In the section discussing components of light, we reviewed the effects of the atmosphere on irradiance. Now, we discuss how to model the transparency of a hypothetical sky that is "clear" of the effects of clouds. Clouds are a major contributor to the reduction of terrestrial irradiance; however, particles in the atmosphere also play a significant role. There are physical properties of the sky that we cannot see with our eyes, but which strongly affect the solar resource on a clear day. As we will see in the reading, aerosols and water vapor present in the atmosphere play an important role in scattering light, and they may be present on "clear sky" days when visible clouds are absent.
Daily or monthly irradiance data is required for proper design of any solar energy collection system. However, this data is not always available. This requires the use of well-designed models to estimate irradiance. Hence, the need for clear sky models.
Clear sky models are used to estimate what is called a clearness index. For a location, a clear-sky model must be properly calibrated to provide an accurate measure for the clearness index. We will be looking at two modeling approaches, the Bird Clear Sky Model (which we can download and use as a spreadsheet) and the REST2 model by Gueymard (which can also be downloaded and run as an executable file--we will not do so here).
The Bird Clear Sky Model was developed by Richard Bird and a number of other scientists at what is now the Department of Energy National Renewable Energy Laboratory. The model requires the following input data
Many openly available codes incorporate this model. The output is a "clear sky" estimate for the total or global horizontal irradiance (GHI), direct normal irradiance (DNI) and diffuse horizontal irradiance (DHI, or DIF) across wavelengths from 305nm to 4000nm. The model calculates these conditions for a single point in solar time, given the latitude (), longitude () and the Time Zone.
Download RReDC: Bird Clear Sky Model [3]
Note: You will be asked to use this tool in this week homework assignment.The REST2 model has been found to be most accurate, as we shall observe in our reading. We will only need to explore one method (Bird) for this class, but it is important that a resource professional is also aware of the modern application of clear sky modeling. REST2 accepts atmospheric inputs of:
The REST2 model will then output estimations of diffuse horizontal irradiance (DHI), direct normal irradiance (DNI), and global plane of array (POA) irradiance.
Links
[1] http://www.sciencedirect.com/science/article/pii/S0038092X11004221
[2] https://www.nrel.gov/docs/legosti/old/761.pdf
[3] http://rredc.nrel.gov/solar/models/clearsky/
[4] http://en.wikipedia.org/wiki/Angstrom_exponent
[5] http://earthobservatory.nasa.gov/GlobalMaps/view.php?d1=MODAL2_M_AER_OD