Computation of solar radiation from sky cover

Abstract

A procedure for estimating global solar radiation from sky cover is developed from the records of 47 stations in the United States with long periods of radiation observations during the 10-year period, March 1961 through February 1971. It fits a general parabolic equation of the form Y --B + (1 -B)(I -N) P to the observations, where Y is the observed global solar radiation divided by clear sky radiation and N is the sky cover. The variables B (the point at which the parabola crosses the y axis) and P (a variable parameter less than 1.0) are selected to minimize the sum of the errors (Y -Yc.•c) •', where Yc.• is the calculated value of Y. The equation Y -B + (1 -B)(I -N) ø'6• is selected as most representative, and the B values in this equation that minimize the sum of the errors squared for the individual stations are shown in Figure 1. The average absolute error of the 5306 data points is 1.18 MJ m -2, or 7% of the average observed radiation. Because of the uncertainties of the observed global solar radiation and observed sky cover, the procedure should be used with caution, particularly for periods of less than I month. [NTRODUCTION X•r = 1/N(Srad/Crad q-N -1 ) (3) Engineers and hydrologists are frequently faced with the problem of estimating global solar radiation for a point where no measurements of radiation are available. Due to the limited number of stations in the United States where global solar radiation is measured it usually is necessary to estimate the radiation from percent of possible sunshine or sky cover. Hamon e! al. [1954] developed a graphical method for converting percent of possible sunshine into daily values of incident solar radiation for stations between 25øN and 50øN. Larnoreaux [1962] used this relation in his work of adapting modern evaporation formulae for use with computers. Baker and Haines [[969] attempted to expand the solar radiation network through correlation of radiation data with frequently

Campbell-stokes data for radiation calibration purposes in East Africa

“…We confirmed this result using more recently published [4] long term monthly climatological cloudiness and CampbellStokes averages for Tanzania [5,6]. The conclusion of a large scattering of ground-based observations of monthly averages of point cloudiness plotted against more accurate measurements is in line with results at American stations using Campbell-Stokes data [7] and solarimeters [8]. It is supposed to be due to the nature of sky cover observations, to geometrical differences between measurements of point and areal cloudiness which depend on cloud distributions over the sky and to different detector sensitivities [7,8,9].…”

Water, Energy, and Biogeochemical Model (WEBMOD), user’s manual, version 1

“…In the absence of weather fronts moving across the watershed, the temperature range observed at a given station on a cloudy day will be less than the temperature range observed on a clear-sky day, making temperature range a useful proxy for incoming solar radiation (Tangborn, 1978). The hyperbolic relation between observations of cloud cover and solar radiation established by Thompson (1976) for 43 stations in and around Oklahoma City, Oklahoma, is shown in figure 14. The procedure is applicable to humid regions with variable cloud cover.…”

“…cre is suggested by Thompson (1976) The estimated incoming solar radiation has a maximum value of radmax(one) fraction of potential incoming solar radiation. The estimated incoming solar radiation is further reduced on days when the average basin precipitation [basin_ppt(one)] exceeds the threshold ppt_rad_adj(nmonths), in inches.…”

“…The procedure is applicable to humid regions with variable cloud cover. The final incoming solar radiation is calculated by the following equations: B is the best fit coefficient obtained by Thompson (1976) (fig. 15), dimensionless…”

Use of a watershed-modeling approach to assess hydrologic effects of urbanization, North Fork Pheasant Branch basin near Middleton, Wisconsin

“…The model routine used to simulate the initiation, accumulation, and depletion of a snowpack in each HRU (daily computation) was based on the conceptual model of Obled and Rosse (1977). Daily shortwave radiation and sky cover was estimated from maximum and minimum daily air temperatures (Thompson, 1976;Tangborn, 1978).…”