Wells, limits & cooling

Short answer

Yes, if the planned 400 MW IT load uses evaporative heat rejection. Dry and evaporative cooling have very different water requirements, and the final cooling design has not been published.

The local baseline

The advertised 2.5 MGD demand is about 1.25 MGD below the annual permit limit. That arithmetic does not account for peak demand, leakage, fire flow, drought, existing commitments or well limitations.

Cooling designs

Dry cooling

A sealed loop carries heat to radiator-like dry coolers and rejects it to outdoor air. Ordinary cooling-water consumption can be near zero, although the site still needs water for initial fill, maintenance, humidification, cleaning, employees, landscaping, and fire protection.

Evaporative cooling

A cooling tower rejects heat by evaporating water. It continuously needs makeup water and also discharges concentrated blowdown. An internal server-water loop may be “closed” while the final outdoor heat-rejection system still consumes substantial water.

What the agreement already settles, and what it does not

The signed development agreement prohibits open-loop and once-through cooling and requires closed-loop cooling or other technology “designed to minimize water consumption” (§11). It also bars private water wells and private sanitary sewer, requiring Liberty to apply to the City of Americus for service (§5(C)). These terms rule out once-through cooling and place demand on the city system instead of private wells.

The agreement does not cap consumption. “Closed-loop” describes the circuit, and a closed circuit may still reject heat through evaporation. The agreement contains no water balance.

The draft ordinance passed by the Zoning Committee on August 18, 2026 does not mention water use: no cooling technology, source, water balance, consumption limit or capacity certification. Section 11 of the PDA agreement is a contract term, not a city zoning standard. The city therefore cannot enforce it through this ordinance. Any contractual enforcement available to the PDA is separate from city code enforcement.

Scale scenario, not a project forecast

The Department of Energy publishes a full-load table of cooling-tower water use in gallons per day, indexed by chiller tonnage and by cycles of concentration: how many times water goes around the loop before it is discharged as blowdown. Fewer cycles means more blowdown and more makeup water. The table runs from 3 cycles to 8.

That table is indexed by tons, and a data center is described in megawatts, so getting from one to the other takes three steps. Each can be checked with a calculator:

1. Megawatts to tons. One megawatt of IT heat is 3.412 million Btu per hour; one ton of cooling is 12,000 Btu per hour. So 1 MW ≈ 284 tons, and a 25 MW load is about 7,100 tons.
2. Tons to gallons. The DOE table, at 24-hour operation, works out to about 54.8 gallons per ton per day at 3 cycles (its 100-ton entry is 5,480 gal/day) and about 43.8 at 6 cycles, consistent with its 5-cycle and 8-cycle entries, which come to 45.9 and 41.9 gallons per ton per day.
3. Cycles set the width of the band. The range below is that table read from 6 cycles at the efficient end down to 3 at the water-hungry end.

For the 25 MW row: 25 MW × 284 tons/MW = 7,108 tons. At 43.8 gal/ton/day, that is 311,000 gallons a day; at 54.8, it is 389,000. The annual column multiplies the daily figure by 365.

These figures assume full load around the clock and that every watt of IT heat is rejected through the tower. The 400 MW row applies the published project IT load to the same DOE method; it is a cooling-tower scenario, not a water forecast. Weather, utilization, economizers, auxiliary heat, and hybrid or partly dry designs all change the result. A design that rejects part of its heat to air uses correspondingly less water.

Other water risks

• Drawdown affecting municipal or private wells
• Diesel, transformer-fluid, or chemical spills
• Firefighting runoff
• Cooling-water chemicals and blowdown
• Construction erosion and concrete washout
• Abandoned or improperly constructed wells creating contamination pathways

Questions that resolve the issue

1. What is the final outdoor heat-rejection method?
2. What are average-day, maximum-day, maximum-month, and annual demands at full buildout?
3. Will the project use city water, a private well, reclaimed water, or a combination?
4. Has the city certified capacity after peak demand, leakage, drought, fire flow, and existing commitments?
5. Where will blowdown go, and what chemicals will it contain?
6. What does independent pumping and drawdown analysis predict for nearby wells?

The cooling schematic, water balance, and city capacity certification are records held by the city and the Development Authority. See how to obtain the records and reach the officials, and the contact page for the same roster with each official's published e-mail address.

Important historical context

An older USGS study identified a small Providence-aquifer cone of depression around Americus. The same study also concluded that groundwater resources in its broader study area were not significantly impaired by 1990 use. That decades-old evidence supports obtaining a current study; it does not prove present harm or present safety.