Global Wind Power Tracker Data

It is the world's wind fleet as typed rows instead of a map you have to squint at. The Global Wind Power Tracker, produced by Global Energy Monitor (GEM), monitors utility-scale wind power development worldwide and aims to capture every project phase larger than 10 MW - onshore and offshore, announced through retired. The February 2026 release spans 166 countries and areas and 32,227 phases: roughly 1.1 TW of operating capacity plus about 2.65 TW sitting in announced, pre-construction and construction stages - over 33,000 projects in total.

The threshold is what keeps the file usable: nothing under 10 MW appears as its own row, so the rows are the projects that actually move grids and markets. Smaller installations are not discarded - they sit in a separate below-threshold set wherever data is readily available. Captive industrial wind, including wind dedicated to hydrogen production, is tracked alongside grid-connected plants rather than filtered out, and a green-hydrogen flag added in June 2024 marks phases earmarked for electrolysis.

Validation is a multi-source lift, not a single feed. Research draws on government plans and permits, company reports and news media, with regional rollouts phased across releases. GEM reports that some 97 GW of wind came online in 2025 while pipeline additions fell 13% year-on-year, and that between 2026 and 2030 nearly 365 GW of onshore and around 200 GW of offshore capacity are planned - still leaving global tripling goals short by roughly 1 TW. Within Datadory's catalog of 1,744 datasets across 159 viable industries, this slice scores 9/10 on our quality rubric.

Exactly as they land after normalization - three of the 32,227 phases, chosen because they show three different points in the lifecycle:

phase-id=L1XXXXXXXXXXXX  status=operating
  project=<wind farm name> / <phase name>
  capacity_MW=301.0  install=onshore  start-year=2022
  owner=<operating group> [100%]
  country=United States  region=Americas
  lat,lon=35.201700,-118.334500

An operating onshore farm: coordinates resolved against satellite imagery, start year filled because commissioning already happened. The second row shows the other end - a 1.2 GW offshore complex under construction, ownership split between a utility holding company and an infrastructure fund, start year still blank pending completion. The third is why the status column earns its keep: a 400 MW floating proposal marked shelved, retained in the file with its coordinates, owner and green-hydrogen flag intact so pipeline analyses count it as stalled rather than vanished.

One join key per phase, one eight-value enum for lifecycle state, one enum for how the machines meet the ground: the shape stays constant whether you pull ten phases or all of them.

Thirteen verified core fields, written against the published schema rather than guessed. Identification comes first: GEM Location ID and GEM Project ID are GEM's stable keys in L1+11-digit and G1+11-digit formats as of June 2024, followed by Project Name and Phase Name - a project may comprise several phases installed under separate permits and contracts.

Physical attributes are Capacity (MW) (nameplate rating per phase), the eight-value Status enum - announced, pre-construction, construction, operating, shelved, cancelled, mothballed, retired, with inferred suffixes applied after two and four years without observed progress - and Installation Type, which splits onshore from offshore and subdivides offshore into hard mount for seabed-mounted units generally at up to ~60 m depth, floating for deeper water, and mount unknown. Timing lives in Start Year: the year of commissioning or expected commercial operation.

Corporate attribution is Project Owner, researched from corporate disclosures, and Operator, which since June 2024 has been actively maintained only for China. Geography stacks three deep: Country/Area (renamed from 'Country' in June 2024 and not strictly bound to political borders), Region / Subregion for rollups, and Latitude / Longitude placed visually against mapping imagery, with proposed projects located from permit documents and company filings. The June 2024 Green Hydrogen boolean closes the dictionary. Anything adjacent confirmed during sample preparation folds under additional fields on request rather than being promised blind.

Geography: global by design. Every phase above 10 MW is in the file wherever it sits - 166 countries and areas spanning Asia, Europe, the Americas, Africa and Oceania, with region and subregion columns carrying the rollups. Coverage is comprehensive at and above the threshold; below it, projects appear in a separate set only where the data is readily available, so treat sub-10 MW comparisons between countries as uneven by construction.

Temporal: a rolling census rather than a history table. The live file captures today's operating fleet plus the full development queue, and the February 2026 edition is the reference for this page. New capacity additions are additionally summarized by country and year, so growth curves can be built without reassembling old releases - useful when the recent story is as lopsided as it is: 97 GW commissioned in 2025 against a pipeline that shrank 13% in the same breath.

Granularity: one row per wind project phase, rollable to project, country/area, region or installation type because the identifiers survive the aggregation. The onshore/offshore split - and within offshore, hard mount versus floating - holds at every level of aggregation, which is what makes supply-chain segmentation possible on a single file.

API, files, or your warehouse. Daily, weekly, or hourly.

Pick the channel your stack already speaks. The rows arrive identical either way - one record per phase with its identifiers, status, capacity, installation type, ownership chain and coordinates - extracted so nothing in your pipeline re-assembles a tracker by hand.

Every delivery ships with the field dictionary above plus the sample rows for validation, so your first join attempt happens against evidence, not hope.

• Market researchers and consultants size the wind market project by project and track which developers are still adding capacity; worked flows continue on our market sizing page.
• Investors and quant researchers split the 1,129 GW operating fleet from the ~2.7 TW queue and price pipeline quality off the status mix before committing capital; see investors and quants use cases.
• Grid and storage planners screen the construction queue and the green-hydrogen flag to find where flexible demand is materializing next.
• Data scientists and ML engineers get a typed, geocoded panel that joins cleanly on phase ID and maps without geocoding; see data scientists use cases.
• Journalists and academics cite project-level records assembled from permits, company disclosures and local reporting rather than press-release totals; see citation-grade research.
• Sales and growth teams work the pre-construction and construction lists, where turbine suppliers, EPCs and logistics firms are choosing vendors right now.

Market researchers and consultants get the structural layer - 32,227 phases, 166 countries, both installation modes - that turns a headline terawatt figure into a ranked customer or competitor list. Investors and quant researchers roll phase-level capacity up owner chains and read the health of a development platform from its status mix, not its announcement count. Journalists, academics and students get a research trail behind every farm - government plans, permit applications, company reports, news media - that makes 'the world installed another hundred gigawatts' checkable one project at a time. And strategy teams inside the supply chain get the queue itself: a floating-offshore filter over the 2026-2030 planned window is a shortlist of ports, vessels and foundations waiting to be sold.