Direct Air Emission Measurements from Livestock Pastures Using an Unmanned Aerial Vehicle-Based Air Sampling System

Quantifying air emissions from livestock pastures remains challenging due to spatial variability and temporal fluctuations in emissions due to weather conditions. In this study we used a small unmanned aerial vehicle (sUAV) equipped with real-time sensors and an air sample collection system to directly measure carbon dioxide (CO2), methane (CH4), ammonia (NH3), nitrous oxide (N2O), nitrogen dioxide (NO2), hydrogen sulfide (H2S), total volatile organic compound (VOC), and particulate matter (PM1, PM2.5, PM10) emissions across two dairy pastures, two beef pastures, and one sheep pasture in Wisconsin. Emission rates were calculated using the Lagrangian mass balance model and validated against ground-level dynamic flux chamber (DFC) measurements. UAV-based CO2 concentrations showed a strong correlation with DFC measurements (R2 = 0.86, RMSE = 21.5 ppm, MBE = +9.7 ppm). Dairy 1 yielded the highest emissions for most compounds, with average emission rates of 0.50 ± 0.28 g m−2 day−1 head−1 for CO2, 8.48 ± 2.75 mg m−2 day−1 head−1 for CH4, and 0.20 ± 0.60 mg m−2 day−1 head−1 for NH3. The sheep pasture, on the other hand, had the lowest CH4 and NH3 emission rates, averaging 0.35 ± 0.22 mg m−2 day−1 head−1 and 0.02 ± 0.05 mg m−2 day−1 head−1, respectively. Rainfall events (≥ 5 mm within five days of sampling) significantly elevated N2O emissions (0.56 ± 0.40 vs. 0.13 ± 0.17 mg m−2 day−1 head−1). Particulate matter emissions were significantly affected by forage density. PM2.5 emission rates reached 1.25 × 10−4 g m−2 day−1 head−1 under low vegetative cover. It was concluded that emissions were affected by both animal species and the environmental conditions. The findings of this study provide a foundation for further development of emission inventories for pasture-based livestock production systems.