Leaf area index estimates obtained for clumped canopies using hemispherical photography

A method has been developed to improve estimates of leaf area index determined from hemispherical images of plant canopies, The approach used has been to modify existing techniques for analysis of gap fraction at a range of zenith angles. Conventional analytical techniques determine the gap fraction for a number of concentric annuli and then calculate the logarithm of this quantity. This approach requires the assumption that all foliage elements are randomly arranged within each annulus. The new approach divides each annulus into a number of small segments. The gap fraction of each segment is calculated and the average of their logarithms is calculated for each annulus (log-average or segmented method). These values are then used for the analysis of leaf area index and mean leaf angle using conventional techniques. Leaf area index estimates from hemispherical images were compared with data from a destructive harvest of a sparse canopy of Gliricidia sepium in Mexico. The log-average method was shown to significantly reduce the underestimation of leaf area index obtained from analysis of hemispherical images of clumped canopies. Conventional analysis of hemispherical images resulted in an underestimate of 50% compared to the destructive harvest. The segmented analysis reduced this to 15%. A comparison of the conventional and segmented analysis of hemispherical canopy images led to the development of a simple clumping factor calculated as the ratio of segmented and conventional estimates of leaf area index. This variation was found to vary between samples and location. For a canopy of Quercus pubescens in Scotland, the clumping factor was found to vary between 1.20 and 1.75, being highest under sparse canopies, This was attributed to the effects of tree spacing and the presence of a number of canopy gaps in the forest. Thus alternative methods that apply a uniform correct factor to conventional analysis may produce erroneous results where the scale of clumping is not uniform throughout the canopy. (C) 1999 Elsevier Science B.V, All rights reserved.