Performance Comparison of the Prediction Models for Enteric Methane Emissions from Dairy Cattle

Area of Science:

• Environmental Science
• Animal Science
• Agricultural Science

Background:

• Enteric methane (CH4) emissions from dairy cattle are a major source of anthropogenic greenhouse gases.
• These emissions represent a significant energy loss for the animals, impacting animal productivity and farm economics.

Purpose of the Study:

• To critically evaluate the predictive accuracy of existing models for estimating enteric methane (CH4) emissions in dairy cattle.
• To identify the most reliable and robust model for quantifying CH4 emissions in this livestock sector.

Main Methods:

• A comprehensive database was constructed from 135 treatment means across 81 peer-reviewed studies.
• Forty existing dairy cattle CH4 prediction models were assessed using statistical metrics including RMSPE, CCC, RSR, and error decomposition (ECT, ER, ED).

Main Results:

• Model 38 exhibited the lowest RSR (0.71) but showed substantial prediction errors.
• Model 21, which utilizes dry matter intake (DMI) as a predictor, demonstrated superior overall performance with an RSR of 0.83 and CCC of 0.58.

Conclusions:

• Model 21 is recommended as the most robust option for estimating enteric methane (CH4) emissions from dairy cattle based on current data.
• Future research should focus on expanding the database and refining models for diverse dairy farming systems to enhance accuracy and applicability.