Best Practices of Uncertainty Estimation for the National Solar Radiation Database (NSRDB 1998-2015): Preprint

Research output: Contribution to conferencePaper


It is essential to apply a traceable and standard approach to determine the uncertainty of solar resource data. Solar resource data are used for all phases of solar energy conversion projects, from the conceptual phase to routine solar power plant operation, and to determine performance guarantees of solar energy conversion systems. These guarantees are based on the available solar resource derived from a measurement station or modeled data set such as the National Solar Radiation Database (NSRDB). Therefore, quantifying the uncertainty of these data sets provides confidence to financiers, developers, and site operators of solar energy conversion systems and ultimately reduces deployment costs. In this study, we implemented the Guide to the Expression of Uncertainty in Measurement (GUM) 1 to quantify the overall uncertainty of the NSRDB data. First, we start with quantifying measurement uncertainty, then we determine each uncertainty statistic of the NSRDB data, and we combine them using the root-sum-of-the-squares method. The statistics were derived by comparing the NSRDB data to the seven measurement stations from the National Oceanic and Atmospheric Administration's Surface Radiation Budget Network, National Renewable Energy Laboratory's Solar Radiation Research Laboratory, and the Atmospheric Radiation Measurement program's Southern Great Plains Central Facility, in Billings, Oklahoma. The evaluation was conducted for hourly values, daily totals, monthly mean daily totals, and annual mean monthly mean daily totals. Varying time averages assist to capture the temporal uncertainty of the specific modeled solar resource data required for each phase of a solar energy project; some phases require higher temporal resolution than others. Overall, by including the uncertainty of measurements of solar radiation made at ground stations, bias, and root mean square error, the NSRDB data demonstrated expanded uncertainty of 17 percent - 29 percent on hourly and an approximate 5 percent - 8 percent annual bases.
Original languageAmerican English
Number of pages7
StatePublished - 2017
EventEuropean PV Solar Energy Conference and Exhibition - Amsterdam, The Netherlands
Duration: 25 Sep 201729 Sep 2017


ConferenceEuropean PV Solar Energy Conference and Exhibition
CityAmsterdam, The Netherlands

NREL Publication Number

  • NREL/CP-5D00-70165


  • guide to the expression of uncertainty in measurement
  • GUM
  • National Solar Radiation Database
  • pyranometers
  • uncertainty


Dive into the research topics of 'Best Practices of Uncertainty Estimation for the National Solar Radiation Database (NSRDB 1998-2015): Preprint'. Together they form a unique fingerprint.

Cite this