Dunno...
While this might be a fun thing to do I am not sure if it is waste of precious development time for the moment.
Every single machine that I have hanging around, going as far back as 2002 has built in sound environment that comes from the chipset of the motherboard. Those can be tailored to give whatever preset you want, whether shower tiles, concert hall, or mountain range, etc. I use the last option which gives you an unflat ambient sound and a little bit of echo/reverb, even a little pink noise hiss.
If anything I would like to see changes to how stationary sound sources are handled, expecially consists that have two or more engines of the same type and hence the same sound sources. Both MSTS and OR suffer from
http://en.wikipedia.org/wiki/Flanging and
http://en.wikipedia.org/wiki/Phasing which requires the user to de-tune the frequency of engine sounds that use the same .wav files by 250 Hz so that your speakers and ears are not faced with a horrible "beating" frequencies. This means specially tuned eng/sms files for EACH engine. Some payware vendors already do this.
Adding a bit of pre-Doppler for stationary sounds sources might help, since the reason for Doppler effect is the shifting of frequencies when a sound source either moves toward or way from the listener (compression and expansion of soundwaves). In practice how this would work would be to have any sound source that is farther away from the listener have a frequency shift applied even if the source is not moving. Goodbye flanging/phasing!
Something to think about.....