Intake time as potential predictor of methane emissions from cattle.

Measuring enteric methane (CH 4 ) produced by cattle is central for the construction of national greenhouse gas inventories and for devising strategies to mitigate the environmental footprint of ruminant livestock. Efforts to estimate or measure enteric CH 4 emissions in cattle include techniques such as the Green-Feed system, laser detectors, respiration chambers, and the SF6 technique. However, using these tools on a large scale is complicated and expensive. Therefore, it is necessary to develop alternative methods to estimate CH 4 emissions from cattle. The objective of this work was to evaluate the potential use of intake time (IT) to estimate enteric CH 4 emissions in cattle. The CH 4 emissions of seven steers were measured in respiration chambers. A computer vision technique, called Object Detection (OD), was used to estimate IT. The images recorded from the top of the feeder were used for OD. Additionally, the dry matter intake (DMI) was measured by the difference between the offered and rejected food each day. CH 4 emissions were predicted by two linear equations using DMI or IT as explanatory variables. The equation using IT versus that one using DMI presented lower root mean square error (RMSE, 13.8 vs. 20.2 g/d), relative prediction error (RPE, 9.1% vs. 13.3%) and RMSE to standard deviation of observed values ratio (RSR, 0.38 vs. 0.56), respectively. Our results indicated that, under our experimental conditions, IT is a potential predictor of enteric CH 4 emissions in cattle. However, more experimental data are needed to validate our findings.