Google “Project Raspberry” Data Center Campus

About this report

Clear, sourced information to inform public discussion

This independent report examines the proposed data center campus using publicly available information as of July 2026.

It addresses the primary concerns raised by many residents — water supply and rates, electricity rates, noise, and environmental impacts — alongside the potential fiscal benefits to the County. The analysis is factual and does not advocate for or against the project. Its purpose is to provide residents with clear, sourced information to inform public discussion and decision-making.

Independent & unaffiliated. This analysis is compiled entirely from public information and is not affiliated with, endorsed by, or produced on behalf of Google, Botetourt County, the Western Virginia Water Authority, or any agency. Its only aim is to present the best available data. Spotted an error or have a better source? Share an idea or correction →

Executive Summary

Key findings for Botetourt residents

Many Botetourt residents have expressed strong concerns about the proposed Google data center campus, particularly regarding water supply and future rates, electricity rates, noise from cooling systems and generators, and environmental impacts (wetlands, air emissions, land use, and drought resilience). This independent analysis directly examines those issues alongside the fiscal implications for the County.

Major Risks & Community Concerns

Section 01

Project Description and Current Status (July 2026)

1.1 Overview

Google (through affiliate/developer) is proposing “Project Raspberry,” a mission-critical hyperscale data center campus within the Botetourt Center at Greenfield industrial park in Daleville, Botetourt County, Virginia. The site encompasses approximately 343.6 acres. Plans include three data center buildings of roughly 300,000 square feet each (nearly 1 million sq ft total), three electrical substations, a ~28,000 sq ft office/support building, access roads, parking, stormwater management, and associated utilities. This would be Google's first major data center investment in the Roanoke Valley region.

1.2 Key Contractual Commitments (Performance Agreement)

• Minimum $1 billion capital investment per data center.
• Minimum 50 full-time permanent jobs per data center with median annual salary of approximately $86,000.
• 20-year term per data center once operational.
• Land sale from County Economic Development Authority (~312 acres for ~$14 million in June 2025) plus $4 million additional commitment to County initiatives over five years (already providing upfront funding for public safety equipment and other priorities).
• The project is the only location in Botetourt County currently zoned to allow data centers.

1.3 Current Status (as of early July 2026)

The project remains in the pre-construction and permitting phase. Grading permit applications were received by the County in February 2026, with on-site grading anticipated before the end of 2026. A Virginia Water Protection (VWP) permit application was submitted to DEQ in January 2026 and is under review. A Minor New Source Review (NSR) air permit application for backup generators was received by DEQ in May 2026 and is under review. The U.S. Army Corps of Engineers issued a public notice for Section 404 wetlands/stream impacts in March 2026. Appalachian Power is advancing the Daleville Area Transmission Improvements project (line rebuilds, substation upgrades) with public open houses held; Google is expected to fund attributable portions. A community open house was held in June 2026. No full-scale construction has commenced. Timeline for revenue generation (taxable status) is therefore later than some early projections and subject to successful completion of permitting, financing, and construction milestones. Community opposition has been vocal at public meetings, citing water use, noise, visual impacts, and proximity to schools/residential areas.

1.4 Scale Context for Botetourt County

Botetourt County's annual general revenue budget is approximately $99 million. A data center campus generating $10M+ per building in annual property taxes (as publicly projected by County officials) would represent a transformative addition to the tax base once fully operational and assessed. However, the County is small and rural; infrastructure commitments and resource demands must be evaluated against limited administrative and financial capacity to absorb overruns or long-term obligations.

Section 02

Tax Revenue Projections — Independent Model

2.1 Methodology and Key Assumptions (Transparent)

This model is built independently from public records, County economic development statements, Virginia assessment practices, and typical data center depreciation/refresh cycles in the industry. It does not rely on any prior third-party fiscal model.

Core Assumptions

• Three data centers brought online on a staggered realistic cadence reflecting current permitting status: first building taxable status ~2029, second ~2031, third ~2033 (conservative relative to some early announcements; delays are common in permitting-heavy projects).
• Taxable investment per data center: Minimum $1B as per performance agreement. Split for modeling: ~$300–400M real estate/improvements (relatively stable) and ~$600–700M business tangible personal property/equipment (servers, cooling, electrical — subject to depreciation and refresh cycles). Actual split and returns are confidential; sensitivity tested.
• Real estate tax: $0.70 per $100 of assessed value (2026 adopted rate). Assumes assessed value tracks modeled taxable real property value with limited disputes.
• Equipment tax: Subject to performance agreement provisions that include a grant/rebate structure above a defined effective rate on depreciated taxable value. Model uses an effective rate consistent with publicly described agreement mechanics (approximately $2.40 per $100 range after rebate; nominal general personal property rate is higher at $2.94 but not fully applicable to covered data center equipment).
• Depreciation/assessment: Virginia/Botetourt business personal property uses cost-based declining balance cohorts (typical 90%, 70%, 50%, 30%, 10% by acquisition year, then residual). Servers and IT equipment in data centers often depreciate rapidly; model tests both “no refresh” (value falls to low residual and stays) and “refresh every 5 years” (reset at original cost, consistent with industry statements that technology is upgraded on 4–6 year cycles).
• No major reassessment disputes or valuation appeals modeled in base case (real-world risk exists and would reduce revenue).
• One-time items: Land sale proceeds (~$14M) and community investment commitments (~$4M) already realized or in process; counted as upfront benefits in cumulative totals.

2.2 Annual Revenue Projection Table (Nominal $ Millions, Rounded)

2.3 10-Year Cumulative Revenue Summary (Independent Estimate)

• Real estate taxes (full campus, mature): ~$45–55M cumulative over decade.
• Equipment taxes — No refresh case: ~$80–110M cumulative (declining after initial years).
• Equipment taxes — With 5-year refresh: ~$120–160M cumulative (higher and more stable).
• Upfront land + community payments: ~$18–19M (already realized or committed).
• Total direct inflows (full buildout, before costs): Roughly $160–230M nominal over 10 years depending on refresh and exact valuation. This is a substantial addition relative to the County's ~$99M annual revenue base.

Section 03 · Highest Risk

Water Infrastructure — Detailed Risk & Cost Analysis

3.1 Current System Context

The Western Virginia Water Authority (WVWA) serves Botetourt County and the broader Roanoke Valley. Primary source for the relevant area is Carvins Cove Reservoir (6.42 billion gallons at full pond, ~630-acre surface, large watershed). Treatment capacity at the Carvins Cove plant is 28 million gallons per day (MGD). The system also includes Spring Hollow Reservoir (3.2 billion gallons) and smaller/interconnected sources. As of June 2026, Carvins Cove levels were approximately 13 feet below full pond (~69% capacity) amid ongoing dry conditions; the Authority has implemented conservation messaging and is supplementing from other sources. Historical drought of record (2002) saw levels drop more than 34 feet below full. The Authority states it has not had to implement mandatory restrictions in its ~20-year history due to source diversity, but acknowledges long-term planning for additional supply is prudent given regional growth.

3.2 Projected Data Center Water Demand

Public records and utility statements indicate the project reserves up to 2 MGD for Day 1 (initial phase) and up to 8 MGD total at full buildout (Day N). This scale is consistent with hyperscale data center evaporative cooling demands for large, high-density facilities (AI/cloud workloads tend toward the higher end). Industry data shows hyperscale facilities commonly consume hundreds of thousands of gallons per day per building via evaporative methods, with ~70–80% typically consumed (evaporated) rather than returned. Wastewater discharge would also increase load on treatment systems. The 8 MGD figure represents a very large industrial user relative to the existing residential/commercial base served by Carvins Cove.

3.3 New Water Source Need and Cost Reality

WVWA and County officials have publicly stated that a new regional water source was already contemplated in long-range planning (roughly 2040–2060 horizon) to meet baseline growth. The data center accelerates both the timing and urgency of that planning. Studies are now underway (with Google funding initial phases per agreements). Options under discussion include Carvins Cove expansion, new reservoir/intake, groundwater, or non-potable reuse — but no preferred alternative, engineering cost estimate, or detailed timeline has been made public as of July 2026.

County Exposure: Existing WVWA-County agreements establish a framework under which Botetourt has funding obligations for water supply development services and a cost-share formula that can place the majority of early project costs on the County (tapering at higher total project costs). Fixed deposits into a development fund are modest (low tens of millions cumulatively). The variable exposure — actual construction, land acquisition, permitting, mitigation, engineering, and debt service — is the material risk. Pro-rata or “developer pays attributable share” language exists and would reduce net County burden, but the baseline obligation and uncertainty around scope/timing/allocation remain. If a $300M+ project is accelerated into the decade and County share is substantial, debt service alone could add $10–20M+ annually for 20+ years depending on financing terms.

3.4 Water Risk Summary for Residents

• Scarcity & Drought: Adding a large consumptive user to a system already showing drought stress increases the probability and severity of future restrictions and the need for costly emergency measures.
• Cost Allocation Uncertainty: Without published detailed cost estimates and a clear, enforceable pro-rata methodology, residents cannot assess whether they will ultimately bear rate increases or tax-funded debt for infrastructure primarily benefiting a single large industrial user.
• Timing Risk: Studies are ongoing; if a new source is required sooner than modeled or costs exceed expectations, the fiscal and service impacts arrive earlier.
• Regional Equity: Carvins Cove and the broader WVWA system serve multiple jurisdictions. Botetourt-specific commitments have implications for Roanoke and other ratepayers.

Section 04

Power and Other Infrastructure

4.1 Power Demand and Upgrades

Hyperscale data centers are extremely power-intensive (tens to low hundreds of MW per building depending on density and IT load; AI-optimized facilities trend higher). The campus will require substantial new or upgraded transmission and substation capacity. Appalachian Power is advancing the Daleville Area Transmission Improvements Project, which includes rebuilding/upgrading approximately 17 miles of 138 kV lines, upgrading multiple substations, and constructing a new substation. Public information indicates this project supports growing demand including the Google campus while also improving reliability for existing customers. Google is expected to pay for upgrades specifically attributable to its project. Total project cost indications in public reporting are in the range of ~$135 million; the developer-funded portion for attributable assets represents a lower direct fiscal risk to the County than water. Indirect effects (any system-wide costs recovered through rates, or future capacity needs) should still be monitored. Statewide, data center growth is contributing to utility capital plans and, in some jurisdictions, rate pressures.

4.2 Other Infrastructure & Environmental

• Wetlands/Streams: USACE and DEQ permits required; impacts documented in public notices. Mitigation (credits or on-site) adds cost and schedule risk — primarily developer responsibility but with public review process.
• Air/Generators: Large backup diesel or alternative generation requires air permitting (NSR application under review). Emissions and noise are community concerns.
• Stormwater & Local Roads: Standard development requirements; developer bears primary construction cost, but long-term maintenance and any off-site impacts are County considerations.
• Overall: Most direct “project-specific” infrastructure is structured to be developer-funded. The larger, shared, or accelerated regional systems (especially water supply) are where County and resident exposure is greatest.

Section 05

Jobs, Economic Activity & Other Benefits

Direct Employment: Performance agreement requires minimum 50 full-time jobs per data center (150 total for three buildings) at median salary ~$86,000. This is typical for modern hyperscale facilities, which are highly automated. Construction employment will be significantly higher during build-out (hundreds of workers over multiple years) but temporary.

Economic Multipliers: Data center projects generate indirect and induced jobs in construction trades, maintenance, security, landscaping, professional services, and local vendors. County and Google statements reference multipliers in the range of several additional jobs per direct position and substantial state-level economic activity per facility. These are real benefits but secondary to the direct fiscal (tax) analysis for County budget purposes.

Other Benefits: Upfront land sale and community investment payments (~$18M total) have already enabled accelerated public safety and other capital projects without new borrowing. Potential for workforce development partnerships and Google's broader community/resilience investments (e.g., solar/battery projects in other Virginia areas). Long-term, a stable large taxpayer can support lower tax rates or enhanced services for residents if net fiscal positive after costs.

Section 06 · Illustrative

Scenario Analysis — Net County Impact

The following scenarios are illustrative ranges derived from independent research on buildout timing, tax mechanics, water cost benchmarks, and typical project risks. They are not predictions but tools to understand sensitivity. Actual outcomes depend on final permitted scope, exact taxable values, refresh behavior, water project costs/scope/allocation, and financing terms.

Interactive: model your own scenario

Adjust the assumptions below and the illustrative 10-year net direct fiscal impact updates live. It uses the same cumulative figures from Section 2 and the water-cost logic behind the scenarios above. This is a directional teaching tool, not a prediction.

Interpretation of Scenarios

• Full 3-DC, No Major Water Construction Cash in Decade: Strongest outcome — tax revenue ramp dominates. Cumulative net positive in low-to-mid hundreds of millions range before discounting.
• Full 3-DC + Moderate Water Cost ($150M County Share): Still positive but materially reduced. Debt service or cash outlay offsets a large portion of tax gains.
• Full 3-DC + High Water Cost ($250M+ County Share): Marginal or low positive; long-term obligation significant. Risk of net negative if costs overrun or pro-rata recovery is lower than expected.
• Partial Buildout (1 DC) Scenarios: Much lower revenue base. Even without major water cost, benefit is modest; with water cost, easily negative. Highlights buildout risk.

Added Research · Modeled Range

The Likely Outcome — a Monte Carlo Model

The scenarios above show fixed "what-if" cases. This model instead treats every major unknown as a range, runs 10,000 randomized simulations, and reports the full distribution of 10-year net fiscal outcomes — so we can talk about what is likely, not just what is possible.

Each simulation independently draws a value for every variable below from a distribution grounded in the research (fully sourced), then computes the County's net direct fiscal impact as gross tax inflows − the County's water cost. Running it thousands of times produces the range and "most likely" figure shown in the bottom-line bars at the top. The result updates live in your browser; the code is inspectable.

It is run over two horizons — 10 and 20 years — because timing changes the story. Over 10 years, the new regional water source has barely begun, so the County's water outlay is modest; over 20 years, that project is largely paid for, but the tax base has also had a decade longer to compound. The variable ranges below are for the 10-year model; the 20-year model raises the per-building tax figures accordingly, weights full 3-building buildout more heavily (Google's record is that campuses complete eventually), and models the water cost as the County's tiered, debt-financed share of the full project (100% to $100M, a partial share to $300M, WVWA beyond) rather than just early-phase spend.

The variables and where each range comes from

How to read it honestly

• The probabilities for buildings and refresh are informed judgment, not hard frequencies — public data on these is thin. Reasonable people could shift them; the model is transparent so you can.
• It models direct County general-fund impact only — not ratepayer water bills (see Precedents & Bill Impact), electricity rate effects, jobs multipliers, noise, or environmental costs.
• It is undiscounted nominal dollars and deliberately uses wide ranges. Treat the median as "central tendency," not a forecast.

Added Research · July 2026

Comparable Precedents & Resident Bill Impact

How other Virginia localities have actually fared with data centers — and an illustrative estimate of what a new water source could mean on a household water bill.

This section extends the base analysis with additional public reporting and the Virginia legislature's own oversight study (JLARC, December 2024). Every figure is sourced in Sources & References.

Virginia data center precedents

Virginia hosts the largest concentration of data centers in the world. The three localities below are the state's most mature markets and show both the fiscal upside and the infrastructure and ratepayer pressures that accompany it.

Illustrative resident bill impact

Method: Level annual debt service is spread evenly across the Western Virginia Water Authority's ~69,000 water accounts (serving Roanoke City, Roanoke, Botetourt and Franklin counties, Vinton, and Boones Mill). Debt service = P × r ÷ (1 − (1 + r)⁻ⁿ), where P is the principal, r the interest rate, and n the term. Interest rates (4.5–5.0%) and terms (20–30 years) are stated assumptions typical of municipal utility debt, not a specific bond issuance.

Caveats to state plainly

• The ~69,000 figure is water accounts system-wide; spreading a Botetourt-driven cost across the entire regional base is a simplifying choice. A Botetourt-only allocation would raise per-household figures sharply.
• Interest rate and term are assumptions, not a specific bond issuance.
• No published total cost for the new water source was found. The $150M and $250M scenarios are modeled, bracketed by reporting that the County could face “up to $100M” and “potentially $400M total” in water-supply project costs.

Section 07

Comprehensive Risk Register

Section 08

Recommendations for Residents and Decision-Makers

Before Irreversible Commitments

1. Water Project Transparency: Require immediate public release of the full water source study scope, preliminary engineering cost estimates (range and base case), trigger criteria for Day N/new source, and detailed pro-rata/allocation methodology. No major debt or construction commitment without this.
2. Independent Review: Commission a third-party fiscal and infrastructure analysis (including water/power cost sensitivities, NPV, and ratepayer impact modeling) by a qualified firm with no conflict. Make findings public.
3. Contractual Protections: Seek hard caps or clearer limits on County water exposure above certain project cost thresholds. Strengthen enforceability of pro-rata recovery and developer payment for attributable share.
4. Annual Public Scorecard: Establish by agreement or resolution a mandatory annual public report with aggregate (non-confidential) data on: taxable equipment values by facility, taxes paid, any grants/rebates issued, water usage and charges, water fund balances and invoices, debt issued for infrastructure, and milestone progress. Include independent verification option.

Ongoing Oversight

5. Quarterly or semi-annual public updates on permitting status, construction progress, and any material changes to scope or timeline.
6. Cost-of-service and drought management study that explicitly models data center incremental impact versus existing ratepayers and includes enforceable large-user curtailment provisions.
7. Comparative economic development analysis: Evaluate opportunity cost of land/water allocation versus diversified industrial or mixed-use scenarios.
8. Regional coordination: Engage Roanoke and other WVWA member localities on shared water planning, cost allocation, and equity to avoid inter-jurisdictional conflict.

Section 09

Conclusion

The proposed Google data center campus represents one of the largest single economic development opportunities in Botetourt County's history. Under a full three-building buildout with contained infrastructure costs and timely delivery, it would deliver a substantial and lasting expansion of the County's tax base — likely transformative relative to the current ~$99 million annual revenue budget. Upfront payments have already provided tangible community benefits.

However, this is a high-stakes transaction with asymmetric risks. Water infrastructure is the clearest and largest uncertainty: significant incremental demand on a drought-vulnerable primary source, acceleration of a major new supply project whose cost and County net exposure are not yet publicly quantified in detail, and long-term obligations that could offset or exceed tax gains depending on allocation. Power upgrades are better structured (developer pays attributable share). Buildout, equipment valuation, and permitting timelines carry meaningful downside. Residents and ratepayers face resource security, potential rate, and opportunity cost exposures that are not fully bounded in current public information.

This independent analysis is offered to support informed public discussion and decision-making. It is based on the best available public information as of July 2026 and makes no claim to inside knowledge or legal advice. Actual outcomes will depend on final permitted scope, contractual details, construction execution, and future economic/technological conditions.

Ideas & Corrections

Spotted something? Help improve the analysis

This is a living, independent analysis, and accuracy matters. If you have a correction, a public source, or an idea to make the data clearer, please send it along.

Your note goes privately to the author — you won't be emailing anyone directly, and nothing you submit is posted publicly. Sharing your own email is optional and used only if a reply would help.

Independent Research · July 2026

Sources & References

All information is drawn from publicly available primary sources. No confidential or non-public data was used. Links below go to the original documents so readers can verify every claim independently.

Project & Agency Filings

• Botetourt County — Google Data Center project hub: official overview, land purchase, timeline, zoning approvals, and documents.
• Botetourt County — Google Data Center economic impact page: County's projected tax revenue, jobs, and infrastructure funding.
• Virginia DEQ — Project Raspberry page: VWP application received Jan 21, 2026 (PEEP #25-1919) and Minor NSR air permit received May 13, 2026 (PEEP #21819-1); documents impacts to Tinker Creek and Rulman Branch tributaries.
• Virginia DEQ — Virginia Water Protection (VWP) / Wetlands & Streams permit program.
• Virginia DEQ — Air permit program (Minor New Source Review).
• U.S. Army Corps of Engineers (Norfolk District) — Section 404 public notice NAO-2025-01857 (issued Mar 10, 2026; comment period through Apr 9, 2026).
• Appalachian Power / AEP — Daleville Area Transmission Improvements Project: ~17 miles of 138 kV line upgrades, substation upgrades, and a new substation.

Water Supply

• Western Virginia Water Authority — Proposed Google Data Center at Greenfield Industrial Park.
• WVWA — Reservoir levels (Carvins Cove conditions) and Capital Projects index.
• Henrico County — Cobbs Creek Reservoir project: ~$280M, 14.8-billion-gallon reservoir used here as a municipal water-supply cost/scale benchmark.

Broader Studies & Industry Data

• Virginia JLARC — Data Centers in Virginia (December 2024): authoritative state study on fiscal, water, and electricity-ratepayer impacts (full report PDF · briefing deck).
• Industry and government data on hyperscale data center water consumption (evaporative cooling rates, Google facility averages).
• Broader Virginia data center economic and infrastructure analyses (local fiscal analyses and utility capital reports).

Precedents & Ratepayer Analysis (Section ✦)

• Cardinal News — Botetourt Google data center FAQ: buildings, 2→8 MGD demand, ~$8M near-term County payments, up to $100M / potentially $400M County costs, and Google pro-rata reimbursement.
• Botetourt County — Google Data Center: Water: pro-rata cost-share mechanism and new-source timeline.
• WDBJ7 — Carvins Cove levels amid drought: current production vs. capacity; Google to be the system's largest customer.
• WVWA — General Information: ~69,000+ water accounts and service area (ratepayer-base assumption).
• Loudoun County FAQ and Loudoun Water — Water Reclamation: data-center share of budget and reclaimed-water program.
• Bay Journal: Loudoun data-center water use (~900M gal, 2023).
• WTOP and the Prince William County 2024 Data Center Revenue Report: TY2024 revenue and counterfactual rate analysis.
• Richmond BizSense and VPM: Henrico / White Oak scale and zoning review.
• Statewide JLARC coverage: DataCenterDynamics, WRIC 8News, WVTF (the $18B-by-2040 and $14–$37/month figures).

Independence note: This is an independent report and is not affiliated with, endorsed by, or produced on behalf of Google, Botetourt County, the Western Virginia Water Authority, or any agency cited. The executed EDA performance agreement is referenced across the County's project pages; the full document can be requested from Botetourt County directly.