About the data

Physical potential is an InfraMap model of the net energy a plant’s dated equipment could have produced from the weather it actually experienced, assuming full availability and no curtailment. Solar uses NSRDB irradiance, wind uses HRRR or ERA5 winds at hub height, and thermal potential uses seasonal net capability.

Expected generation estimates a plant’s monthly non-storage net generation from its physical potential and its own earlier reported output. It is a retrospective hindcast, not a forecast, and it uses no price inputs.

Both are model outputs, not reported data. A gap between reported output and the model is a lead to check, not a measured loss.

This is a model output, not observed data. InfraMap labels it as a model everywhere it appears.

Potential: Solar and wind 2010 to 2025, with a 2026 archive to October 1, 2026; thermal monthly for 3,473 projects, January 2010 to August 2026
Expected generation: January 2010 to September 2026; thermal to August 2026
Recent days: From September 1, 2026, refreshed about every six hours
Validation: WAPE on retrospective months: solar 14.2%, wind 15.2%, thermal 26.4%

Access

Part of InfraMap Pro. See pricing.

Where it appears: Output tab, agent, MCP.

MCP tools that return it: get_project_generation_potential, explain_output_with_weather. The MCP server is part of Pro; setup is on MCP server.

Sources

National Solar Radiation Database (NSRDB)(opens in a new tab), National Laboratory of the Rockies, formerly NREL
High-Resolution Rapid Refresh (HRRR)(opens in a new tab), NOAA, via AWS Open Data
ERA5 reanalysis(opens in a new tab), ECMWF, Copernicus Climate Change Service; retrieved through the Open-Meteo archive API
EIA-860 and EIA-860M(opens in a new tab), U.S. Energy Information Administration
EIA-923 Power Plant Operations Report(opens in a new tab), U.S. Energy Information Administration
U.S. Wind Turbine Database (USWTDB)(opens in a new tab), U.S. Geological Survey

Periods and update notes for each source are on Data sources.

Known gaps

Limits

Output below potential is not assigned a cause. It can reflect curtailment, outages, other limits, physical losses or model overprediction.
Expected generation uses realized weather and revised EIA data, so it is not what could have been predicted at the time.
Snow, shading and bifacial gain are not modeled. All values are provisional, recent values more so.
It covers today’s fleet, not retired plants. Aggregate error does not guarantee accuracy for any single plant.

Missing values stay missing: a null means the source has no usable record, never zero. Zero and negative values are shown as reported.

Method and validation

How the model is built, what it was tested against and where it is weak is on Methodology.

Questions it answers

How did a given solar plant’s 2025 output compare with its physical potential each month?
In which months of 2025 did a given wind farm produce below 60% of what its weather predicts?
Which wind plants in SPP produced furthest below their weather-based expectation in 2025?

Ask the analyst

Terms: Capacity factor, Curtailment.

Monthly power plant generationMonthly net generation, battery charging and discharge, and fuel heat input for US power plants from EIA-923, January 2000 to July 2026.Plant and airport weatherHourly point weather at every US wind and solar plant from 2010, and NOAA GHCNh hourly weather at about 2,300 US airports from 2010.Modeled curtailmentExperimental InfraMap estimate of the energy US wind and solar plants likely withheld, from physical potential and the real-time price at each node.