|Göckede, M; Thomas, C; Markkanen, T; Mauder, M; Ruppert, J; Foken, T: Sensitivity of Lagrangian Stochastic footprints to turbulence statistics, Tellus, 59B, 577-586 (2007), doi:10.1111/j.1600-0889.2007.00275.x|
This study tests the sensitivity of a Lagrangian Stochastic footprint model to the turbulence statistics describing the flow field, with a focus on the within canopy processes. Representative profiles of the input velocity statistics are taken from a long-term dataset of turbulence measurements within and above a tall spruce canopy. Based on a wavelet tool which allows a detailed analysis of coherent structures along the vertical profile, we characterize several typical states of coupling and decoupling between surface, canopy and atmosphere. For each coupling regime, three flux footprints using different sources for turbulence statistics are compared: the first based on conditionally-averaged measurements, the second on a simple numerical solution, and the third on measurements taken from literature. The effects of profile smoothing and connecting measured canopy data to parameterized atmospheric surface layer profiles are considered. Significant differences between footprints based on modeled and measured profiles were found for exchange regimes with the lower section of the profiles decoupled from the atmospheric surface layer. As such cases are likely to occur for tall canopies with moderate density, our results suggest that the accuracy of Lagrangian Stochastic footprint modeling could be improved by using better turbulence profiles for different exchange regimes.