The computation of foliage clumping index using hemispherical photography

Hemispherical photography (HP) is extensively used for both canopy architecture such as leaf area index (LAI), and solar radiation regime determinations under forest canopies. This is done mainly by assuming that foliage elements occur in a spatially random manner. However, the majority of world forests occur in heterogeneous ecosystems and topography with rather complex canopy architecture. To improve the estimates of canopy structures and solar radiation regimes, a non-random spatial distribution parameter called clumping index (CI) has been used. We compared varying methodologies of CI determination on real HP acquired in contrasting forest types growing on sloping ground, and on simulated HP representing different aggregation levels of foliage elements. The major aim was the comparative analysis of the effects of forest types, forest density, slope and gap fraction acquisition accuracy on estimation of CI using the five different approaches. The result indicated that CI estimates based on gap size distribution approaches performed the best and were less affected by topography and forest density compared to approaches based on solely logarithmic gap averaging techniques.