Improved prediction by enteric methane emission models in ruminant production systems by integrating climate...

Area of Science:

• Agricultural Science
• Animal Science
• Environmental Science

Background:

• Mathematical models exist to predict enteric methane (CH4) production in ruminants.
• These models are often developed from data from specific regions and production systems, limiting their universal applicability.
• Evaluating existing models in diverse South American production systems is crucial for accurate methane emission estimations.

Purpose of the Study:

• To assess the predictive performance of existing mathematical models for enteric CH4 emissions in ruminants.
• To evaluate model applicability across different production systems, climate zones (Köppen classification), and diet compositions (NDF, EE, starch, digestibility) in South America.
• To identify the best-performing models and provide recommendations for improving CH4 emission predictions.

Main Methods:

• Selected existing mathematical models were evaluated for predicting enteric CH4 emissions (g/d) in cattle (dairy, beef) and sheep.
• Models were assessed using data from diverse production systems in Argentina, Brazil, Chile, and Uruguay.
• Performance was ranked using the root mean square prediction error (RMSPE) and the ratio of RMSPE to the standard deviation of observed values (RSR).
• Climate and diet classifications (high/low NDF, EE, starch, NDF digestibility) were applied to evaluate model performance under specific conditions.

Main Results:

• Model performance varied significantly across climate zones, animal categories, and diet types.
• In temperate, hot summer climates, all models performed poorly for dairy cattle, but some performed well for high-NDF diets.
• For beef cattle in high-NDF diets, the Yan et al. (2009) model was best (RSR=0.85). For sheep, Congio et al. (2022a) performed best (RSR=0.63), with Belanche et al. (2023) excelling for high-NDF diets.
• In temperate, warm summer (Cfb) climates, Mills et al. (2003) models performed best for dairy cattle (RSR=0.78), and van Lingen et al. (2019) for beef cattle (RSR=0.91).
• Models incorporating NDF intake (NDFI) and ether extract (EE) were best for low-EE diets in dairy cattle.
• All models performed poorly for sheep in Cfb climates and for beef cattle in tropical savanna climates.

Conclusions:

• Existing models often show poor predictive accuracy for enteric CH4 emissions in South American ruminant production systems.
• Model performance is highly dependent on climate zone, ruminant category, feed intake, and detailed dietary composition.
• Future model development should integrate these factors, including potential mitigation strategies and feed digestibility, for improved regional accuracy.