Geospatial records (GeoTIFF) for photovoltaic energy production on optimally inclined plane (kWh/kW) with a) satellite-based solar radiation from SARAH3 and b) ERA5 reanalysis data
Gounari, Olympia; Martinez, Ana; Falangas, Alexandros; Taylor, Nigel; Alexandris, Nikos (2026): PVGIS 6 Geospatial Solar Radiation and PV Performance Data. European Commission, Joint Research Centre [Dataset] doi: 10.2905/JRC.KVTK9NG PID: http://data.europa.eu/89h/131b88ed-74b7-4ceb-81da-b48b49a47ce1
photovoltaic energysolar radiation
Geospatial records (GeoTIFF) for photovoltaic energy production on optimally inclined plane (kWh/kW) with a) satellite-based solar radiation from SARAH3 and b) ERA5 reanalysis data. NB uses orignal PVGIS crystalline silicon performance mode land system losses of 14%
Values of electricity yield (kWh/kWp) for PV systems calculated by PVGIS. These are calculated using the PVGIS 2025 crystalline silicon performance model and assume system losses of 14%. There are 4 sub-sets:
1) for a horizontal system with SARAH3 irradiance values
2) for an optimally tilted system with SARAH3 irradiance
3) for a horizontal system with ERA5 irradiance values
4) for an optimally tilted system with ERA5 irradiance values
This layer gives the average hourly T_2m values from hourly data 2014-2024 from ERA5-LAND (land areas) and ERA5 (coastal areas).
Values of solar irradiation (kWh/m2) extracted and processed by PVGIS from satellite (SARAH3) and reanalysis (ERA5) data . There are 4 sub-sets:
1) on a horizontal plane with SARAH3 irradiance values
2) for an optimally tilted PV system with SARAH3 irradiance
3) for a horizontal plane with ERA5 irradiance values
4) for an optimally tilted PV system with ERA5 irradiance values
The Joint Research Centre's Photovoltaic Geographical Information System (PVGIS) has provided solar radiation data for over 20 years. PVGIS relies on meteorological datasets from satellite observations and reanalysis data which require periodic updates to capture climate change effects. PVGIS 6 presents a complete rewrite in Python that introduces a modular architecture, a FastAPI‑based backend, and a versatile command‑line interface. The aim of this report is to describe the datasets provided in PVGIS 6 and present the main differences users should expect in comparison with PVGIS 5.3. The new system leverages high‑performance libraries (NumPy, xarray, zarr, netCDF4) and provides database‑less, chunked access to time‑series data. PVGIS 6 incorporates updated meteorological inputs—CMSAF SARAH‑3 satellite irradiance (2014‑2024) and ECMWF ERA5/ERA5‑Land reanalysis (2014‑2024)—with higher‑resolution digital elevation models (1 arc-second SRTM and 30 arc-seconds GTOPO30) and a globally merged horizon‑profile product. As in PVGIS 5.3, a combination of ERA5 and ERA5-Land datasets is used, to fill coastal gaps in ERA5‑Land, namely ERA5LandPlus. Comparative analyses between PVGIS 6 and the legacy version 5.3 (covering the overlapping period 2014‑2023) show that global irradiation, energy yield, and temperature outputs are highly consistent, with average differences of ≤ 0.6 kWh m⁻² for irradiation, ≤ 15 kWh kWp⁻¹ for energy yield, and essentially zero for air temperature; the largest discrepancies occur in mountainous regions due to subtle shifts in horizon profile, air temperature and DEM datasets. These results are further supported by comparisons of air temperature and global horizontal irradiance with ground-based measurements from Baseline Surface Radiation Network stations (BSRN) located within the SARAH-3 footprint. However, none of the BSRN stations are in coastal areas covered by the extended dataset, and this should be addressed in further studies if other sources of reliable long-term measurement data can be identified.
| From date | To date |
|---|---|
| 2013-01-01 | 2024-12-31 |