Factors only ecoinvent states
The ecoinvent Centre characterises this substance where the JRC does not — most often because EF 3.1's flow list has no such compartment, and 3,990 of the 4,372 are in silvicultural soil, an unconfined aquifer or industrial soil (#431). Adopting the number is a real option and often the right one, but it is somebody else's number about a flow the method's own publisher did not characterise, so it is offered rather than taken. Where one substance is proposed the same number in every context it appears in, the row says so: that is what makes a ruling about a compartment possible rather than one about 4,372 flows.
proposed-factor
Nothing in this queue: the two implementations agree wherever both of them speak.