One of the indicators of Snowball Earth is the presence of dropstones in places that were then equatorial. I’ve got two questions about that:
How do they know that the dropstones were dropped by ice glaciers and not salt glaciers? Or glaciers of some other substance?
How do they know that the layers above where the dropstones were dropped were not eroded away by glaciers at a later date?
Dropstones are only found (or at least can only be readily identified) in fine-grained distinctively layered water-deposited sedimentary rocks, which are indicative of deposition in relatively still and somewhat deep water like a lake or a sheltered bay. So you know the large dropstones weren’t transported by normal sedimentary process. The way you can identify a dropstone is that the older layers below the rock are deformed and displaced from the rock hitting the still-soft sediments. Where the rock intrudes into the newer sediments, those sediments are draped over the rock in a somewhat similar but distinct manner.
There’s only a handful of phenomenon that would explain a large rock falling directly down in an otherwise calm-water environment, and most of them only work for certain types of rocks (i.e. volcanic dropstones are going to be volcanic rocks) and for stuff like recycled sedimentary or metaphorphic rocks the only plausible mode is iceberg drops. (Or I suppose biological rafts, but that’s not too likely since we’re talking about the Precambrian. It’s also not salt glaciers because even if they did reach a large body of water and calve some saltbergs, they’d dissolve quickly and also wouldn’t float.)
Note that dropstones aren’t quite the same thing as glacial erratics, which are rocks that are transported and deposited by glaciers and are usually identified simply by being an out-of-place rock type. The defining characteristic of dropstones is that they’re found still in situ where you can see evidence of them hitting the underlying sediments and the overlying sediments being deposited on top.