I have a Chia city (a city constructed entirely on pause, with no sim activity allowed) with the water pump area separated from the rest of the city. There is only one pipe leading from the pumps to the city so it is an easy thing to turn the entire water supply on and off, while leaving the rest of the city entirely the same (in every respect: when the water to the city is turned of the pumps are still running and drawing power, etc.).

What I did then was to set up 4 trials, consisting of two periods of ten years each. I ran each trial 5 times to get a fair average.

Trial 1	Total Pop	Res Pop	Ind Pop	Com Pop	Res Zone	Ind Zone	Com Zone
1st 10 years Watered	204,994	84k	48k	71k	3489	1250	2716
2nd 10 years Watered	214,722	86k	48k	79k	3526	1257	2873
Trial 2	Total Pop	Res Pop	Ind Pop	Com Pop	Res Zone	Ind Zone	Com Zone
1st 10 years Dry	185,954	74k	47k	64k	3525	1203	2878
2nd 10 years Dry	183,518	74k	43k	65k	3523	1183	2885
Trial 3	Total Pop	Res Pop	Ind Pop	Com Pop	Res Zone	Ind Zone	Com Zone
1st 10 years Dry	184,766	74k	47k	63k	3513	2858	1216
2nd 10 years Watered	210,834	86k	47k	76k	3519	1240	2814
Trial 4	Total Pop	Res Pop	Ind Pop	Com Pop	Res Zone	Ind Zone	Com Zone
1st 10 years Watered	202,026	83k	47k	70k	3451	1225	2715
2nd 10 years Dry	214,578	86k	48k	79k	3529	1262	2962

1. It looks pretty clear to me that water has no effect at all on industry. While there is a downturn in industry in the second period of the 2nd trial (dry-dry) I think it was the result of continued poor Residential and Commerce figures (lack of customers?).
2. Water has a strong impact on Residential density. You'll notice that the figures for the number of residential zones is very consistent, but without water the density goes from 24.39 (Trial 1) down to 21 (Trial 2) sims per tile. It also seems to return to the higher level if the water is turned on (see Trial 3).
3. Commerce is effected also. Density decreases from 27.5 to 22.5. I'm not sure if this is a result of the lack of water or a secondary effect of the residential population decrease. I think I would have to design a new city to test this.
4. This is the oddest thing I see. In Trial 4, when the water was turned off it didn't seem to have any effect at all. The results from Trials 1 and 4 are nearly identical. This would seem to indicate that it only matters that the tile received water at some point. Or possibly that the effect of being denied water takes a long time to appear?

I used a res=6, com=3, ind=2 tax structure (I didn't differentiate between industries). This produces (when the city is full grown) enough revenue to cover city services and replacement of the power sources (2 fusion generators =2x$40,000/50 (years) =$1600/annum)) plus a small surplus. Since the entire point of a chia city is to be able to let it grow without any action on the part of the Mayor (after it starts) you have to use a cheat to get enough money to pay for services until the city grows to it's max size ( as well as funds to build the city, of course).

This is just a start on figuring out what water does. Any other opinions will be welcomed.

Bill Cooper