Iron Islands are suggested to have a thousand ships. While it is true that, as pointed out by Blackfish, “Even the great bureaucratic machines of Rome couldn’t afford to keep more than 1% of their population mobilized”, reason for this lies in the nature of Roman military at the time. It was a fully professional standing army – men who had no job other than fighting. Much later, after Islamic invasions of 7th century, Roman military restructured itself on a semi-professional thematic model: each soldier was also a landowner, and used these lands to support himself. With this model, 11th century Roman military managed to field a total of 250 000 soldiers of ground force and at least additional 42 000 of the navy (this number is for 899). With population of 12 million, this comes to 2,4% of total populace. China during Tang dynasty managed to field army that was 1/65 of populace (or 1,5%). In 840, Roman army fielded 120 000 soldiers out of populace of 8 million as well as ~40 000 men in navy, for a total of 150 000 – 160 000 men, or 1,9 – 2% of populace. In the West, Alfred of Wessex could sustain 5 000 mounted troops out of 450 000 populace (1,1%).
However, Iron Islands are based neither on ancient Rome nor on its Medieval iteration. They are, if anything, Vikings of Westeros. Now, in AD 1000, Norway had 200 000 total populace, Sweden 500 000 and Denmark 400 000. However, only Norwegians and Danes raided British isles (600 000 total). Now, raids could be anywhere from 3 to 30 ships; largest raid – Canterburry and London in in 850s – involved 350 ships, which would mean an army of 10 000 at least (at 30 – 40 per ship, it would come out to 10 000 – 14 000 men). This means mobilization rate of 1,7% – 2,3%; but it is unlikely both Norwegian and Danish Vikings were in the raid, so number could be as high as 7%. Paris raid had 300 boars carrying 6 000 – 8 000 men (20 – 25 per boat); so 350-ship-raid would still mean at least 7 000 men, or 1,2% of populace.
Of course, short-term defensive mobilization of poorly-trained militia can and did reach 10% of total populace, but that is not what we are looking at here. So using numbers suggested in the post, that is 30 000 seamen in total, we are looking at total population of anywhere between 430 000 and 1 500 000. Fact still remains that Iron Islands almost certainly do not have that kind of population – in fact, a number fifth to tenth of that is more likely. If wall is 300 miles in length, total area of Iron Islands cannot be more than 10 000 square miles. Assuming population density of 4 people per square mile – a likely number, considering Iron Islands’ historical model and apparently not very good living conditions – total population would be 40 000 people. Even high-end population density of 32 people per square mile (~12 per square kilometer), only historically achieved in Mediterranean societies, still results in total population of 320 000 people. This means that – using high mobilization rate of 7% and low crew of 20 per boat – absolutely largest number of warships that Iron Islands can mobilize comes out to 1 120. However, actually reasonable number is 2 800 men and 140 ships.
Links for details: