PSXII-14 Fecal near infrared reflectance spectroscopy for indirect prediction of methane emissions from grazing cattle

Abstract

Abstract Methane (CH4) production from enteric fermentation of ruminants accounts for over a quarter of the greenhouse gas emissions from the agriculture sector in the U.S. Accurate and precise quantification of CH4 is essential to identify strategies for mitigating CH4 emissions. Numerous techniques and methodologies exist for measuring enteric CH4 emissions from ruminants, however, many were designed for specific research purposes, potentially limiting their applicability on commercial farms. Our objective was to evaluate the application of fecal near infrared reflectance spectroscopy (FNIRS) to provide inputs for indirect CH4 prediction calibrations. The Grazingland Animal Nutrition Laboratory (GAN Lab) has produced FNIRS predictions of grazing livestock diet quality since 1995 in conjunction with animal performance (weight/body condition change) generated by the Nutritional Balance Analyzer (Nutbal) Software package. FNIRS is a non-invasive approach that provides a significant opportunity to inform models for grazing cattle CH4 emissions. From a comprehensive U.S. dataset of approximately 30,000 samples, we selected a subset representing beef cattle grazing the gulf coast region of Texas from 2016 to 2019 (n = 85). Inputs of diet crude protein (CP), diet digestible organic matter (DOM), live weight (LW), and estimated live weight change (LWC) were used to predict feed intake (FI) employing a published equation and, from outputs generated by Nutbal, i.e. FI_Eq and FI_Nut, respectively. The FI (kg DM/animal.day) generated by both methods were applied to established prediction equations to generate preliminary estimates of CH4 emissions (CH4 FI_Eq and CH4 FI_Nut, respectively). Differences among estimations were assessed through ANOVA, with significance determined at P < 0.05. The estimated FI was different (P < 0.01) for both methodologies, resulting in different (P < 0.01) predictions of CH4 emissions. The CH4 prediction for CH4 FI_Eq was 0.24 ± 0.03 kg/animal.day, compared to 0.38± 0.06 kg/animal.day for CH4_Nut. A difference was observed between years for the CH4 predictions estimated with CH4 FI_Eq (P < 0.01). The FNIRS/Nutbal technique provided useful inputs to the established CH4 calibrations but will need to be validated against CH4 reference methods. The Gan Lab FNIRS/Nutbal dataset offers an opportunity to not only estimate U.S. grazing ruminant CH4 emissions over the past 30 years, but also predictions of current and future emissions at individual operation to national scales. This technique will enable producers with the knowledge to optimize forage resources management, mitigate CH4 emissions, and enhance the sustainability of grazing animal agriculture.