Challenges and advances in calculating light transmission through forest canopy over large extents

Abstract:

Shortwave radiation is one of the primary components of the sub-canopy energy budget, exhibiting spatial and temporal variabilities in snow surface shading at metre and minute scales in heterogeneous forest canopies. Canopy shortwave transmissivity can be successfully predicted at these high resolutions using hemispherical photography or high resolution laser scanning data, yet both data and computational requirements invoke challenges when modelling canopy transmission over spatial extents > 1km².

We present and compare a series of shortwave canopy transmission models requiring varying complexity of canopy structure input data, from airborne laser scanning to a simple canopy height model. All models are based upon the concept of synthetic hemispherical images, which can be used to simultaneously determine both the diffuse and time-varying direct shortwave radiation components. The most complex and computationally demanding model is capable of resolving individual trunk and branches within the canopy, however these complexities limit easy application of this model to areas < 1 km². The least complex model uses canopy height models to resolve individual tree crowns and 2D canopy heterogeneity, while generalising the lower canopy structure. This simpler model is able to be efficiently applied at metre-resolution scales over large spatial extents well beyond 1 km² while still enabling calculation of sub-canopy snow surface shading and sun-flecks within shadowed areas.

Comparison of the models show that at metre and minute scales, individual trunks and branches are an important aspect for accurate radiation transfer modelling. Averaging model output across coarse grid scales shows the simpler model has a similar performance compared to the more complex models, while still resolving canopy structure at high spatial resolution. These results demonstrate representation of individual tree crowns and 2D canopy heterogeneity is still important for accurately representing spatial distribution of sub-canopy shortwave radiation.