So here's the deal: working on a short film, I compressed (track compressor) some of the audio with a 2:1 compression, but not all. That which I didn't compress was shot outdoors, and I decided to leave the compression off 'cause it seemed to bring up the wind noise/ambient noise. It's EQ'd, and I rode the audio throughout the movie with volume envelopes. However, watching the movie on regular TVs, I notice it becomes distinctly less punchy, and to my ear, harder to discern dialogue, once you go from the compressed portions to the uncompressed.

I did the compression at a track level, so all I have to do to compress the now uncompressed audio is add the track compressor plugin and set it to 2:1. However, this changes the volume levels. I'd like the peak levels to stay where I put 'em, and just have the low stuff brought up. Is there a simple way to compensate for the change, or do I have to go back and redo the envelopes for that whole portion?

You could use expansion instead/after a compressor, but with lots of volume envelopes, it's guaranteed to sound fairly processed/false.
Were it me, I'd remove all the envelopes, set up each set of tracks to busses of related information, and put a static compressor on the bus, set to whatever you need it to be. Then individually use volume and/or compression on the track. Compression generally shouldn't be automated, although of course it can be depending on the compressor.
Then use small volume changes to build the intensity or lack of it, on the remaining sections.