Thanks for the info Dan and welcome to the boards!
Just wanted to add:
the break-even point of a harbor buyout as soon as the city is built (420 gold) with an average of surplus food of X/turn (on standard speed) is given in the following table:
FpT - Turns Needed for profit after buying Harbor at pop 1 (unmodified):
------------------------------------------------
1.0 - 200 Turns
1.5 - 161 Turns
2.0 - 139 Turns
2.5 - 124 Turns
3.0 - 114 Turns
3.5 - 106 Turns
4.0 - 100 Turns
4.5 - 94 Turns
5.0 - 90 Turns
5.5 - 86 Turns
6.0 - 83 Turns
Looking at the above chart, you're probably thinking... Why are there diminishing returns on pumping more food quickly to cities? If anything, you'd think it opposite... follow the math:
Here's an extreme example to fetch the point:
New City is made and purchases a harbor for 420 Gold. You play golf with the mayor of every maritime C-S to the effect that it generates surplus +10 Food, on average.
By turn 4, it's already Size 3. It's just now starting to turn a profit (not much), because at size 2, the 3 gold upkeep on a harbor is still -.5/turn. We're still down 420Gold on our investment however
By turn 16 it turns size 6, but due to the 3.25GPT was earning our gross revenue is still only 23 gold.
On turn 30, we're up to size 8, revenue about 95 gold (but now we're up to 7 GPT)
Turn 40, Size 9... 158 Gold.
Turn 50, Size 10... 245 gold
Turns out we have to wait until turn 67 to our revenue to reach our initial 420G investment. Keep in mind, this is purely for trade purposes. In 67 turns, at that growth rate, the city will be massively profitable, just not from trade value before that time.
The moral of the story is that you for sure want to wait until Pop 3 to build the harbor, and may want to wait until pop 6-8 to buy one outright.
Note: Anybody feel free to check my math, no guarantees it's correct at such a late hour