Abstract

A solar spectral model that describes the solar radiation flux on a clear day at any given location was developed and tested. The model computes spectral fluxes of global, global photon, direct and diffuse solar radiation incident at the surface. Input parameters describe location and atmospheric characteristics. Location is described by latitude, altitude, slope orientation and surface albedo. Atmospheric characteristics described are turbidity, precipitable water vapor and total ozone content. The model was constructed using a one-layer homogeneous atmosphere with refinements, which are: 1. (1) Use of climatological data to predict the total ozone content, if it is not known. 2. (2) A more advanced treatment of infrared solar radiation (0.8–4.5 μm) absorption. 3. (3) A more complex scheme for predicting diffuse radiation. 4. (4) The capability of handling a spectral albedo. 5. (5) Inclusion of albedo dependence on zenith angle. Input parameters are minimized and several simplifying features are incorporated for ease of handling variables not routinely measured. Turbidity and total ozone content are treated as climatological estimates if specific location measurements are not available. Precipitable water vapor can be predicted using surface vapor pressure, since the sounding network is not dense. These features allow researchers outside the field of solar radiation to use the model. Because complete measurements with needed location and atmospheric characteristics could not be found, the validity of the model was tested by comparing it with a more complex, multilayered atmosphere model by Dave et al. [3]. Calculated fluxes of total direct, diffuse, and global radiation from the model presented were 11.5, 20.1 and 13.2 per cent lower, respectively, when they were adjusted for differences in extra-terrestrial solar radiation fluxes. Direct spectral fluxes closely agreed in spectral composition, with slight exception in the 0.8–0.95 μm region. Diffuse spectral flluxes were slightly higher in the UV region lower in the rest of the spectrum than were those in the multilayered atmosphere model. A sensitivity analysis of the model was also conducted: the most influential inputs were found to be latitude, slope orientation and turbidity, and the least influential was total ozone content. The hourly integrated values for the model compared very well with measured values for clear days at Davis. Discrepancies between predicted and measured values were due to lack of local turbidity coefficients.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call