Solar radiation transmission through conifer canopies

Energy budget dynamics under forest canopies are strongly influenced by the large spatial variability of radiative and turbulent transfers in this environment. Incoming solar radiation under canopies has a particularly high degree of spatial variability. Transmission of solar radiation through a forest canopy varies with the size and location of the canopy gaps, as well as canopy leaf area. Modeling this transmission has proven challenging owing to the highly variable nature of the gaps within and between tree crowns, particularly in discontinuous canopies. This study describes and simulates the solar variability incident on the snow surface beneath two conifer forests. Objectives of this work are (1) to evaluate the variability of incoming solar radiation data with respect to canopy structure and cloudiness, (2) to correlate measured solar radiation transmission with predicted solar transmission based on analysis of hemispherical photographs, and (3) to examine the impact of measured and predicted transmission factors on the seasonal net radiative exchanges and snow ablation, based on snow process modeling. Observations were made during the winters of 2002 and 2003 in two predominately lodgepole pine (Pinus contorta) stands, one discontinuous and one relatively uniform, at the Local Scale Observation Site (LSOS) in Fraser, CO, USA, as part of the Cold Land Processes Experiment (CLPX). The canopy structure of all trees in the 0.8 ha triangular plot was measured and mapped in detail. We measured incoming global solar radiation at the snow surface beneath the discontinuous and the uniform canopies using arrays of 10 upward looking pyranometers at each site. Incoming global solar radiation was also measured above the canopy. Hemispherical photographs were taken with a Nikon Coolpix 995 with a fisheye converter at each radiometer location in both canopies, and were analyzed with Gap Light Analyzer (GLA) software. We found good agreement between measured and GLA-predicted transmissivities (r(2) = 0.86) when all data from both years were considered. Transmission factors derived from hemispheric photos and GLA software can be used to specify the distribution of solar flux under a canopy, instead of direct solar flux measurements, without degradation in snow model melt predictions. (C) 2004 Elsevier B.V. All rights reserved.