Abstract
The surface energy balance, i.e., the partitioning of the energy exchange between the Earth’s surface and the atmosphere, plays an essential role in determining the characteristics and the evolution of the atmospheric boundary layer. An accurate assessment of its components is, crucial for a variety of applications, including weather forecasting, climate modeling, and ecosystem monitoring. However, measurements of the surface energy balance terms are still affected by uncertainties. In particular, turbulent heat fluxes measured with the eddy-covariance technique generally do not balance the available energy. Several studies claim that the main reason for this imbalance is connected to the advection induced by secondary circulations, which can be present even over homogeneous surfaces under convective conditions but are more common over heterogeneous and complex terrain as a consequence of differential heating.
In our research project we aim to evaluate the uncertainties connected to the measurement of the surface energy balance at different sites in the Alpine environment, where processes related to the lack of closure are expected to be particularly significant. Measurements at sites located in different contexts allow the investigation of the relationship between the non-closure of the surface energy balance, the surface heterogeneity, and the consequent development of different types of local and mesoscale thermally driven circulations. The use of UAVs allows spatially distributed measurements around the eddy-covariance sites, which are crucial for estimating advection and understanding the horizontal and vertical variability of the flow field.
In preparation for the UAV field campaigns, Computational Fluid Dynamics (CFD) simulations and wind tunnel tests were carried out to optimize the UAV measurement configuration and to validate the accuracy of the mounted anemometers. CFD simulations were performed for a wide range of test conditions, varying wind speed (up 5 m/s), drone inclination (up to 45°), and propeller rotational speed (up to 7000 RPM), leading to a total of 288 possible combinations. From these, 95 representative cases were simulated based on fluid dynamic relevance, convergence behavior, and dimensional analysis. Complementary wind tunnel experiments were conducted at the WindShape facility in Switzerland, using a dedicated setup for UAV testing equipped with a 6-degree-of-freedom robotic arm to precisely control the drone orientation. A total of 36 experimental cases were carried out to validate the CFD results and to characterize the aerodynamic disturbances generated by the propellers.
These tests also helped identify the optimal placement of sonic anemometers on the UAV to minimize flow disturbances and vibration effects. Mounting sensors too close to the propellers introduces significant errors due to turbulence and recirculation, while placing them too far increases mechanical imbalance and reduces flight time. CFD and experimental results were thus combined to determine the most suitable locations for reliable flow measurements during flight operat