River Forest · Forsyth, Georgia · Neighbor Briefing
How far will the sound actually carry?
The Forsyth Technology Campus — roughly 12 million sq ft of data center west of I-75 — sits about 1.5 to 2.5 miles from our neighborhood. Sound fades fast with distance, but not evenly, and not the same at 2 a.m. as at noon. Here is a plain-English, physics-based look at what we would and wouldn't hear — with the honest caveats, so it holds up in front of the council.
01 The distance picture
Sound drops by about 6 decibels every time the distance doubles
That's the inverse-square law — the core of how noise fades across open land. The rings below show the estimated hum level at each distance from the cooling equipment. Two dashed paths show why the same facility sounds different by day and by night.
louder / near fainter / far night sound path
Levels are estimates from the inverse-square law plus typical outdoor losses, anchored to a mitigated source of ~55 dB at 1,000 ft. Real numbers depend on the final equipment and how well the sound walls perform — which is exactly why we should demand a measured study (see §07).
02 What each level sounds like
Putting a decibel number next to something you know
Decibels are logarithmic, so the jumps are bigger than they look — every +10 dB is roughly a doubling of loudness to the ear. Here's where the data-center hum is estimated to land at our distance, against everyday sounds.
110
Car horn, close up· rock concert
90
Gas lawn mower· motorcycle passing
85
18-wheeler passing at 50 ft· garbage disposalGenerator test — up close
The key isn't the hum's raw number — it's whether it rises above the background. At ~30 dB by day it sits under our 45–50 dB daytime background and disappears. At ~35 dB on a still night it can climb above a 25 dB background and become noticeable.
03 What we'd actually hear
Same facility, two very different experiences
Whether a distant hum is audible depends less on the source than on everything between it and our ears — and the air itself behaves differently by day and night.
☀︎ Midday ~12:00 PM
Likely not noticeable for most people, most of the time.
1Background masks it. Birdsong, breeze, lawn equipment and distant I-75 traffic sit around 45–50 dB — well above a ~30 dB hum, so it's buried underneath.
2Warm air lifts sound away. Daytime thermals rising off the ground bend sound waves upward, over our heads instead of toward us.
3Distance eats the highs. The air absorbs higher pitches over two miles, leaving at most a faint low tone that blends into everything else.
4Trees and terrain scatter it. The wooded buffer and our rolling ground break up and soak up part of what's left.
☾ Deep night ~2:00 AM
A faint, steady low drone is possible on calm, clear nights — like a highway that never stops.
1The background falls away. Ambient drops to ~25 dB, so a ~35 dB hum now stands above the silence instead of under the noise.
2Cool air bends sound down. On still, clear nights a temperature inversion refracts sound back toward the ground, so it carries farther with less loss.
3Wind direction matters. On nights the breeze blows from the site toward us, add a few decibels; the other way, it's pushed away.
4Low tones get through walls. The deep hum that survives the distance is the same frequency that passes through windows and walls most easily.
A neighbor's perspective
We already live next to the interstate
Here's the honest gut-check: I live right up against this community, and on a quiet evening I can already hear I-75 from my yard. If we ever hear the data center at all, it will be far softer than the highway we've all long since stopped noticing. That's how steady, distant sound works — the brain files it away as background, the same way it does a refrigerator or an HVAC unit humming in the next room. Within a week you don't hear it anymore. It just becomes part of the quiet.
The wildcard — and it's not in the permit
The monthly generator test
The continuous cooling hum is the steady background story. The louder, more disruptive event is the backup generators. At the city hearing, the facility was described as having roughly a thousand diesel generators — each about the size of an 18-wheeler — that must be run at least monthly to prove they'll work in an emergency.
That is a fundamentally different noise than the hum: louder, lower, and concentrated into a testing window. Right now nothing in the approved permit limits when it can happen or how loud it can be at our property lines — the city's noise ordinance has no decibel standard, and the council attached no noise condition. This is the single most important thing for the community to get a binding limit and schedule on before construction.
04 The water question
The scary number is a ceiling, not a daily draw
Water is the part that worries people most, and there's a number that gets quoted a lot. Here's what it means and what the approved permit actually requires — the honest version, good and bad.
☑ What eases the worry
1The 1.5M gallons/day is a design ceiling. That figure in the state filing is the maximum for peak-heat conditions, not a guaranteed daily draw — actual use is expected to be far lower.
2Closed-loop cooling recirculates. Instead of evaporating water away like old cooling towers, a closed loop reuses the same sealed water over and over. Comparable closed-loop campuses run on the order of tens of thousands of gallons a day — a small fraction of an evaporative design of the same size.
3The city runs its own treatment plant. Forsyth treats and distributes its own water and buys more from neighboring counties under contract. (Where the raw water is drawn from — and whether that source is shared — isn't spelled out in the project filings, which is one reason it's worth asking; see the next column.)
⚠ What to keep an eye on
1It's the same water we drink. The data center draws from the City of Forsyth system — the same one River Forest is on. So the real question isn't wells or aquifers, it's total capacity. The city's permitted withdrawal is around 4 million gallons/day; a 1.5M ceiling is a meaningful share of that.
2Drought is the pressure point. Forsyth and Monroe County have both asked residents to conserve during hot, dry spells. Adding a large industrial user is worth watching against those peak-demand periods specifically.
3"Closed-loop" isn't automatically near-zero. Some closed-loop designs still fall back on evaporative cooling in peak heat, which uses real water — and the electricity itself (partly from a coal plant ~20 miles away) has its own water footprint.
4The smart ask. A written cap on daily withdrawal, priority for residential supply during drought, independent verification of the closed-loop design, and a clear answer on where the city's raw water is drawn from and whether that source is shared.
Bottom line: the data center pulls from the same City of Forsyth water we do — so the honest issue is shared capacity, not a threat to private wells. The reassuring part is the closed-loop design and the city running its own treatment plant. The part worth pinning down in writing is where the raw water comes from, how the city protects residential supply, and a cap on the data center's draw during peak heat and drought.
05 What about our bills?
Will our water and power rates go up?
The most common worry after noise and water: does a giant new customer raise costs for the rest of us? Nobody can promise otherwise — but here's the honest push and pull, and the one thing to make the city commit to.
☑ The case it helps
1Tax revenue is the whole pitch. The project is projected to generate more than $220 million a year in local taxes — the stated goal was to ease the burden on residents and fund the city's aging water system.
2They pay for their own hookup. The approval requires the developer to fund new fire stations, water lines, and infrastructure — so residents aren't directly subsidizing the data center's connection.
3A big ratepayer can spread fixed costs. In principle, a large customer sharing the cost of the system can hold down the per-household share — if the contract is structured that way.
⚠ The case to stay watchful
1Capacity costs can flow downhill. If the city has to expand its water plant or buy more power to serve the data center and everyone else, those costs can land in everyone's rates.
2Power is the big one. Data centers draw enormous electricity, and large load additions have pushed up power costs for regular customers elsewhere in Georgia. Forsyth runs its own electric utility, so how they write the data center's power contract matters a lot.
3"Taxes will go down" is a projection. That figure comes from the people who backed the project — a promise, not a guarantee. Worth holding them to it rather than taking it on faith.
4The smart ask. Get it in writing that the data center's water and power costs are billed to the data center — not spread onto residential rates — and that residents see the promised tax relief, not just the developer.
Bottom line: the honest answer isn't "relax" or "brace yourself." The pitch is that the tax revenue lowers our burden — but that's a promise. The reassurance worth standing behind is a written commitment that the data center pays its own water and power costs, not us.
06 What about home values?
The one that worries people most — and the research is reassuring
A neighbor's honest opinion
I'm not losing sleep over this one. A data center isn't a warehouse, a highway interchange, or a factory — it's a quiet, fenced, low-traffic building most people forget is there. Could a buyer hear "data center" and hesitate on headline fear alone? Maybe. But two miles of woods between us and it, in a gated golf community, with a tax base that helps fund our schools and roads — I'd bet on this being a wash or a net positive long-term, not a hit. That's my read, not a guarantee — and the early research actually backs it up.
What the research actually shows
1The largest market on earth showed no drag. A 2025 George Mason University study of Northern Virginia — the world's biggest data-center hub — found no statistical evidence that being near a data center lowered home values. Homes closer actually sold for slightly more.
2A four-county study found them essentially even. Homes within 1.5 miles of data centers across four Indiana counties grew about 42% in value over the study period — nearly identical to the 41% growth in the surrounding market.
3A rigorous before/after study found "small and slightly positive." A Virginia analysis using a causal design ruled out substantial price declines near data centers — including newer AI facilities like this one.
4Distance is on our side. Where any discount does show up, it's for homes right next door — within a few hundred feet — not two miles away behind a wooded buffer. At our distance, the studies point the other way.
Honest caveat: these markets differ from ours and the research is still young, so it's genuinely reassuring rather than a promise. But every formal study so far points the same direction — proximity to a data center has not been the property-value hit people fear, especially at a couple miles' remove.
07 What this is — and isn't
Read this before you quote a number
What it is
A physics-based estimate using the inverse-square law and typical outdoor sound losses. It's a reasonable first picture, not a measurement.
The main assumption
A mitigated source of ~55 dB at 1,000 ft from the cooling equipment. If the real equipment is louder, or the sound walls underperform, every number here shifts up. If mitigation is better, they shift down.
What we don't know yet
The exact equipment location and orientation within the 1,630 acres isn't public, and the final cooling design (described as "closed-loop") isn't locked. Distance to the nearest boundary is estimated at 1.5–2.5 miles.
The regulatory gap
The site is inside City of Forsyth limits; most of River Forest is unincorporated Monroe County. The city ordinance has no decibel limit and enforcement requires an officer to witness the noise in person — a weak tool against a distant hum.
What to ask the city and developer for
1An independent acoustic study modeling day and night levels at River Forest property lines, including low-frequency content and inversion conditions.
2A binding decibel limit at the property line, written into the permit — since neither the ordinance nor the current conditions set one.
3A generator-testing schedule and noise cap — fixed daytime windows and a hard dB limit, not "at least monthly" with no ceiling.
4The signed conditional-use permit and site plan via a Georgia Open Records request, so we can see equipment placement and every condition as written.