Comparison of direct and indirect methods for assessing leaf area index across a tropical rain forest landscape

Mean direct-harvest LAI above 1 m was 5.5 ± 0.3 SE. Plant area index (PAI, leaves + wood) was 5.1 ± 0.2 for the LAI-2000, and for the hemispherical photographs was 3.9 ± 0.2, analyzed using Gap Light Analyzer (GLA), and 4.9-6.0 ± 0.2 using WinSCANOPY software. Correction for leaf clumping (non-random distribution of leaves) generally improved LAI estimates of the hemispherical photographs. At the local scale, direct-harvest LAI was not significantly correlated with LAI estimates for either indirect method. However, correlations between direct-harvest LAI and both indirect methods along vertical canopy transects from forest floor to the canopy top were significant. Relationships between harvest LAI and canopy closure (from which indirect LAI values are derived) showed very small changes in closure with large changes in LAI at LAI values > 6, indicating that the estimations of LAI using canopy closure values were reaching an asymptote. As a result, at high canopy closure indirect LAI is underestimated. Overall, the LAI-2000 performed better than hemispherical photography for estimating direct-harvest LAI at landscapes scales. However, with corrections for leaf clumping, hemispherical photography can be effective for estimating and characterizing landscape level LAI of tropical rain forest.