An implementation of the advanced footprint analysis for UAV-based BVOC measurements

Biogenic volatile organic compounds (BVOCs) are emitted into the atmosphere by plants and other living organisms and play a significant role in various plant functions, such as growth, reproduction, and defense. BVOCs are also an essential part of many chemical reactions in the atmosphere and contribute to the formation of ozone and secondary organic aerosols and affect the radiation balance.

Investigating the atmospheric vertical profile concentrations of BVOCs has become an important focus for understanding these processes. There are various methods that can be used to study the atmospheric vertical profile, including towers, balloons, aircraft, and unmanned aerial vehicles (UAVs). Among these methods, UAVs offer greater flexibility for local air sampling by hovering over a target area and can reach altitudes of up to 1000m, making them ideal for permanent BVOCs monitoring that requires repeated measurements in a specific spatial and temporal domain. However, the source contribution area of the obtained BVOCs concentrations is often not identified, potentially leading to inaccurate conclusions about the exchange between the surface, vegetation, and atmosphere above the target area.

In this study, the vertical profile of BVOCs concentrations was obtained at SMEAR Estonia (Station for Measuring Ecosystem Atmosphere Relations) to analyze the composition and distribution of these compounds in the near-surface layer. The vertical samples were collected in 2020-2021 using a commercially available pump equipped with cartridges filled with adsorbents and mounted on a UAV. The UAV was used to collect samples from heights between 0 m and 90 m.

To address the issue of space-time representativeness related to the source signal, a footprint analysis was conducted. Micrometeorological data for four target areas were obtained from three SMEAR Estonia flux towers. The Flux Footprint Prediction model (Kljun et al., 2015) was used for the footprint calculation. We determined temporal and spatial changes in roughness length (z0) and zero-displacement height (zd) for each day when BVOCs measurements were carried out using meteorological and geospatial data on land cover types and corresponding canopy heights. Due to the presence of surface heterogeneity, z0 and zd varied significantly for each wind sector. Therefore, we ran the spin-up of the FFP model with updated input parameters at each step. For the measurement results interpretation, we also evaluated the representativeness of the obtained footprints over the target areas in the space-time domain and analysed the land cover composition and vegetation characteristics.

In this work, we show how the source contribution area of BVOCs concentrations can vary in size and shape depending on atmospheric conditions, and spatial and temporal variation and thus have an effect on the obtained species composition of BVOCs.  Based on the presented findings we discuss the potential implementation of this approach for similar research and its future development.