Radiative transfer and carbon assimilation in relation to canopy architecture, foliage area distribution and clumping in a mature temperate rainforest canopy in New Zealand

The objective of this study was to characterize the forest canopy in terms of species composition, tree size classes and spatial distribution of leaf area of a mature temperate rainforest to provide a basis for scaling up shoot-level processes of energy and mass exchange using a one-dimensional canopy model. Stem number per unit area was high (1660 stems ha−1) but species diversity was low (n = 10). Two species (Dacrydium cuppressinum and Weinmannia racemosa) equally accounted for 70% of the stem numbers, with the remaining 30% distributed among the other eight tree species. However, D. cuppressinum alone accounted for 72% of basal area and 75% of crown volume. Across the site Le ("effective" leaf area index, uncorrected for foliage clumping and woody surface interceptance) varied between 2.2 and 5.6, with a mean value of 3.5. Half total woody surface area per unit ground area (W) was 1.1. The mean foliage clumping index within shoot elements (γE) was 1.2, while that between shoot elements (ΩE) was 0.87, giving an overall foliage clumping factor (Ω = ΩE/γE) of 0.73. After corrections for foliage clumping and woody surface interceptance, half total leaf area per unit ground area (Lhc = Le/Ω − W) was 3.7. For individual shoots, γE increased from 1.0 to 1.5 with height in the canopy, possibly reflecting shoot structural adaptation to local irradiance. Simulations with the canopy model over 1 year using mean clumping factors revealed that clumping reduced total canopy radiation interception by 5%. Shoot-level irradiance was reduced at the top of the canopy but increased in lower canopy layers resulting in a more even canopy irradiance distribution. This enhanced annual canopy photosynthesis by 8% over a canopy with randomly distributed foliage. When the simulations were repeated to include the observed variation of γE with height, total canopy radiation absorption was 11% lower than for a random canopy and canopy photosynthesis was 12% greater than a random canopy. The combination of reduced canopy radiation absorption and increased photosynthesis due to foliage clumping resulted in a considerable enhancement of canopy light-use efficiency. These analyses reveal a significant advantage of clumped foliage over randomly distributed foliage in terms of carbon gain, even for a forest canopy with moderate leaf area index.