Sensitivity analysis of parameters, emission factors, and coefficients for estimating animal emissions of ruminant species in the Global Livestock Environmental Assessment Model (GLEAM)

Abstract

PurposeAnimal emissions account for nearly 60% of total greenhouse gas emissions from the livestock sector. To estimate these emissions, the Food and Agriculture Organization of the United Nations (FAO) developed a dedicated module within the Global Livestock Environmental Assessment Model (GLEAM). Although previous studies have explored selected inputs for specific animals and emission types, a comprehensive analysis of all 92 inputs (parameters and emission factors) had not been conducted. This study aimed to identify the most influential inputs affecting ruminant emissions in GLEAM.MethodsUsing global data from GLEAM to build representative samples, a one-at-a-time (OAT) sensitivity analysis was conducted by varying each input individually while holding the others constant. Parameters-specific ranges were defined, and sensitivity was assessed using regression coefficients for methane, nitrous oxide, and their sum as total emissions.ResultsSensitivity was determined for 70 of the 92 inputs, based on a high R2 between each input and the predicted emissions. Three parameters: gross energy of the diet, diet digestibility, and age at first calving, were the most influential with a negative correlation to animal emission, with diet digestibility emerging as the most sensitive. In contrast, parameters related to animal weight and two emissions factors: the methane producing capacity of manure (Bo) and urinary energy as a fraction of gross energy (UE), were the most influential with a positive correlation, mainly due to their impact on methane, which accounts for nearly 90% of total animal emissions. Nitrous oxide emissions were highly sensitive and positively correlated with the nitrogen content of the diet, while showing moderate sensitivity with a positive correlation to the emission factors for direct N2O emissions from manure (EF3), for nitrogen volatilization and redeposition (EF4) and for N2O from leaching/runoff (EF5). Regarding manure management systems, methane emissions were most affected and positively correlated with manure managed in liquid systems, while nitrous oxide emissions were most influenced with a positive correlation to manure managed as dry lot and deep litter. In contrast, changing manure management to compost, burned for fuel, or daily spreading showed the greatest potential to reduce animal emissions.ConclusionsThe study identified the most and least influential parameters and emission factors based on individual effects but did not evaluate interactions between them. The findings support prioritizing data quality improvements for the most influential inputs while using default values for less influential ones, helping to improve the accuracy and efficiency of livestock emission assessments.