Characterizing turbulent plume dynamics of methane emissions in cattle: Field measurements and parametric modeling

This work examines the dispersion of enteric methane emissions within the first 2 meters of the animal’s muzzle to clarify the factors that reduce the accuracy of methane estimates from sniffer methods. We examined turbulence during respiration in low-tropic cattle and proposed models to estimate methane mass that incorporate the physics of gas dispersion. Methane concentration was measured using a laser device, which captured 97 emission events across 11 Girolando animals. Peak concentrations occurred between 10% and 23% of the event duration and declined exponentially, consistent with the dispersion relation. Evidence from two-dimensional spectral measurements indicated that the flow involves a two-stage mechanism: the initial jet, in which the flow is first released, and the interrupted jet that follows the cessation of the momentum source. The events displayed a time-averaged methane concentration of 76.4 ppm, an average event duration of 37.9 s, and turbulent fluctuations reaching 243% and 76% for the primary and secondary clusters, respectively. Turbulence analysis demonstrated that flow near the nostrils is anisotropic. A physics-based, time-invariant model was created to estimate daily methane mass per animal, incorporating a dilution correction. Although the predictions match previous observations, the estimate’s uncertainty highlights the challenges of obtaining accurate measurements with sniffer methods. Additionally, models based on lognormal, sinusoidal, and Gaussian probability distributions were developed to forecast methane concentration over time and to account for turbulent fluctuations. This study offers insights into cattle methane turbulence, clarifies sniffer estimate errors, and suggests corrections to improve their design and use. • CH 4 peaks within 10% of enteric emission, decaying according to the dispersion relation. • Emission plume shows starting (momentum-) and interrupted (buoyancy-driven) jet stages. • Time-invariant model incorporate first order features. • Lognormal, Sinusoid, and Gaussian models capture time-varying CH 4 concentration.