Well the 100-150 kings estimate is generally based on the number of Trícha Cét which a later construction. If you see a correspondence between trícha cét and tribal kingdom, then there is no reason why we cannot use it for population estimates. Perhaps 500,000 is far to much, but 9,000 isn't worth being called a minimum. I have never seen anything that suggest certain kings were only a cenn fine. Look at the size of the trícha cét as Paul McCotter has them, over 170 in total. It would be impossible to imagine any of them contained a mere 60 individuals.
I'm also remembering that population explosion that occured in the 5th and 6th centuries as a result of the importation of superior agricultural practises, not all of which occured post-Christianity.
I would be fine with that, perhaps more as I feel the population was rising, but 9,000 took me aback.
Ah yes, the 9,000 was more of a conversation in my head of how the O'Neill Y DNA would have needed to overwhelm a minimum of 9,000 people, and if I couldn't imagine the Y DNA overcoming that few, there was no point on my brain continuing further down, Because the DNA more or less says that he produced 21% of the males (who had children) in Northern Ireland throughout Irish History.
Still Id say that your right that much of the population explosion happened prior to Patrick. Id say it actually started around 280 AD and started reaching critical mass around 320 AD when Irish Raiders started appearing in Roman Britain.
Still I think that the population explosion was tempered by migration to Britain. The Early Welsh Kingdoms were very heavily Linked to Ireland so I think that a large amount of "Irish" in c. 350 AD were migrating over to Wales (and maybe Amorica).
Overall I think of Ireland in c. 400 AD as a Celtic type of Greece. The whole Irish Sea was similar to the Aegean Sea in Ancient Greek Times. Many of the "Northern Irish Clans" That existed in that time period swapped shores to Scotland/Pictland, and I believe that many in general swapped to Wales.
I guess theres also a question of how one counts that overflow, and at what "snapshot in time" the picture is taken.