How do they come up with the estimated number of asteroids larger than 1 km? Is it based simply on sampling a volume of observed space and extrapolating? Or do they have some way of inferring based on gravitational effects and anomalies?

> Is it based simply on sampling a volume of observed space and extrapolating

Yes, that's pretty much it. It's not as bad as perhaps it sounds. The age of the solar system means that the asteroid belt is mostly stable, which means the distribution can be inferred.

The Kuiper belt and Oort cloud may throw some outliers in occasionally, though. We don't know enough about them yet.

Is 93% really an acceptable amount of the 'giant asteroids that could smash into Earth' to track?

Clearly, its a better number than, for example, 50%; and its great that NASA is doing any tracking; but this is an species level existential threat that's being talked about.

I'd really prefer to live in a world whereby a joint international effort (ideally; or just anyone) was tracking 99.9999% of such asteroids, as a matter of priority.

I think that the LHC, or the moon landings, were great projects - but shouldn't we do this, first?