Greenhouse Gas Emissions: Estimation, Uncertainty and Sensitivity Analyses in a Cow-Calf Operation.

Mathematical models have long been used to estimate emissions from main greenhouse gas (GHG) sources, determine mitigation strategies and comply with market requirements in different scales. The objectives of this project were to (i) estimate the carbon footprint of a cow-calf operation in Florida, (ii) access the uncertainty in the results and (iii) perform a sensitivity analysis of the model in order to prioritize future research. The Intergovernmental Panel on Climate Change (IPCC) methodology was used to estimate GHG emissions with parameter values for the U.S. and/or for Florida, when available. Activity data was obtained from Buck Island Ranch. Mean and standard deviation for the 45 parameters analyzed were obtained from literature. Uncertainty analysis was performed generating random parameter values simultaneously in normal distribution. Morris sensitivity analysis methodology was used considering all of the parameters, and Fourier Amplitude Sensitivity Test (FAST) performed using parameters responsible for 95% of direct effect found by Morris. Results show that enteric fermentation is the main source of GHG, followed by animal waste. The most important GHG emitted is methane (CH₄) from enteric fermentation and animal waste. Nitrous oxide (N₂O) is second, from animal waste and N fertilizers. Uncertainty analysis indicates a large variability in results obtained, particularly in total GHG, enteric fermentation and animal waste emissions estimation. There is a relatively small number of parameters that greatly influences total estimated emissions, mostly related to N₂O and CH₄ emissions: emission factor for N₂O from dung and urine deposited on pasture, N excretion from cows, feed digestible energy, emission factors for N₂O volatilization from N fertilizers and N leaching and runoff, methane from soils and wetlands, methane conversion rate, N₂O and CH₄ global warming potentials. A new project is underway to provide site-specific values for pastures with and without grazing for several of these parameters.