CENTA NERC - Air flow through small heterogeneous woodlands and its role in CO2 exchange

Assessment of carbon exchange between large-size homogeneous forests (LHF) and the atmosphere largely relies on the FLUXNET approach[1] in which long-term eddy-covariance measurements from flux towers above forest canopy are used. This approach assumes spatial homogeneity with a large fetch and ignores many other factors (e.g. horizontal heterogeneity below the sensor level)[3]. Therefore, this approach cannot be applied to small-size heterogeneous woodlands (SHW) because previous studies showed that, depending on tree/leaf density, the length scale (from the edge) for a scalar to reach equilibrium can be tens of tree heights, much longer than the length scale for momentum[6]. SHW cover a significant proportion of land in the UK, Europe and and other parts of the world. The statistics, especially the intermittency, of wind and turbulence inside such woodlands play a crucial role of transferring CO2[2]; in addition, enhanced turbulence near the edge raise the exchange of CO2 there (Figure 1)[4]. In order to improve the assessment of CO2 exchange associated with SHW, we must enhance our understanding of the key player, turbulence, and its impact on CO2 transport for woodlands. Scientific questions are: (i) what are the 3D characteristics of wind and turbulence inside and outside the tree canopy? (ii) Can we model in-canopy mixing processes in order to improve the knowledge that is currently lacking? (iii) Compared with LHFs of the same tree species, can we express the additional carbon uptake in terms of some measure of the SHW shape (e.g., edge length, area)? (iv) What impact does accounting for SHW make to estimates of the global terrestrial carbon sink?
This project aims to address the scientific questions listed above and to shed light on the impact of SHW on the CO2 exchange with the atmosphere aloft. Particular focuses will be on the spatial varibility of CO2 exchange across a SHW and the enhancement of the exchange per unit area with reference to LHF. The project has the following innovative elements: (1) the first LES for turbulent flows inside a SHW; (2) the first use of a FACE facility to assesss the mixing capability of woodland.

LES will be adopted to reveal detailed in-canopy transport processes[4,5] that affect CO2 fluxes. To evaluate the model, turbulence data from 5 sonic anemometers inside and above the canopy of the BIFoR woodland and other data (met and CO2 concentrations) will be used. The BIFoR facility will also be used as a unique “field dispersion laboratory” where the CO2 amount required to maintain a concentration 150ppmv above ambient over several patches can be used to infer the mixing capability of the woodland.
Once evaluated, the model will be run for CO2 exchange scenarios (respectively for SHWs and LHF) by imposing various CO2 concentrations above the canopy, thus enabling an assessment of the CO2 exchange for SHWs and LHF. We seek functions to describe the difference between SHWs and LHF based on geometric and/or meteorologial parameters, and then assess the scaled-up estimates using land-cover maps and the new parameterisation for SHWs.

Keywords: woodland, forest canopy turbulence, CO2 exchange, computational fluid dynamics, CFD, large-eddy simulation, numerical modelling, BIFoR FACE