> If there's even a small chance of success in establishing a self-sustaining colony [...] then any degree of personal risks levied on individual colonists are "worth it" from the species level

No, the risk v reward curve has to be better than an equivalent risk v reward curve for a terrestrial species survival project to justify the mission.

I would argue that just about any PhD in epidemiology, disaster relief, or geopolitics will have a risk/reward curve far better than any Mars mission.

People keep making this argument that, absent any other information, a two planet species is more robust than a one planet species. But that's not the choice we're facing. The choice is a two-planet species versus a more prepared one-planet species.