To fully characterize measured or modeled solar resource data, the data set should beaccompanied by a statement of uncertainty that will help the analyst to correctly apply the information and will provide the necessary context for the reliability of each value. For example, a full characterization of uncertainty provides a basis to assess the predicted output of planned solar conversion systems and is thus a key factor when determining the bankability of the project. Uncertainty can be thought of as the confidence one has in the data. However, it is important to determine the uncertainty using a standard methodology that others also can use and will obtain identical results. The Guide to the Expression of Uncertainty in Measurements (GUM) (ISO 2008) is an example of how to determine the uncertainty in measurements. GUM has been formalized by several organizations, including the International Bureau of Weights and Measurements (French acronym: BIPM), and published by the International Standards Organization (ISO).
In this chapter, the uncertainties associated with the measured or modeled solar resource data are discussed along with the validation of physical or empirical models that use such data. Precise methods to measure and model the solar resource are difficult to develop because of the rapidly changing nature of solar irradiance. While instrumentation is improving, the measurement or modeling of incident irradiance can have large uncertainties, depending on circumstances. The GUM methodology for quantifying uncertainty in either measured (Section 6.1) or modeled values (Section 6.2) is discussed in what follows. Note that the uncertainty in modeled data is typically obtained by comparison with reference measurements, which is why this development comes first.