Evaluation and development of prediction models for enteric methane emissions from cattle in India.

Area of Science:

• Agricultural Science
• Environmental Science
• Animal Science

Background:

• Ruminants in India contribute significantly to global warming through enteric methane (EntCH4) emissions.
• Accurate EntCH4 emission estimates are vital for evaluating mitigation strategies, but direct measurement is impractical at scale.
• Existing EntCH4 prediction models often lack accuracy for the diverse Indian cattle population.

Purpose of the Study:

• To evaluate the accuracy of existing EntCH4 prediction models for Indian cattle.
• To develop a new, more accurate model for predicting EntCH4 emissions from both dairy and nondairy cattle in India.

Main Methods:

• Evaluated six existing EntCH4 prediction models (IPCC_DMI, IPCC_GEI, Ribeiro_DMI, Ribeiro_GEI, Patra_DMI, Patra_GEI) using independent datasets.
• Developed a novel empirical EntCH4 prediction model using a linear mixed-effects framework.
• Assessed model accuracy using relative prediction error (RPE) and mean bias error (MBE), with RPE <20% considered acceptable.

Main Results:

• None of the six evaluated models accurately predicted EntCH4 emissions for nondairy cattle (RPE <20%).
• The Ribeiro_DMI and Patra_DMI models showed RPEs of 22.6% and 22.9% for dairy cattle, respectively.
• The newly developed linear mixed-effects model (Alam's model) achieved a significantly lower RPE of 9.48% for both dairy and nondairy cattle.

Conclusions:

• Existing EntCH4 models are insufficient for predicting emissions from nondairy cattle in India.
• The new linear mixed-effects model offers superior accuracy for EntCH4 prediction in both dairy and nondairy Indian cattle.
• This model provides a valuable tool for assessing mitigation strategies and understanding India's EntCH4 emissions.