Dairy cattle

Genetic parameters of body weight traits from a 3D camera system in Nordic Holstein and Red Dairy Cattle

---

Abstract

Body weight (BW) is increasingly used in genetic evaluations because it contributes to the feed-efficiency complex and serves as an indicator for functional traits, such as energy balance and fertility. Frequent and accurate weighing is difficult to obtain routinely in large commercial herds, but three-dimensional (3D) imaging enables high-frequency, noninvasive estimation of BW from camera images. Our objective was to estimate genetic parameters for BW and BW-change traits in Nordic Holstein and Red Dairy Cattle (RDC) using phenotypes from the 3D camera-based CowFIT system (Microsoft Xbox One Kinect V2). Body weight was predicted using a partial least squares regression model, and validated with a scale-measured reference dataset (accuracy = 0.94). We analyzed 7,511 complete lactations recorded from 2019 to 2025 in 11 Holstein and 8 RDC herds. Traits included calving weight (BWcalv), average BW during the first 305 d in milk (BWavg), early BW change (ΔBWearly), and later BW change (ΔBWlate). Daily BW time series were smoothed using LOESS in R, fitted per cow and lactation (parities 1-4), and variance components were estimated separately by breed using animal models in DMU. Descriptively, BWcalv and BWavg increased with parity (Holstein ≈ 600-746 kg; RDC ≈ 585-713 kg). Cows reached the lowest BW later in lactation with higher parity numbers (Holstein 49 to 92 d; RDC 48 to 78 d), and ΔBWearly became more negative with parity. Estimated heritabilities were moderate to high for BW-level traits (0.42-0.60) and low to moderate for BW-change traits (0.13-0.33). BWcalv and BWavg were strongly phenotypically ( >0.75) and genetically ( >0.85) correlated. BWcalv and ΔBWearly were negatively associated, with moderate phenotypic correlations (−0.45 to −0.30) and genetic correlations ranging from 0 in RDC to about −0.20 in Holstein, indicating that heavier cows tended to lose more BW early in lactation primarily at the phenotypic level. Notably, ΔBWearly and ΔBWlate were negatively genetic correlated (−0.30 to −0.50), suggesting different genetic mechanisms underlying BW mobilization and deposition phases during lactation. BWavg and ΔBWlate were favorably correlated both phenotypically and genetically (0.15-0.25), whereas BWcalv and ΔBWlate were weakly correlated. In commercial herds, 3D camera-estimated BW provides biologically meaningful, heritable information on BW level and its change during lactation, consistent with traditionally measured BW traits. Incorporation of these phenotypes into routine evaluations could improve selection accuracy for feed efficiency and functional traits related to energy balance and fertility.

Keywords: 2026

How to Cite:

Bengtsson, C., Kupisiewicz, K., Stephansen, R. & Lassen, J., (2026) “Genetic parameters of body weight traits from a 3D camera system in Nordic Holstein and Red Dairy Cattle”, World Congress on Genetics Applied to Livestock Production Digital Archive 2026(1): 2281847. doi: https://doi.org/10.31274/wcgalp.23458

Rights: 1