Unfolding the Link between Forest Canopy Structure and Flow Morphology

Both realistic surface boundary conditions and an accurate representation of unresolved processes are critical elements for reliable Numerical Weather Prediction (NWP) models. Traditionally, vegetated canopies are represented in NWP models by means of enhanced surface roughness and associated displacement height (J.R. Garratt, 1992; J.C. Wyngaard, 2010), making use of similarity theory, which has been shown to work well above the canopy (I.N. Harman and J.J. Finnigan, 2007) thank to the reduced resolution of standard NWP models. Operational configurations have the first vertical grid point above 20 m height, and the horizontal resolution is close to 3 km. However, with the progressive increase in computational power and the advent of new computing techniques, this resolution will increase by almost a factor ten (~O(100 m)), hence being able to partially resolve the heterogeneities in vegetated canopies. In this new scenario, similarity theory will not hold any longer (I.N. Harman and J.J. Finnigan, 2007); therefore there is a practical need for new canopy models that are capable of partially reproducing the heterogeneity of vegetated canopies.

The focus of this project is on canopy heterogeneity associated with canopy lacunarity with scales ~ O(100 m) associated with a uniform under-canopy roughness and dominated by weak thermal stratification. Such near canonical cases describe inhomogeneous momentum transport in an otherwise planar homogeneous flow when the canopy is absent. These canonical configurations serve as a logical starting point for the more complex cases, where heterogeneity affects additional aspects of the flow field. More specifically, in this work, we explore the use of dispersive fluxes as a means not only to quantify perturbations induced by canopy heterogeneity on the mean flow but also as an opportunity for developing new parameterizations. For this purpose, Large Eddy Simulations of the atmospheric boundary layer with varying geostrophic forcing will be used. A high-resolution representation of different vegetated canopies, with a changing degree of "gapiness" and heterogeneity, is included. Results will quantify the perturbations induced by the canopy heterogeneities on the mean flow an higher-order statistics. Furthermore, preliminary results of a new scaling relating the contribution of canopy-induced dispersive fluxes as a function of canopy heterogeneity will be presented.