The repeatability and heritability of traits derived from accelerometer sensors associated with grazing and rumination time in an extensive sheep farming system

Introduction The automated collection of phenotypic measurements in livestock is interest to both researchers and farmers. Real-time, low-cost, accurate phenotyping can enhance precision management could lead the optimized utilization pasture breeding efficient animals. Wearable sensors provide tools for develop novel phenotypes across all production systems, which especially valuable grazing conditions. objectives this study were estimate repeatability heritability traits related rumination activities their correlations with other traits. Methods This was conducted on a commercial Merino farm west Victoria, Australia, from 4 May 2020 29 2020. A total 160 ActiGraph embedded halters attached left side muzzles sheep (M = 74, F 86) aged 10–11 months while pasture. Support vector machine (SVM) algorithms classified sensor output into categories grazing, rumination, walking, idle, activities. These further daily time (GT), number events (NGE), length (GL), (RT), (NRE), (RL), walking (WT), idle (IT). data analyzed using univariate bivariate models ASReml-SA repeatability, heritability, among Results GT estimated be 0.44 ± 0.23, whereas had estimates close zero. moderate high, highest being (0.70 0.03) lowest RT (0.44 0.03). intraclass correlation or at 1-day interval (i.e., 2 consecutive days) high traits, steadily reduced when between longer than 1 week. Discussion showed that wearable SVM methods are reliable recording pasture, have potential application ranking animals selective breeding.