Biases in LI-COR Plant Canopy Analyzer estimates of seasonal light interception by black spruce and trembling aspen canopies

Abstract

This article examines four sources of bias in the estimation of seasonal light interception by black spruce (Picea mariana (Mill.) BSP) and trembling aspen (Populus tremuloides Michx.) canopies using the LI-COR LAI-2000 Plant Canopy Analyzer: (i) geometric averaging of gap fractions, (ii) unrepresentative elevation angle intervals for radiation detectors, (iii) assumption of isotropic radiance distribution, and (iv) inclusion of light intercepted by stems in the estimate of canopy light interception. Bias source i caused overestimates of canopy light interception, source ii caused underestimates, source iii caused little bias, and source iv caused overestimates. The magnitude of bias from sources i, ii, and iv increased as fractional light interception by canopies decreased. The average bias in seasonal light interception for all sources combined was 5% for black spruce canopies and 17% for trembling aspen canopies. It is recommended that canopy light interception and understorey light level estimates be calculated in a way to reduce bias from sources i and ii. For estimates of canopy interception alone, bias from source iv should also be reduced.