Albedo data
Modelling recommendations
In PV simulation software, monthly albedo values are typically needed as input. Accurate albedo values are more important in systems where the reflected irradiance can have a high impact, such as bifacial systems [1] and systems with high tilt.
- Satellite-based data sources: For estimating albedo outside the winter season for areas where the ground cover will not be changed, and where the area is larger than the pixels in the satellite data, we recommend satellite-based data sources such as Solargis (paid service) or data from NASA based on MODIS measurements (free).
- Data from NASA [2] can be downloaded from AppEEARS. Search for MCD43A3.061 and select Albedo_WSA_shortwave and Albedo_BSA_shortwave. To obtain actual albedo, black sky (BSA) and white sky albedo (WSA) must be weighted with the diffuse ratio (kdiff=DHI/GHI) every day at 12: Albedo = WSA*kdiff+(1-kdiff)*BSA [3].
- Typical albedo of ground cover: For months and locations where satellite-based data is not reliable, or if accurate values are not needed, we recommend estimating albedo based on typical values for the relevant ground cover.
- Example – PV system with grass ground cover: 0.2 in months without snow, and 0.8 in months when snow on the ground is expected most days. For months with partial or varying ground snow cover, the value can be weighted according to the expected number of days with snow.
- For PV systems with multiple rows, be aware of the shading loss factor of the reflected irradiance in PV simulation software. As an example, PVsyst assumes a 100% loss of reflected irradiance between rows in multi-row systems for front-side irradiance. This will underestimate the irradiance of high tilt systems in high albedo environments.
Background: State of the art & knowledge gaps
Ground albedo, which is a measure of the amount of reflected irradiance, can vary significantly over time, for example due to seasonal changes in vegetation, water content in the ground, solar elevation, weather conditions, or snow cover. Snow can result in very high albedo, but it varies with snow type, age and ground coverage, making wintertime albedo especially challenging to estimate. For ground-mounted PV installations, groundwork during construction can alter surface characteristics, making historical albedo values for the same area unreliable.
In current PV simulations, albedo is commonly estimated using either the typical albedo of the ground cover or satellite-based data. However, the accuracy of satellite-based data is challenged by frequent cloud cover, difficulties in distinguishing snow from clouds, and mixed land cover within pixels. Wintertime albedo estimations are especially challenging based on satellite data, highlighting the need for methods accounting for snow duration and the temporal variability of snow albedo.
ReFERENCES
[1] M.M. Nygård, M.S. Wiig, N. Roosloot, G. Otnes, M.B. Øgaard, H.N. Riise, E.S. Marstein, Elucidating uncertainty in bifacial photovoltaic gain estimation, Solar Energy, 2025.
[2] NASA, LAADS DAAC, MCD43A3 - MODIS/Terra+Aqua BRDF/Albedo Daily L3 Global - 500m, dataset.
[3] W. Lucht, B.S. Crystal, A.H. Strahler, An algorithm for the retrieval of albedo from space using semiempirical BRDF models, IEEE Transactions on Geoscience and Remote Sensing, 2002.
Last update: 27.4.2026