Based on comments, would the moderators please move this to Astronomy.SE?
In the novel, some agency disrupts the moon. The moon is left in 7 major chunks that are still gravitationally bound. (Reference that a week later, it's still just a cluster of large rocks.)
As the rocks interact they divide into smaller rocks.
About two years later the world ends in a "White Sky" Debris from the moon is hitting the atmosphere at a rate that the upper atmosphere is glowing white hot. Earth is cooked.
My gut level reaction (my physics intuition) insists that if the impact resulted in a mutually gravitationally bound cluster, it would relatively quickly reform into 'scrambled moon'.
I have a few problems with SevenEves, but don't know how to attack the problem in terms of its physics.
Problem 1: Rock isn't very strong compared to it's mass. If the fragments are hitting each other, they are going to be pounded into much smaller rocks in a hurry. This however takes energy which will come from the relative speed of the pieces. This favours reassembly into a single moon.
Problem 2: By the equi-partition idea eventually all the rocks have roughly the same energy -- much like a mix of oxygen and hydrogen gas, the hydrogen has an average larger speed. Right? I recall this being a process that cleans dwarf stars out of globular clusters. This acts to eject smaller rocks from the cloud, reducing the relative energy of remaining rocks to their own center of mass. In effect cooling the cluster. This favours reassembly into a moon.
It's not clear to me how to calculate how many interactions it takes to reach some sort of equilibrium distribution of speeds, nor from that how to calculate the rate of debris hitting the Earth's atmosphere.
Problem 3: The moon is beyond the earth's Hill sphere. Half of the ejected chunks are lost for very long periods of time. Of the chunks with velocities that take them into the Hill sphere, they are coming in with the sum of their ejection velocity from the cluster plus the orbital velocity of the cluster, plus the escape velocity of their apogee. Most paths will be hyperbolic. This means that a large number of them have ONE chance at hitting earth before being flung out into space. (Some fraction that are ejected in the direction opposite to the cluster's orbital motion will be in an elliptical orbit with multiple chance to hit earth, although if they miss the first time, they don't really get another chance until perturbed with another encounter with the cluster.)
Problem 4: The earth will have an effective cross sectional area larger than it's diameter due to gravitation. This will depend on the velocity of the incoming chunks. By analogy to the nucleus having different cross sections depending on the speed of the incoming neutron. This favours a white sky.
Problem 5: If clusters eject significant mass this way, why do the Jovian trojans still have a large distribution of sizes. There are also 'Collisional Families" groups of asteroids that appear to be from a single parent body, with similar orbital elements. This favours clusters of rocks being at least meta-stable.
I feel that I'm missing some major concept here.
Is my intuition correct? Are the problems I see valid objections?