I didn't mean to suggest that the crater was oval. Rather, I arm-wave that the smaller anomaly is an ice dome that is localized not on the central peak structure, but is developed at another weakness within the crater, say at the intersection of a ring fracture and a radial fracture.

I'm picturing a thicker crust, where only the bigger impacts result in penetrating this ice mantle.

Or, it is only the bigger impacts that fracture the overburden to the point where the ductal ice mantle begins to move to the surface as a diapir.

Maybe the ice is not a mantle, but a layer, and a discontinuous one at that? There does seem to be a tendency for higher albedo anomalies at the upper pole of the sphere.

Observational bias, I suppose. I've been interested in Upheaval Dome (Wikipedia) ever since I stood on it's rim back in the mid 80's. The debate has long been salt dome verses impact. I like both models. An impact occurs in a column of sediments that include a highly ductile layer at depth. The impact doesn't penetrate that layer, but does fracture the overburden to allow the deep ductile layer to move upwards.

I also picture one of those geodesic domes from Silent Running (1972).

(links added post-post)