rather than trying to fit a full planetary ecosystem into a ship, a better approach might be to create a smaller, custom designed, and possibly engineered, ecosystem instead.

I can't imagine the size of the ship necessary to take an entire planet's full of ecosystem. That's millions of species of just animals alone, let alone any other kingdom. You'd have to take a breeding population of all of them at least and even more of species that support others by providing food, shelter, pollination etc.

In terms of transplanting an ecosystem it would definitely have to hover above a minimal ecosystem. Minimal as in "this is the bare minimum to keep everything alive and sustainable" but with a bit extra to add a buffer in case of accident. Then on the other side you breed numbers up and use your exowomb technology to introduce new species.

Genetic engineering and bots could shrink that number down as we've suggested if you can design things like robotic pollinators, soil species that fill multiple niches etc.

One element of such a shipsystem might be that it 'tops out' with its largest organisms and/or apex predators actually being rather small and harmless by our standards. Cats or weasels instead of wolves or lions (or tigers or bears, oh my). Another might be lifeforms engineered to work better in the closed environment of the ship. Perhaps a rabbit gengineered to have a slower breeding rate, for example.

I assume here your idea is that humans don't draw resources from this ecosystem? If the apex creatures are cats then they likely couldn't, humans requiring a lot more resources. But I suppose that if the journey was short enough you may be able to get away with stores of food, recycling technology and life support all to keep the humans alive until they can arrive and expand the ecosystem.

For a generation ship or anything like that I suspect you wouldn't be able to do this. You'll need to use the ecosystem for everything from recycling of waste into food to providing breathable air to microbiomics.

b) I would also suggest that for a fairly short range/short duration ship such as Aurora (I've seen discussions of ships taking centuries to thousands of years to get around and SF back in the 70s and 80s had many stories about such), that a better approach would be to go for less of an ecosystem and more of a system designed to support humans reliably over a fairly long time scale. So perhaps something more like farmland, living areas, some amount of plants and animals, but not trying to cram whole self-sustaining ecosystems into such a small space. A certain amount of machinery doing some jobs as well might be in order

Later in the setting technology could do the job of an ecosystem. An artificial ecosystem could replace the air, recycle waste and advanced medical technology might be able to provide all the symbiotic organisms humans need to properly survive. That's a deceptively monumental challenge though replacing the biological environment humans have spent their entire lineage evolving to fit within. Vecs would definitely have an advantage in colonisation.

Which sets up a second strategy quite nicely: building an ecosystem from scratch. As difficult as it would be to transplant an ecosystem this is orders of magnitude more. You have to not only synthesise millions of organisms representing thousands of species but somehow apply them to an environment in the right order (likely artificially preparing that environment first) so that they thrive rather than die out in a huge wave. A lot of scope for failure there. Hmm....ideas for an entry about an early colonisation effort are percolating

Along those lines do we have any examples in the setting of robotic interstellar ships, lead by an AI, that build an ecosystem and raise human children in it? I'm feeling some good scope for OA level weirdness there.

A few early missions used the Parental Vec system.
Uoagranyu was mostly a success;
Pandya nearly failed because of inadequate immunology;
Diwali here the environment was deemed too dangerous for biological humans so it became a purely vec colony;
On Caph the system worked too well, and a population explosion occurred.

Ah thanks for that I have the outline of an idea along these lines. An early mission that consisted of a few AI with exowomb technology that were intended to build an ecosystem and raise humans in it. Only their eco-engineering theories were flawed and the ecosystems keep dying out. Over time this drives the AI mad (because they were designed to cherish life) and they get stuck in a cycle. Several centuries later a probe arrives to find a system littered with dead habs and a mad AI that keeps building new ones, raising human children in them and watching them die of habitat loss over and over.