How much sun: The good, the bad, and the typical

Supplemental material for the Feature Article "Home photovoltaic systems for physicists" PHYSICS TODAY, July 2008, page 42.

Thomas W. Murphy Jr

People who live in the southwestern US enjoy its generally sunny climate. Inhabitants of the Pacific Northwest, on the other hand, think of the Sun as an occasional, unscheduled visitor during the winter months. In quantitative terms, the Southwest receives a greater flux of solar energy, or insolation. From 1961 to 1990, the National Renewable Energy Laboratory assembled data on insolation from 239 stations in the US (see http://rredc.nrel.gov/solar/pubs/redbook). In the continental US, the station reporting the greatest insolation was Daggett, California, in the Mojave Desert. Quillayute, Washington, on the Olympic Peninsula, received the least insolation. Saint Louis, Missouri, typifies the results from the continental US.

The table presents the number of hours per day of peak-equivalent sunshine—the number of hours the Sun would have to shine straight overhead on a cloudless day to give the total energy received. Cloud cover and illumination geometry reduce that value from the daily number of hours the Sun is above the horizon. The first numerical entry in each of the table's data cells is the daily average for peak-equivalent Sun hours at each station. The second number shows what is theoretically possible with no clouds and no horizon obstruction. Atmospheric extinction is considered, though; otherwise, perfectly tracking the Sun (two-axis tracking) would give a yearly average of exactly 12 hours.

The yearly average and best-month values vary by only a factor of two between the most and least insolated places in the continental US. The worst month sees a greater variation. Average daytime cloud cover can be surmised by taking the ratio of actual to theoretical Sun hours when two-axis tracking is applied. Note that the fixed-tilt scenario achieves a greater fraction of its theoretical yield than does two-axis tracking. The relatively poor performance of two-axis tracking indicates that the Sun is more often obscured when it is low in the sky, and thus the theoretically possible reward of two-axis tracking is diminished.