Is the spherical leaf inclination angle distribution a valid assumption for temperate and boreal broadleaf tree species?

Angular distribution of leaves is an important parameter determining the transmission and reflection of radiation by vegetation canopies. When inverting canopy transmittance measurements for estimating the leaf area index or canopy clumping, incorrect assumptions on leaf angles may lead to considerable errors. To address this issue, we measured an extensive data set of leaf angle distributions for 58 deciduous broadleaf tree species commonly found in temperate and boreal ecoclimatic regions. Additionally, a small subset of species was sampled several times during the growing season. We found that reliable estimates of leaf angle distributions at the level of a whole tree can be obtained by measuring the leaf inclination angles of 75 leaves distributed across the vertical tree profile using a simple technique based on digital photography. According to our measurements, the common assumption of a spherical leaf angle distribution often used when no measurement data is available is not a valid assumption for most of the studied tree species. Our multitemporal measurements indicated no seasonality of leaf inclination angles of the selected tree species, except directly after budburst. Finally, using a gap fraction model, we illustrate the influence of different leaf angle distributions on the clumping factor and leaf area index. We advocate the use of species-specific leaf angle distributions from actual leaf inclination measurements. However, a planophile or a plagiophile leaf angle distribution appears to be a more appropriate assumption than a spherical one for modeling radiation transmission through temperate and boreal broadleaf stands when no leaf inclination angle measurements are available.