Yes, satellites are a problem. However they transmit in narrow frequency windows so you deal with this by simply notching out those frequency channels. You do need to make sure that there isn't so much power in the intereference that it does nasty things like send amplifiers into compression.

(source: I am a radio astronomer)

An educated guess, but I suppose LEO is far enough for them to be able to filter out the signal in software.

Kind of would block out a whole range of frequencies though, right? Dozens of transiting satellites, transmitting at a fairly constant volume, but of random data?

Not necessarily. From what I heard, with an array antenna you can actually separate the scalar and vector (directional) component of incoming signals and ignore those coming from particular direction. So you could either track (in software) and ignore signals from particular points (satellite), or pretty much configure your array to be directional.

That's fascinating. But then also makes you wonder why they have gone to such extents to extinguish any radio signals in the vicinity of the project

If I were to guess, here are several reasons:

- radio signal power goes down with square of the distance, so closer sources are much stronger than far ones

- any (be it software or hardware) filtering is not perfect, so a strong enough signal will probably still drown the weak ones

- it may be tad more difficult to filter out a close source because you can't assume it radiates in parallel lines, which to some extent you can with far ones (like for most practical purposes here on Earth we assume that sun rays are parallel)

- strong signals close to the ground means lots of reflections, which means more strong signals coming from random directions

- it's probably cheaper to make an exclusion zone in the middle of a desert rather than shielding yourself and upgrading the tech even more to filter out all the surrounding RF noise