"That's assuming you can distinguish between lunar and asteroidal materials lying on the surface with a spectroscopic instrument. I don't know if that's true or not."

Seems unlikely. And the higher the velocity the more difficult it would be.

Let's look at reference:
http://adsabs.harvard.edu/full/1980LPI....11..112B
[can't copy and paste this material]

It models an impactor with same mass [1.67 x 10^8 kg] at velocities starting at 2 km/sec to 25 km/sec.
At 2 km/sec the apparent crater volume is 5.39 x 10^6 cubic meters.

If we assume a cubic meter has mass of say 2000 kg, then the comparison of the mass of the crater ejecta to the mass of the projectile is:
5.39 x 10^6 times 2000 = 1.08 x 10^10
The ratio is 64.5 kg existing material to 1 kg impactor.

At 10 km/sec the apparent crater volume is 4.57 x 10^7 cubic maters- so almost 10 times more existing surface mass compared to the impactor mass.
At 25 km/sec the apparent crater volume is 1.29 x 10^8 cubic meters.

Of course with low velocity the impactor may more or less stay *somewhat* partially intact, higher velocity might have more mixing, and +25 km/sec you going to more vaporization of impactor and what it hits, and the resulting plume largely could reach escape velocity and what doesn't escape diffusing over a large area.

As a guess perhaps the part of crater in the middle which uplifts could have some residuals of a impactor