Solcast vs NREL NSRDB
The NREL NSRDB is a specification-limited free source of medium-quality global satellite-derived irradiance data, with a regional focus on North America, which provides data that ends between 2019 and 2021, depending on the region. The NSRDB database is free-to-access but has fewer features, limited spatio-temporal coverage, and higher uncertainty when compared with Solcast data.
About NREL NSRDB
The NSRB (National Solar Radiation Database) is a dataset created by NREL (the US National Renewable Energy Laboratory). The dataset, which runs through to the end of 2019 to 2021 (depending on the region), has a particular focus on North America, with data available for other regions with lower specifications.
Data Features and Capabilities
| Solcast | NSRDB | |
|---|---|---|
| Free trial with instant access and data download? | ||
| Price point | $$ | Free |
| Download wait time | <1-30 seconds | 5 minutes to several hours |
| Comprehensive, global, independent validation | ||
| Validation Sites | 46 | 9 (USA only) |
| Satellite based estimation | ||
| Global Coverage | ||
| Resolution of satellite data used | 1-2 km | 4 km |
| 15+ years of satellite data at full temporal resolution | ||
| Ignores older, less reliable satellites |
Source: Evaluation of the National Solar Radiation Database (NSRDB Version 2): 1998–2015
Inputs and Algorithms
The NSRDB and Solcast methodology is relatively similar. Both are semi-empirical and are satellite-derived. Both rely on validated, published models to build a clear sky model, and use proprietary cloud detection. The NSRDB cloud model has a lower resolution than the Solcast model, based around a 2-10 km grid, compared to Solcast’s 1-2km. Solcast has real-time data available, whereas NSRDB ends in 2019 to 2021 (depending on region). The NSRDB coverage begins between 1998 and 2017, and ends between 2019 and 2021, depending on the region. For some regions, this temporal coverage is insufficient for sampling interannual variability and production of long term averages. In other regions, including North America, NSRDB uses data from lower-resolution, pre-21st century satellites. Solcast only uses data from recent generation geostationary meteorological satellites (GMS). We do not use data prior to 2007 due to climate change and satellite data quality issues. This maximises data quality and validity, while still providing 15+ years of data history from which to sample for interannual variability.
Validation and Accuracy
Meta analysis of North American validation results: GHI results
| Solcast | NSRDB | NSRDB | |
|---|---|---|---|
| Performed by | DNV | IEA PVPS | Yang & Bright 2020 |
| Year published | 2023 | 2023 | 2020 |
| No. of sites | 38 | 28 | 15 |
| Mean Bias | +0.77% | +0.86% | +2.00% |
| Bias Std. Dev. | ±1.40% | ±4.05% | ±3.91% |
| Mean nMAD (nMAE) | 9.38% | 12.83% | Not Published |
| Mean nRMSD (nRMSE) | 14.75% | 18.13% | 20.29% |
Meta analysis of North American validation results: DNI results
NSRDB and Solcast are both independently validated, globally, NSRDB at 70 sites and Solcast at 207 sites.
| Solcast | NSRDB | |
|---|---|---|
| Performed by | DNV | IEA PVPS |
| Year published | 2023 | 2023 |
| No. of sites | 22 | 28 |
| Mean Bias | +0.85% | +7.34% |
| Bias Std. Dev. | ±3.32% | ±7.43% |
| Mean nMAD (nMAE) | 20.01% | 22.05% |
| Mean nRMSD (nRMSE) | 32.44% | 33.27% |
References
Yang, D., 2018. A correct validation of the national solar radiation data base (NSRDB). Renewable and Sustainable Energy Reviews, 97, pp.152-155.
Yang, D. and Bright, J.M., 2020. Worldwide validation of 8 satellite-derived and reanalysis solar radiation products: A preliminary evaluation and overall metrics for hourly data over 27 years. Solar Energy, 210, pp.3-19.
Forstinger, A., et al. (2023). Worldwide benchmark of modelled solar irradiance data (2023 PVPS Task 16): Solar resource for high penetration and large-scale applications. ResearchGate.
Cuevas-Agulló, E., et al. (2023). A new global high-resolution solar resource dataset. Zenodo