GRIDMATCHGRIDMATCHDC Vulnerability Score — how sensitive a community is to data center development, and to what specifically. It combines observed public actions (moratoria, zoning restrictions, lawsuits, contested hearings — each cited to source) with modeled indicators (water stress, electricity rates and energy burden, climate vulnerability, labor markets, farmland). Where a real event exists, it always takes precedence over the model.
DC Demand Score — how much data center demand a county's population and economy generate. The DC Demand Score is an original GridMatch estimate; no standard method exists for attributing data center demand to local geography. It allocates national data center electricity consumption (LBNL 2025 Update, cross-checked against IEA 2025) using county digital-economy activity (BEA, BLS, Census), broadband- connected population and income, and a flat per-capita share. AI-training load serves national and global model development and has no meaningful "demanding county" — it is allocated per-capita by stated convention. Cryptocurrency-mining consumption (EIA estimate) is excluded from Demand entirely; crypto facilities appear only in Contribution where hosted.
DC Contribution Score — how much data center capacity a county physically hosts, from open facility inventories (Server Country, Data Center Tracker), weighted by project status. Free-source MW inventories undercount; most counties carry B/C confidence tags — this is expected and shown honestly.
DC Equilibrium Score — a county's net position. A high Equilibrium score means a county hosts more data center infrastructure — and bears more community sensitivity — than its own demand for digital services explains. A low score means a county consumes digital services largely hosted elsewhere.
Every score is flagged OBSERVED or MODELED with a confidence level (A/B/C). MODELED values are statistical estimates from environmental, economic, and demographic indicators — they are not observed conditions and do not predict the outcome of any specific project. OBSERVED values reflect documented public actions and facilities, cited to source.
Statewide data center policies (e.g., the New York Responsible Data Center Development Act, 2026) are tracked as first-class records and shown as a badge on every county in the affected state, independent of local county events. Rescinded local measures are never shown as active; they retain a reduced "demonstrated opposition" weight.
datacentertracker.org (CC BY 4.0) · Moratorium Nation — Bommarito, "Moratorium Nation," SSRN working paper 6242898 (CC-BY-4.0) · Bommarito, Michael J. II. (2026). Server Country Data Center Database. Retrieved from https://servercountry.org/data/downloads. · WRI Aqueduct 4.0 (CC BY 4.0) · U.S. Climate Vulnerability Index (Texas A&M University / Environmental Defense Fund) · FEMA National Risk Index · EIA · BEA · BLS · U.S. Census Bureau · USDA · U.S. Drought Monitor · NOAA/NEMAC Climate Mapping for Resilience & Adaptation (CMRA) · DOE LEAD Tool · HUD · GDELT Project · LBNL · Cleanview (cross-check) · Named polls: Gallup (May 2026), Pew (Mar 2026), Washington Post–Schar School (Mar 2026), Marquette (Apr 2026).
This map presents GridMatch's synthesis of publicly available data on community sensitivity toward data center development and the geography of data center demand and hosting. County scores marked MODELED are statistical estimates based on environmental, economic, and demographic indicators — they are not observed conditions and do not predict the outcome of any specific project. Scores marked OBSERVED reflect documented public actions and facilities, cited to source. The DC Demand Score is an original GridMatch estimate; no standard method exists for attributing data center demand to local geography. This tool is informational and advisory in nature. GridMatch LLC is an advisory firm and does not provide engineering, EPC, or construction services. © GridMatch LLC. Delivering certainty. Power. Positioned.