Given the landing inaccuracy of either craft, doing highly detailed analysis to find a specific safe spot is pointless -- the "spot" for either MER or Phoenix covers hundreds of square kms. Existing orbital photography should be good enough to identify areas that are mostly flat and not too rocky.

As Doug points out, given the capabilities and goals of Phoenix, what it needs is a spot with near-surface ice, and from what I understand there are huge areas of the northern plains that look pretty similar, both in topography and hydrogen abundance. MARSIS data will no doubt add to that picture, but those results should be available and analyzed in plenty of time for a 2007 launch.

I bet an MER clone would have more to gain from waiting for data from MRO than Phoenix. There are several other places one might go look for sedimentary rocks, and higher res images with better multi-spectral data might help you choose which pile to look at. Likewise, higher rez might help determine which lava flow looks youngest, whether some ambiguous areas are lava flows or mudflows, etc., depending on what it is you want to explore.

Nick's also right about the state of the missions. If we had another ready-to-fly MER as well as the already built Phoenix, there might be an interesting question about which one merited the use of a rocket in 2007, but that's not the situation we face.