There are two, possibly inter-related, issues here. One is the preposition expected with the root mlk in the hifil; the other is the relationship between the prepositions ʾel and ʿal.
1. MLK + ??
Typically the verb mlk takes the preposition ʿal, "rule over", and in the Hifil it appears so on at least six occasions (1 Sam. 12:1; 2 Ki. 8:20; 1 Chr. 28:4; 2 Chr. 1:9, 11; 21:8; n.b. this data remains a bit rough-and-ready).1 However, in 2 Sam 2:9, the verb appears (initially) with ʾel, "to" and this is (in a brief search) the only occurrence of this collocation.
Discussion of verb + preposition is found in Gesenius-Kautzsch-Cowley - § 119g for ʾel, and § 119aa for ʿal. This, too, suggests the expected preposition would be ʿal. Thus the Septuagint "flattens" them, giving ἐπί in each case = ʿal.2
2. Prepositions
This is most likely a case of the common mix-up between ʾel and ʿal which frequently happens, even in situations of near proximity like this one. See the BDB entry which explains the phenomenon and gives examples:
The preposition entries were prepared by S.R. Driver. He draws attention to the prepositions in this verse (without further comment) in his Notes on the Hebrew Text and Topography of the Books of Samuel (2nd edition; OUP, 1913), p. 242, adding some other examples where the same combination is found.
This line is extended (by way of explicit reference to Driver again) in the treatment of ʾel by P. Joüon & T. Muraoka, A Grammar of Biblical Hebrew (2006), pp. 456-7 = § 133b, which I take the liberty of quoting in full:
Note that the corresponding entry on ʿal (§ 133f) concludes with:
For the frequent confusion of על and אל, cf. § b.
And to bring comment on this phenomenon to its most recent statement, here is Ernst Jenni in his article on "Preposition: Biblical Hebrew" in the Encyclopedia of Hebrew Language and Linguistics, gen. ed. G. Khan (Brill, 2013):
The occasional apparent interchange in the usage of אֶל ʾεl and עַל ʿal especially in Samuel, Kings, Jeremiah, and Ezekiel (BDB 41, 757), has been put down to phonological fusion by copyists and semantic overlap in transferred meanings; in all likelihood the influence of Aramaic (where אֶל ʾεl is on the decline) also plays a role (Kropat 1909:41–42; Brockelmann 1956:104; Joüon and Muraoka 2006:456).
Yes, he's reporting. But I doubt there's anyone on the planet who knows more about preopositions in classical Hebrew than Ernst Jenni.
Summary
It looks to me like one of the frequent cases of ʾel - ʿal exchange, and the difficult thing to explain is why ʾel is there in the MT. Joüon-Muraoka go a long way to explaining these cases, noting the influence of Aramaic and the graphic challenge posed by these forms.
2 Samuel 2:9 is far from an isolated case, however, and a "complete" explanation would need to consider all the other passages in which this pattern is also found. Given this linguistic scenario, I have a hard time seeing that geography, etc., has anything to do with it.
Notes
- Note that with the more common Qal, mlk (vb) takes ʿal 33×, but never takes ʾel.
- See also P.K. McCarter, II Samuel (Anchor Bible 9; Doubleday, 1984), pp. 71, 82 with the similar occurrence at 2 Sam 1:24, who reads these ʾel prepositions as ʿal.