Earth is missing a huge part of its crust. Now we may know why

Ah yeah. That crater is about 30 km wide, which is pretty big. Most craters follow roughly a 1:10 scaling between depth and diameter [1], so we can approximate a 3 km initial depth for this new new Hiawatha Crater. This crater is quite young (12 ka to 3 Ma est.), but if it had existed 720 Myr ago, then our estimates suggest it would have been borderline -- probably erased unless it got lucky and landed in one of the regions with little or no Neoproterozoic glacial erosion

The number one condition for fast glacial erosion is what's called "wet-based ice": if pressure and temperature at the base of the ice sheet (warmed by geothermal heat flow) is near the melting point, then ice tends to slip against the rock and erode rapidly; if it's too cold, then the ice will flow internally instead and not erode very well. Coupled climate-icesheet models so far [2] suggest it's possible to get wet-based ice even in a "hard snowball" state with limited precipitation from sublimation alone (though not under all parts of the icesheet at once).

[1] https://agupubs.onlinelibrary.wiley.com/doi/pdf/10.1029/JB07... [2] https://www.sciencedirect.com/science/article/pii/S0012821X0...

Good questions!

1) Perhaps counter-intuitively, it's generally thought to be easier to enter a Snowball state when most of the continents are at the equator, the reasons being (A) silicate weathering of the continents consumes CO2 and is faster at low latitudes and (B) the albedo (reflectivity) contrast between ice and water is greater than the albedo contrast between ice and land. Ice is very reflective, water is not, and land is in between. In other words, if you grow sea ice at the poles, you make Earth more reflective, and the change in light (and heat) reflected per unit area of ice cover is greater for ice grown over oceans than ice grown over land. There is some background on the history of the Snowball Earth hypothesis, including the above points about albedo, here: https://en.wikipedia.org/wiki/Snowball_Earth

2) Definitely. I don't know the right answer yet, but here's a relatively recent (and relatively accessible) Scientific American article on the topic: https://www.scientificamerican.com/article/a-new-idea-on-how...