the Opportunity landing site is near the crest of a low dome, not on low-land at all. "Etched terrain" to the north, south and west are in lowlands, and are peppered with mesas and layered exposures of sulfate/salts with modest hematite inclusions, but nothing like the nearly complete hematite "gravel" cover we see around the Opportunity rover. In some ways, the lowlands resemble terrestrial playa,

but the analogue isn't quite correct because the mechanism of formation on Mars is not well understood. Same with the sabkha model.

The lowlands seem very heavily eroded, whereas the highlands do not. As pointed out in another thread, the hematite "gravel" cover on the dome surface where the Opportunity craft is chugging along could have resulted from the erosion of a meter or so of hematite containing sulfate/salts. Not knowing the erosion rates on Mars, it's hard to state how long the surface has been weathering. On the other hand, layered mesas in the etched terrains are dozens or even hundreds of meters high, and they appear to be the product of erosion -- surrounding terrain has been stripped away somehow. So the heavy erosion in the lowlands could be the result of fluid action, wheras the lighter erosion in the uplands could be the result of atmospheric action.

But one of the first findings was that the deposits themselves (upland and lowland) were most likely made in water (or sulfuric acid solution or some such). Was this whole area a sea of some sort? That seems unlikely though not impossible. Could the deposits have been made under an ice cover?

This line of reasoning leads to the notion that the deposits in the lowlands and their erosion were the product of fluid action. The deposits in the uplands have not been eroded to any significant degree except where openings occur (craters and such) that expose the underlying sulfate/salt deposits to the atmosphere. Then erosion may occur fairly quickly, but not nearly as fast as erosion in the lowlands.

Hmmm....