98 Utilizing gas Flux from an Automated Head Chamber System to Estimate Dietary Energy Values in Cattle Fed a Finishing Diet

Abstract

Abstract Automated head chamber systems (AHCS; GreenFeed, C-Lock Inc., Rapid City, SD) provide estimates of daily flux of methane (CH4), carbon dioxide (CO2), and oxygen (O2) from free roaming cattle. This provides the opportunity to conduct energetic research from cattle in their production environment. One opportunity is to evaluate agreement between performance-estimated (P) net energy [Mcal×(kg DMI)-1] for maintenance (NEm) and gain (NEg; pNEm and pNEg, respectively) and gas-estimated (G) NEm and NEg (gNEm and gNEg, respectively). To assess this, yearling steers (n = 54; initial BW = 484.1 ± 26.0 kg) were assigned to one of two pens containing an AHCS and feed bunks designed to measure individual feed intake (Calan Gate; American Calan, Northwood, NH). Steers were transitioned to one of three finishing diets and were then fed for 80 days. Paired day initial and final unshrunk BW were measured on day 0 and 1, and on day 79 and 80, respectively. Following the feeding period, steers were transported to a commercial abattoir where carcass data were collected. Average daily gas flux was utilized to calculate heat production (HP), ignoring the energy losses from urinary nitrogen. Empty body weight (EBW) and empty body gain (EBG) were calculated and used to estimate retained energy (RE). The estimates of RE and HP were used to calculate ME intake and DMI was used to quantify dietary ME content. Through previously established cubic models, gNEm and gNEg were calculated. Ultimately, gNEm and gNEg were estimated from gas flux, DMI, ADG, and initial and final BW. A previously established quadratic model was used to estimate pNEm and pNEg using calculated energy requirements and available carcass parameters. Statistical analyses to compare the P and G energy values were conducted utilizing R (v.4.1.0) to assess precision (Pearson Correlation; r), method agreement (Lin’s Concordance Coefficient Correlation; CCC), Root Mean Square Error of Prediction (RMSEP), and Relative Prediction Error (RPE; RMSEP as a % of average performance estimated values). On average, P provided greater estimates of NEm [0.074 Mcal×(kg DMI)-1] and NEg [0.070 Mcal×(kg DMI)-1] relative to G. The G and P provided values which were highly correlated and had excellent agreement (r = 0.92, P < 0.01; CCC ≥ 0.87) for both NEm and NEg. Additionally, predictive capability was high for NEm (RMSEP = 0.12; RPE = 5.3%) and NEg (RMSEP = 0.11; RPE = 6.8%) between P and G values. The excellent agreement between P and G derived NEm and NEg values suggest that researchers can utilize AHCS to conduct energetic studies with cattle in their production environment. Moreover, the G methodology provides an alternative method for retrospective assessment of animal energetic efficiency.