Evaluating a Flux Footprint Model Using Tracer Release Experiments and Tall Tower Eddy Covariance Measurements

In complex and heterogeneous landscapes, determining the spatial origin of measured fluxes is critical for interpreting eddy covariance (EC) data accurately. To address this, footprint models are used to simulate the transport of turbulence and quantify the contribution of different areas within the source region. These models rely on theoretical assumptions, such as homogeneous and stationary atmospheric conditions, which often deviate significantly from real-world conditions particularly in terrains with uneven topography or land cover. This discrepancy may lead to substantial uncertainties, as the models may fail to accurately represent the true flux contributions under these non-ideal conditions.

To evaluate the reliability of the Flux Footprint Prediction (FFP) model (Kljun et al., 2015) and its performance under real-world conditions, we conducted three tracer release campaigns in the upwind region of a tall tower EC greenhouse gas observation system located at Hove (55.7169°N, 12.2375°E), a rural area west of Copenhagen, Denmark. The experiments utilized acetylene (C₂H₂) as the tracer gas, released at a controlled and precisely known emission rate.  The FFP model were assessed using data from different averaging intervals, enabling a detailed comparison of temporal resolutions and their impact on flux estimates.

The observed fluxes were systematically compared with the model predictions, allowing us to identify discrepancies and provide critical insights into the strengths and limitations of the FFP model, particularly in rural and heterogeneous landscapes. Moreover, the analysis highlights the influence of averaging intervals on the agreement between measured and modelled fluxes. This work also provides a reference for applying tracer release experiments in heterogeneous terrain using the tall tower EC system, contributing to the understanding of experimental design and model validation in such environments.