Lists of power plants

A thread on the openmod mailing list in early‑2021 sought information on power plants globally by installed capacity, commissioning date, geographic location (or country or region), thermal efficiency (or heat rate), annual production, and average CO2 emissions intensity (or annual CO2 emissions):

It seemed therefore useful to collect some of the information sources provided. The following resources are sorted first by role and then by chronology. If background information is available elsewhere via links, say on Wikipedia EN, then it is not duplicated here.

Of course energy systems are more than simply lists of power plants, but that information is often a starting point for assessing national and global decarbonization potentials.

Not all the information below is strictly open data. Some falls into an unfortunate gray zone covered by unwritten not‑to‑sue arrangements — something that European grid operators could fix in an instant by adopting CC‑BY‑4.0 licensing.

I do not think it good practice of continually fork, harvest, and locally enhance this kind of data without mechanisms for propagating useful changes back upstream. Moreover I think that more than about two hops in public is not useful either. Much of the effort below would indeed benefit from being treated as contributing to a genuine common resource and not just something to grab to get the research questions at hand resolved. Work on higher‑level data models, ontologies, and metadata are not included here but nonetheless forms part of the wider information jigsaw.

Methodological issues

Compiling lists of energy system assets is not necessarily that straightforward:

Wikipedia resources

Wikipedia EN has an article covering open energy system databases:

Data portals

Under the terminology adopted here, data portals are intended to be maintained — whereas snapshots are not.


OPSD is Open Power System Data and, among other things, covers power plants in Europe:


PowerExplorer was announced on this forum in April 2018. The PowerExplorer partners include the World Resources Institute, Google, and KTH Stockholm. More here:

Global Electrification Platform

The Global Electrification Platform (GEP), led by the World Bank Group and ESMAP, offers a global portal and also covers grid assets. The associated grid planning model is OnSSET but other frameworks can be supported. More information:


PowerGenome is a tool to quickly and easily create inputs for power systems models:

Open Energy Outlook

The Open Energy Outlook is a project confined to the United States that began in 2020. The project input data is stored in an SQLite database. The evolving documentation comprises a set of jupyter notebooks that themselves are a combination of markdown text, embedded SQL queries to render input tables, and code to create graphviz network diagrams.

Specific studies and snapshots

Reports often archive their datasets for reasons of transparency and reproducibility:

  • Jones, Dave (March 2020). Global electricity review. London, United Kingdom: Ember (previously Sandbag). Landing page. Report under full copyright. Datasets available but not licensed.

  • Kanellopoulos, Kostis, Matteo De Felice, Ignacio Hidalgo, A Bocin, and A Uihlein (2019). The Joint Research Centre power plant database (JRC-PPDB) — Version 0.9 — EUR 29806 EN. Luxembourg: Publications Office of the European Union. ISBN 978-92-76-08849-3. doi:10.2760/5281. Catalogue number KJ-NA-29806-EN-N. Reuse according to Commission Decision 2011/833/EU. Database archive on Zenodo. Dataset files licenced Creative Commons CC‑BY‑4.0.

  • Schram, Wouter, Ioannis Lampropoulos, Tarek AlSkaif, and Wilfried van Sark (2019). On the use of average versus marginal emission factors. ISBN 978-989-758-373-5. doi:10.5220/0007765701870193. Paper at 8th International Conference on Smart Cities and Green ICT Systems — SMARTGREENS held Heraklion, Crete, Greece. Creative Commons CC‑BY‑NC‑ND‑4.0 license. Contains list (table 1) for the Netherlands.

  • World Resources Institute (June 2019). CSV dataset version — 1.2.0. Use to initially stock the PowerExplorer portal (listed above).

Data tooling

Data tooling (for want of a better term) is an empty or populated dataset and code providing a fully‑connected electricity network model and offering additional functionality. That functionality may include integrity checks, simple electrical engineering calculations, and sophisticated storage management. The tooling also collects the results during and following a model run. It may also provide standard and bespoke reports and visualizations.


PowerSystems.jl offers much of the functionality just outlined and is written in the julia language. The associated dataset library project is called PowerSystemCaseBuilder.jl. More here:


Wikipedia supports the infobox power station template which automatically propagates its contents to Wikidata. An example of the kind of information available is indicated on the Drax power station article. (If anyone has made use of this facility, could they maybe comment below?)

Finally, if there are other resources or thoughts, perhaps people could mention them in reply?

And thanks also to those whose input was recycled here: @tom_brown @matteodefelice IA DT.

Power plants in Poland

Spreadsheet published 24 March 2021 by Fundacja Instrat (short and long links):

More background on this email thread:

Platts database no longer maintained

This openmod mailing list posting suggests the commercial Platts World Electric Power Plants (WEPP) database is no longer being maintained:

Dear all,

I wanted to share a final update of the global power plant database and inform you that we are (for now) stopping maintenance of the database as we are out of funding for this project.

We hope to continue updates and be able to scale the work if we can secure more resources. As you know the project is open source on GitHub, so if you want to contribute or maintain it (or know others who want to), please do so.


  1. The latest data update now includes 35,000 power plants, with 3700 solar plants being the main additions from our commercial partners. We also added AI based annual generation for hydro, solar and wind power plant . As we expand these methods to fossil fuel plants, this will deliver more accurate measures of water withdrawal and consumption and pollutants to air .
  2. The country level power sector profiles on RW provide a nuanced picture of efficiency, electrification, exposure to climate hazards, electricity access and opportunities for investment in renewable electricity.
  3. The dataset remains by far the most accessed on RW . For example, researchers identified) that 1/2 of the largest NOx air pollution point sources can be matched to power plants in our database and have built a global power plant dispatch model based on the data

In the future we have several visions that we would like to pursue with you, but we are also happy for you to continue to scale the work through the open source nature of the project.
We hope that in the future this work can be used to:

  1. De-risk energy sector investments by MDBs and private investors by identifying current and future climate risks.
  2. Internalize water risk and impacts on power sector infrastructure planning by combining it with Aqueduct water risk data.
  3. Highlight what populations are most affected by power sector air pollution and enable better compliance monitoring.

Johannes Friedrich
World Resources Institute
WRI is a global research organization that turns big ideas into action at the nexus of environment, economic opportunity and human well-being.
Africa | Brazil | China | Europe | India | Indonesia | Mexico | United States

1 Like

Noting abbreviations used in the above posting:

Thanks for all the resources collected here! I’m wondering, is any of those providing aggregate capacities at sub-national resolution? E.g., how much wind capacity is currently installed in a given NUTS2 (or GADM) region in Europe?

Aggregate power plant capacities in a region

If you have the lat/longitude of power plants such as provided in the GPPD, you can easily assign them to a specific shape (NUTS2 etc.). In Python, this should be possible with “geopandas” or the underlying “shapely” tool.

Here are some shapely examples (it requires that you have the polyfiles for the regions, which are i.e. provided in GADM, and the lat/long location of plants):

1 Like
Text and images licensed under CC BY 4.0Data licensed under CC0 1.0Code licensed under MITSite terms of serviceOpenmod mailing listOpenmod wiki.