Limburgish
Are you sure that the distinction between those is umlaut? They had separate vowels across Middle Dutch, as indicated at vliegen. Can the Limburgish distinction be a result of an older ô ~ ō alternation?
For that I'd like to refer you to a-mutation and middle dutch: Retention of /u/ (did not merge with /o/) and /uː/ (remained as a back vowel). I believe ôo in general turns op as oea/oa/oeë/oe (or ou) instead of oo. As in doead (dead) vs. bloòd (blood) vs. kaore (corn) vs. boum (tree) vs. noót (nut). Strong verbs are an exception to this general change and become oo due to its plural having oo (*flaug > *vloôg > vloog, *flug- > vloge). Middle Dutch had a paradigm in which oô (singular) alternated with ō (plural/particle). In Limburgish the past tense oo alternates with the particle ao, which is the same as the pattern in Old English (flugon vs. flogen), with the exceptions in the singular.
So the oo in the Limburgish past tense is reflex of a preserved Old Dutch u that underwent lengthening, and this gave a different result than lengthened Old Dutch o? Does Limburgish preserve distinct original u anywhere else?
Limburgish preserves /ɒ ~ ɔ/ - /ʊ/ split. In Limburgish this can be seen as o and ó. Compare: góld (gold, no umlaut) - hout (wood, with umlaut and diphtongization), honing (honey, no umlaut) - kaore (corn, with umlaut and lenghthening). Limburgish pretty much follows the rules as described: here. --Ooswesthoesbes (talk) 10:22, 11 November 2017 (UTC)