are we meant for a savanna landscape?

An article in this week's Nature, by Cerling et al., discusses the use of chemical isotopes in soil to to determine a 'paleo-shade' measure for ancient environments. They suggest that the early hominin sites were primarily in savanna environments (defined as having less than 40% tree cover). This has some interesting implications for landscape use among early hominins.

I can't get access to the actual article, so I'm relying on this NSF summary. The summary raises two questions that I hope will be answered in the article:

1) Are there changes through time in the landscapes where we find hominin remains? Other indicators of local environments (microfauna, paleobotanical remains, etc.) suggest that many of the earliest fossil sites were in forested environments. Does this analysis suggest otherwise? What accounts for the discrepancy?

2) Is the concentration of hominin remains in savanna environments a reflection of landscape use, or do we have a taphonomic problem? Tropical forest environments are among the least likely environments to produce fossils, due to the rapid decay of organic material. How do we know hominins weren't using the forest, but in death only found in less humid environments? Will future alien archaeologists say of us "They lived in open or lightly wooded meadows, in long rows of boxes, buried six feet deep."?