Publication Date: 2015.
Methane emission factor (Ym) is directly involved to calculate the worldwide livestock methane inventories, hence it is important to refine the estimation of this parameter for different livestock production systems. The purpose of this work was to generate refined mathematical models to predict CH4 emissions from an extensive compilated database at on-farm level and to compare them with different models already available in the literature. Methane emission predictive models (expressed as Ym, % gross energy intake; and methane production, CH4p, g an−1 d−1) where fitted taken into account the production system, the livestock type and the feed characteristics available at on-farm level within a reasonable uncertainty range. In order to develop the models, only easy available parameters were selected to fit new mathematical models. Hence, the full model included: ruminant types (beef cattle, dairy cattle, and sheep), fibre sources (fresh forage, conserved forage, and straw) and concentrate levels (DM basis) in the diet (Low, <35%; Intermediate, 35–65%; High, >65%). Full models were assessed by the Bayesian Information Criterion (BIC) and terms that did not reach significance level (P≤0.05) were dropped from the model. Furthermore, predicted results were assessed through correlation and regression analyses considering the model significance. Models developed in this study were compared by the degree of adjustment of a simple regression. Additive and technique terms were initially dropped from the full model used to predict Ym because they did not have effect in the prediction (P>0.10). Therefore, the final equation for Model 1 was: Ym(a)=Intercept−0.243(±0.051)×DMI (kg d−1)+5.9×10−3(±1.17×10−3)×NDF (g kg−1 DM−1)+5.7×10−3(±1.63×10−3)×DMD (g kg−1 MS−1) (BIC=559). All terms of this model, intercept factor (type of cattle×source of fibre×level of concentrate), DMI, NDF, and DMD were significant (P<0.0001). DMI was the term with the greatest weight in the model. The predicted Ym value decreased about 0.243 percentage units (P<0.0001) per each additional kg in DMI. When the equation was compared with previous publicated models, our model showed a satisfactory degree of fitting. In conclusion, this new model improved the estimation of the Ym factor from beef and dairy production systems, using different forage quality characteristics from on-farm level to increase precision.
EEA Santiago del Estero
Author affiliation: Jaurena, Gustavo. Universidad de Buenos Aires. Facultad de Agronomía; Argentina
Author affiliation: Cantet, Juan Manuel. Universidad de Buenos Aires. Facultad de Agronomía; Argentina
Author affiliation: Arroquy, Jose Ignacio. INTA. Estación Experimental Agropecuaria Santiago del Estero; Argentina. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina. Universidad Nacional de Santiago del Estero.Facultad de Agronomía y Agroindustrias; Argentina
Author affiliation: Palladino, Rafael Alejandro. Universidad de Buenos Aires. Facultad de Agronomía; Argentina. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina
Author affiliation: Wawrzkiewicz, Marisa. Universidad de Buenos Aires. Facultad de Agronomía; Argentina
Author affiliation: Colombatto, Dario. Universidad de Buenos Aires. Facultad de Agronomía; Argentina. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina
Repository: INTA Digital (INTA). Instituto Nacional de Tecnología Agropecuaria