The closer we get, the more this area looks to me like something similar to glaciation may have carved up the Columbia Hills and other similar landforms throughout Gusev.
My thinking here follows this general timeline:
1) Near the end of the LHB, Gusev is formed by impact into an already impact-shattered and brecciated megaregolithic crust.
2) Tectonic stress during the final cooling and "setting" of Mars' crust (the end of any major crustal plate movement) uplifted a string of hills through the center of Gusev. The Columbia Hills are one stretch of a general line of hills that runs off-center through Gusev and extends beyond it, and thus doesn't appear to be related to impact formations. Ergo, tectonic uplift is implied.
3) Before, after or during this tectonic uplift, Gusev filled with water and became a lake. This may have been a relatively long-lived phenomenon, or a short-lived one. Or even a cyclic one. I'm thinking this happened after rather than before the hill uplift, because while some of the rocks of the hills seem to have been fairly strongly aqueously altered, many more seem to have seen very little liquid water in their histories.
4) As Mars cooled further, the liquid water all froze and glaciers formed. Glacial movement deflated much of the original floor of Gusev, including a lot of the lacustrine materials, piling them up in some places and denuding them from others.
5) The volcanic deposition originating from the Tharsis Bulge finally reached the Gusev region, filling over all but the tallest features from the original Gusev floor in a lake of basalt -- the last lake Gusev would ever hold.
6) Therefore, the currently exposed surfaces of the hills and the inter-hill basins are the result of, first, major glacial deflation, and second, long, slow aeolian deflation and deposition, of the Gusev flooring material that was uplifted into the original hills so long ago. It's the last exposed remnant of the pre-lava-flooding Gusev floor. And while that is exactly what the MER Team was hoping for, the long and eventful history of the site jumbles everything and makes it nearly impossible to gain a clean context for most of the exposed rock beds.
One reason it's all so jumbled is that the original Gusev floor was impact melt and breccia, which is jumbled to begin with. That floor then collected layers of volcanoclastic and lacustrine materials (all the while being peppered with impact craters of all sizes), and then was tectonically uplifted, which is a jumbling process in and of itself. Then when the Gusev floor was heavily glaciated, rocks and soil got transported glacially all over the place. Yet more jumble.
I'm beginning to think that it's going to be almost completely impossible to find any geologically non-jumbled sites on any bodies that still exhibit major scars from the LHB...
-the other Doug
I for one can't see anything in Gusev that looks like glacial landforms. Also please note that glacial deflation requires an active wet-based glacier that moves over the substrate. This requires either melting in the marginal zone or gravity-induced movement (i e downslope). A cold-based glacier, that is one that is frozen to the substrate, leaves little evidence of its presence.