CLEAR-SKY RADIANCE DISTRIBUTIONS IN ULTRAVIOLET WAVELENGTH BANDS

Citation
Rh. Grant et al., CLEAR-SKY RADIANCE DISTRIBUTIONS IN ULTRAVIOLET WAVELENGTH BANDS, Theoretical and applied climatology, 56(3-4), 1997, pp. 123-135
Citations number
34
Categorie Soggetti
Metereology & Atmospheric Sciences
ISSN journal
0177798X
Volume
56
Issue
3-4
Year of publication
1997
Pages
123 - 135
Database
ISI
SICI code
0177-798X(1997)56:3-4<123:CRDIUW>2.0.ZU;2-C
Abstract
In the ultraviolet. the diffuse sky radiation component is often at le ast 50% of the global irradiance for middle latitudes. Thus, knowledge of the sky radiance distribution is important for modeling ultraviole t irradiance in vegetative or urban canopies. In this study, the distr ibution of clear sky radiance was measured and modeled for wavelength bands of ultraviolet-B (280-320 nm) and ultraviolet-A (320-400 nm). Sk y radiance measurements were made in a rural area over a wide range of solar zenith angles using radiance sensors mounted on a hand-operated hemispherical rotation mount. The measured sky radiance distribution in the UVA waveband differed greatly from that in the UVB waveband. Th e sky UVB radiance varied less across the sky hemisphere than the sky UVA radiance, in accordance with theory. A distinct region of minimum sky radiance was commonly found in the UVA waveband, but not the UVB w aveband. Nonlinear regression modeling of the sky radiance in the two wavebands showed that the isotropic sky was a poor descriptor of the U V radiance distribution. An anisotropic model of the clear sky UV radi ance distributions was developed that used the scattering angle betwee n the sun and the location in the sky and the sky zenith angle as pred ictor variables. The model coefficients of determination (r(2)) values were 0.93 and 0.69 for the UVA and UVB waveband models respectively. The model equations can be applied directly to the sky diffuse irradia nce on the horizontal to provide radiance distributions for the sky. I n addition, these distributions can be used to estimate correction fac tors for the shadowband method of diffuse irradiance estimation. Since the surface albedo of the measurement location was low and not explic itly used in the model equations, the distributions modeled may not ap ply to skies over snow cover or other high albedo surfaces.