To further expand the topic of "why are metropoli (so is this the correct plural in english? Good to know
) usually on coasts", I'll go a bit out of "gameplay" topic, so feel free to skip this wall of text while I make a little example:
Between AD 1650 and AD 1700, the world saw the birth of the very first, true metropolis: It was called Edo, the modern Tokyo. Around that age, Edo reached 1 million people, which was an incredible number if you consider that in the meanwhile London had around 500,000 and Paris 300,000.
London and Paris numbers aren't really mindblowing if you think at what you call a "metropolis" today, aren't 'em? And yet, they already weren't self sufficent for food and were importers, of course, but half a million people can still be rather easily fed without intensive farming on land, the surrounding farming cities can be enough. And we are at the doors of the industrial revolution already.
Those, and many more, were built on water because in ancient times, water WAS the best source of food: no fertilizers and no pesticides to discover to maximize production, water was "fertile land" all year long, not just half of it. However, as time goes by, and fishing towns become large cities, agriculture discovers new ways of increasing production, while fishing is really more or less the same now as it was centuries ago. Yeah, we now have ships able to fish in the arctic seas without fear of sinking, but the methods used are just quite like the same, and so neither their "productivity" has gone through the roof as agriculture - and population - did.
Another big point is transportation: cities as Paris and London, demographically speaking, could have had a much higher population back then, even with the higher mortality rate: fact is, without modern transportation vehicles and infrastructures there wasn't any efficient way to ship huge cargoes of foodstuff between longer distances, without fail and in big quantities: once they came to be, food surplus from other areas specialized - as the industrial revolution taught us - in farming let 'em grow in population accordingly, reaching today's skyrocketing numbers.
USA is a good example of transportation relevance: the biggest cities on the east coast were - of course - the first to be settled for a mere geographic reason. While the central area of the continent was being explored, and slowly settled and turned into productive land, they were the political and industrial hub of the nation, but still dependant on shipments from the old world, both of food and machinery. And, while those shipments were consistent, there weren't infrastructures to efficiently transport sizeable quantities of food much further inland.
Once North America was finally completely settled, they already were population heavy cities, and since they were cutting ties with Europe, as such the inland area was dedicated to be the source of their food, as fishing definitely wasn't a main source. Slavery itself was in no small part a "natural" historical consequence of the need of setting up a massive farming output for a populous area somewhat suddenly forced to be self sufficient in a rather underdeveloped land (indipendence or not, there was an ocean between the two continents, shippings could only fulfill up to a certain requirement, especially with the means we had back then). Naturally, to grow food in large quantities, you need a lot of land, not a lot of people, and as such those areas are much less populated.
This demographic map of the USA still today speaks for itself:
Now, how could Tokyo - the only seeming exception - reach that (still relatively) huge population back then? Well, just think at what was the common diet in japan since just a few decades ago (and, to an extent, still today): rice. Fish was for the elites in Edo, most people lived on eating mostly rice and only rice for centuries, much different from what happened in Europe where even if wheat was the main source of food for the common people, they still had a lot of other agricultural products. Why?
Edo didn't born naturally as most metropoli, it was built over a small town just to be the new political capital, as a way for the newborn leadership of the finally unified Japan after the Sengoku period to centralize the power, weakening any potential threat from feudal lords forcing 'em out of "their" land, and imposing on them a huge expenditure to transfer their courts to the new "forced" capital. As such, Edo grew exponentially in a very short period, and to sustain such population agriculture nation-wide was further heavily specialized - almost to the point of leaving no space for anything else - on a food product easy to grow in large quantities even on a
relatively limited amount of land. Literally half of Japan was dedicated as being the "food production line" for just one city!
If they could get by with fishing, they wouldn't have needed such extreme measures, wouldn't they?
Going back to game talking (phew!
), about food specifically, now, what's the problem with CIV at somewhat representing this reality? That it has self-sustaining cities, instead of a general, empire-wide pool for food as it has for strategic resources, gold, science and culture (gold, science and culture are produced locally just as food, but they apply on empire wide features). As such, the city producing more food will be the most populous one aswell, the opposite of what often happens irl, while the gold producing city might see most of its gold income spent on another one. There's no way to really reproduce that complex behaviour unless you completely change how the game works, so we have two choices :
- through indirect changes of the game's balance, we make it so coastal cities are always able to grow the same - if not more - than inland ones, so that once an empire is fully developed its demographics may look a little more like real world demographics. Two issues though in this approach imho: 1) it all happens magically, those cities don't grow more because of how you actually played the game, better representing reality and a "food chain mechanic", they just end up, coincidentally, looking a bit more like it, no gameplay enhancements at emulating history... so... why bother, what's to gain? and 2) de facto, you overpower - from vanilla - a component of the game for the sake of a "reality" which still the game is unable to represent clearly... with all the potential balance issues this might bring to someone's tastes.
- reason with a logic suited for a simplified, alternative version of reality where settlements are exclusively self sufficient. It's not real here, but it's real in this gameworld. If that's so, logic would dictate that yeah, those cities should be smaller, as water shouldn't be as good as farm lands. Good counter argument would be: "ok, but if that's not to be compared too strictly to reality for gameplay balance, why not just pretend water in this gameworld
IS as good as farms?". At which I would reply, yeah, I can take it, and be happy seeing my coastal capital able to skyrocket in population as it probably would in real even if it now happens only because the seas are more productive, but on the other hand I'll also see that scientific frontier outpost in Greenland do almost the same. And that, I can't swallow
Now, if sometimes "weaker" coastal cities were a
gameplay issue, then there would be no point in discussing this, it's a game,
gameplay has the priority over reality. But I don't see a gameplay issue with coastal cities: even if they are "weaker" (something I'm not really prone to say wholeheartedly as I find in my style of playing they usually are quite efficient for me), they don't hurt anything. At worst, you just know that your coastal city will be less productive than your inland one, so you'll base your production queues accordingly, as your opponents AIs do. And coastal cities already often make for some of the best wealth specialized cities, now that TPs aren't so imperative, which isn't "weaker" in my book, just different.
Rather, it seems to me that the reasons encouraging this possible "boost" to water is either for realism sake (on which I already far too much extensively explained why I think it would actually do the opposite) or for a reasoning I don't understand: "land produces so much, as such, for balance, water should too". Well, land produces so much, mountains don't. Why don't boost mountains? Even in reality, they are a source of income, just think how much money from tourism the Alps bring to the surrounding countries... but reasoning like this, it would be an endless boosting of each and every feature, and then back at further boosting the first ones because now they suddenly seem "unrealistically" underpowered compared to the new ones.
The boost to improvements and resources made by this mod was to balance a gameplay issue: underwhelming improvements brought many players to spam trading posts instead of more varied improvements, detracting from the game fun. It was needed. With boosted improvements, now water looks a bit weaker than before on comparison... but even if it is, the gameplay doesn't suffer from it, I believe.
Just my... well... two billions thousands cents, I guess
bonus was added. Placing this bonus on the building instead of the resource itself has the simultaneous effect of improving this building's usefulness, which has been frequently cited on strategy forums as somewhat underwhelming in cost-effectiveness. It also follows precedent from the Mint and Monastery. It would not be logically consistent with a granary concept, however.
per trade route, +1 
per sea tile (better than per resource), +25% production of naval units, +1 Merchant slot 
boosts to water tiles with Harbors and the Research Lab in the latest update of the Terrain Improvements mod.
