Elsevier

Renewable Energy

Volume 18, Issue 4, December 1999, Pages 535-556
Renewable Energy

Data bank
Mapping global, diffuse and beam solar radiation over Zimbabwe

https://doi.org/10.1016/S0960-1481(98)00782-4Get rights and content

Abstract

A database for long-term monthly radiation over Zimbabwe is developed. The meteorological raw data inputs are long-term monthly average records of pyranometer-measured hemispherical radiation, monthly average sunshine records, and satellite-measured hemispherical records over a 2-year period. The sunshine records are incorporated into the database by use of Angstrom-type correlations developed for Zimbabwe, and the short-term satellite data are ‘cultured to long-term ground-measurement basis by means of an empirically derived correlation’ and a ‘time series factor’. Diffuse radiation values are generated from the resulting hemispherical radiation database by a locally developed correlation of the monthly average diffuse fraction of hemispherical radiation with monthly average clearness index. Normal beam radiation is computed from the hemispherical and diffuse radiation using two different methods. The two methods are found to agree generally to within 7%. The results are presented as country-border-contained isolines of radiation. The sensitivity of beam radiation to the accuracy of estimating diffuse radiation is inspected. The sensitivity is quite high, about 1:1 for months and locations with very low clearness. This translates to an almost equal sensitivity of insolation available to tracking tilted apertures, underscoring the importance of developing a local diffuse fraction-clearness index correlation rather than relying on a correlation developed elsewhere.

Introduction

In research problems attempting to verify the detailed performance of a particular solar energy system, instantaneous fluxes of hemispherical (also called global), diffuse and/or beam solar radiation may be required. However, for the purpose of engineering design and the basic assessment of the economics of solar energy systems, only the monthly averages of the above quantities are required. The objective of this paper is to present the distribution of the intensity of monthly average hemispherical, diffuse and normal beam solar radiation over Zimbabwe.

In Zimbabwe hemispherical radiation is measured by pyranometers at 20 stations monitored by the Meteorological Department (Met. Dept)1 These stations also record sunshine duration. Another nine stations record sunshine duration only. The sunshine data can be converted into hemispherical radiation data using Angstrom-type correlations, which were developed using sunshine and radiation data for nearby stations [1]. In addition, Raschke et al. reported values of satellite-measured hemispherical radiation for Africa, for the years 1985 and 1986, at latitude–longitude grid points with a spatial resolution of 2.5×2.5 degrees [2].

These three data sources, pyranometer-measured radiation records, sunshine duration records and satellite-measured radiation records, are blended, after some adjustments as described in section 2, to build a consolidated hemispherical radiation database.

In order to compute the insolation available to inclined planes, one needs, in addition to hemispherical radiation, the corresponding diffuse component. Diffuse radiation is recorded at only two locations in Zimbabwe. To determine the diffuse component of radiation when only the hemispherical radiation is known, the correlation of the ratio, d/h, of monthly average diffuse to hemispherical radiation, with monthly average clearness index, h must be determined.

Several correlations between these quantities have been developed in the literature using data for specific locations [3], [4], [5], [6]. However, the general validity of these correlations for applicability to locations situated very differently from the locations for which they were derived is not known. Therefore, in this study a region-specific correlation, based on the local data is developed, to be used in generating diffuse radiation values from hemispherical values for all locations in Zimbabwe. An important step in the development of this correlation is the appropriate correction of the measured diffuse radiation for the effect that by obscuring part of the sky dome from the pyranometer, the shade-ring cuts off not only the beam radiation but also a significant proportion of the diffuse radiation. Without local measurements of true diffuse radiation, the approach chosen here is to select from the literature a correction model which returns the highest correction factors, in order to make conservative estimates of available beam radiation.

The hemispherical radiation database is built in section 2 using all the available data. The development of the correlation of d/h with h as outlined above is the subject of section 3. In section 4, values of beam radiation at normal incidence, are calculated from the results of 2 Building the hemispherical radiation database, 3 Deriving monthly average diffuse radiation from hemispherical radiation. The results of the study are presented in section 5.

Section snippets

Building the hemispherical radiation database

For the determination of the monthly average hemispherical radiation, h, three data sources were used.

  • 1.

    Records of pyranometer-measurements of daily hemispherical radiation from the Department of Meteorological Services of Zimbabwe (elsewhere called the Met. Dept).

  • 2.

    Records of monthly average sunshine hours per day from the Met. Dept.

  • 3.

    Hemispherical radiation data derived from imaging data of the geostationary satellite METEOSAT2, for 1985 and 1986 [2].

Deriving monthly average diffuse radiation from hemispherical radiation

In Zimbabwe, hourly diffuse radiation is recorded at two stations, Harare and Bulawayo, using the common instrument set-up of a pyranometer shaded with a shadow band. In this study, only the Bulawayo records are used. The data sourced comprise monthly average hourly records of hemispherical and diffuse radiation for 13 years. The diffuse radiation data have already been corrected for the shade ring effect—the effect that the shade ring, by obstructing part of the sky dome from the pyranometer,

Estimating normal beam radiation

In this section beam radiation is calculated using two different methods and the results are compared.

Results

The results of this study are presented as a set of isoline maps of long-term average radiation. A sample of the maps is shown in Fig. 4, Fig. 5, Fig. 6, Fig. 7, Fig. 8, Fig. 9, Fig. 10, Fig. 11, Fig. 12. This gives a visual picture of the spatial distribution of radiation over Zimbabwe which can be conveniently read by potential users.

For the estimation of radiation available to tilted apertures which accept both beam and diffuse radiation the monthly averages of both hemispherical (global)

Conclusion

In this study, figures for annual normal beam radiation for Zimbabwe are obtained using all data on global and diffuse radiation presently available.

Two independent methods are applied to obtain monthly beam radiation averages from monthly global and diffuse radiation data. Results for monthly averages differ by typically less than 7%. Results for annual averages differ by less than 3% for all stations.

A sensitivity analysis reveals that uncertainties in monthly diffuse data may result in up to

Acknowledgements

The authors are indebted to Mr T.C. Temba from the Department of Meteorological Services of the Government of Zimbabwe who provided the raw data for this study.

References (15)

There are more references available in the full text version of this article.

Cited by (0)

View full text