Remember Carl, who wanted us to explain the "why" as well as the "what?"
Bell boxes, made of zinc, are quite often not resistant to corrosion where they are located. Both salt spray and direct burial seem to lead to their rapidly corroding away.
The lack of burial depth is not as simple an issue as it would first appear. Would you rather the assembly be beneath the water table?
Near the surface as it is, it is subject to both movement (shifting sand) and mechanical damage (tripping). Either can result in the wires (and connections) being damaged.
Where you have movement, it is important that your connections be made in such a way that any strain is not transmitted to the conductor connections. The use of "smurf tube" makes sure that the strain is absorbed only by the connections- just the opposite of what we want.
Lest I open a whole new can of worms, this MIGHT be a place for SO cord, strain relief fittings, and PVC (NEMA-4X) boxes. It all depends upon what the circuit is derving!
We often chide people for using plumbing components for electrical work- yet I am not aware of any listed bronze or stainless conduit. If I were to pipe this, I'd consider using SS plumbing pipe.
I don't like UF, for a number of reasons. I don't think UF is subjected to a salt-spray test; nor do I think that there are NEMA-4 strain relief fittings for any but the smallest sizes. I'd have to research this before I rule UF out.
If that box is there only as a splice, I would suggest that a resin/epoxy spliceing kit would be preferable to a box. A buried box is not accessible, in any event.