What is the intended use case for this implementation of the polder? As they are currently balanced, the only use I can find for them is squeezing them in where you otherwise can't get decent farm adjacency. They max out at +2 Food, which I can beat with 3 farms in a triangle. By the time I'm able to build them at all, I likely have or could have Civil Service to get the same food yield from a farm, plus adjacency. It seems I'm trading a very valuable and rather difficult to obtain yield (food) for a bunch of easy to obtain and unhelpful yield (gold). I also get a production and a culture in the swap. When I'm making a choice between 2 Food, or 1 Production 3 Gold 1 Culture, the 2 Food is almost always the better choice.
You made farms give adjacency food bonuses to each other. You balanced your population growth and specialist food consumption to match the now-inflated food potential from the adjacency-fueled farms. If you want a Civ to have a unique improvement that takes the place of farms, they HAVE TO provide equal or superior food. This is why I suggested they buff village food yield instead of gold, and suggested higher Polder yields, in the post I made
here. As a reminder, these were the suggested yields:
Base 3 Food 1 Gold
with Economics Polder gains +2 Gold
with Chemistry Polder gains +1 Culture
adjacent Villages +1 Food
An example from an (effectively) late-game city. Six tiles and their yields, first as Polders and a Village, then as Polders with 2 Villages, then as Farms. Total yield counts from each, in spoiler text and above image.
So we're trading 14 or 16 Food to get 7 or 6 Production, 22 or 26 Gold, and 6 Culture. Conversely, specialists in Information Age cost 6 Food each, so we can support a couple more Specialists with the Farms. We can get 18 Gold, 4 Production, and Great Merchant points using merchant specialists. We can get 18 Production 8 Science from Engineers, 18 Science 4 Culture from Scientists, 18 Culture 6 Production from Artists, 18 Culture 6 Science from Writers, 16 Culture 8 Gold from Musicians, and 6 Culture, Science, and Gold from Civil Servants. This is before Wonders, Policies, and Religion. With Freedom, that becomes 5 more Specialists instead of 2. On top of greater yields for everything but Gold, you also get Great Person points from working these specialists.
If we were to use the yields I suggested earlier, we would have these results:
Polders, 1 Village: 27F 19P 20G 6C
Polders, 2 Villages: 28F 19P 22G 6C
Farms: 34F 17P 0G 0C
This is a trade-off worth making. The 2 village yields show that it takes a river bend where you can get 4 polders adjacent to 1 village to get superior yields compared to just putting a polder there. Looking at these, the polder gold yield is likely too high and should come down by 1.
Something similar to this change needs to happen if polders are to be a useful improvement we want to build when possible. I'm sure this isn't the only solution, it's simply one solution that looks good to me.