Use of hemispherical piiotographs in forest ecology

Hemispherical photographs (hemiphots) taken from the ground looking up at the forest canopy provide a record of the canopy. They can be analysed to compute either relative or, more usefully, absolute measures of solar radiation for the sample point, taking account of the contributions of direct (beam or sun) and indirect (diffuse or sky) radiation. Hemiphots can be obtained more easily and cheaply than instrumental measurements especially in remote areas. After explaining the geometry and terminology of hemiphots and solar radiation this paper describes the methods of computing absolute amounts of radiation, dealing with direct and indirect radiation separately and summing for the total as a final step. The first method uses the product of measured radiation incident above the canopy and site factors computed from the hemiphot. Site factors are a relative measure of radiation defined as the fraction of incident radiation that reaches the ground. Direct and indirect site factors must each be calculated with appropriate weighting for the angular distribution of radiation. The second method is based on the known, constant amount of solar radiation available above the earth's atmosphere (the solar constant) and its attenuation by the atmosphere. Unfortunately, only direct radiation can be computed in this way. Neither method of computing solar radiation from hemiphots is completely independent of instrumental measurements or estimates of incident radiation for the site. The use of hemiphots (or instrumental measurements) for mapping radiation on the ground is discussed. Interpolation between sample points is a sound procedure only for long-term average radiation and where the canopy is homogeneous, having small holes uniformly distributed. In other circumstances there are unpredictable discontinuities in the amount of radiation which make mapping unreliable. Hemiphots are an invaluable tool in forest ecology but they require some instrumental measurements to be able to compute absolute amounts of radiation. Checklists of steps in computation and sources of error are given. Because of the relatively low precision of field measurements of radiation close agreement is not to be expected between computed and measured values. Attention is drawn to the assumptions that must be made in computing radiation from hemiphots as these should always be clearly stated, and tested where possible.