Improving robustness and accuracy of predicted daily methane emissions of dairy cows using milk mid‐infrared spectra

Abstract

BACKGROUND: A robust proxy for estimating methane (CH 4 ) emissions of individual dairy cows would be valuable especially for selective breeding. This study aimed to improve the robustness and accuracy of prediction models that estimate daily CH 4 emissions from milk Fourier transform mid-infrared (FT-MIR) spectra by (i) increasing the reference dataset and (ii) adjusting for routinely recorded phenotypic information. Prediction equations for CH 4 were developed using a combined dataset including daily CH 4 measurements (n = 1089; g d −1 ) collected using the SF 6 tracer technique (n = 513) and measurements using respiration chambers (RC, n = 576). Furthermore, in addition to the milk FT-MIR spectra, the variables of milk yield (MY) on the test day, parity (P) and breed (B) of cows were included in the regression analysis as explanatory variables.RESULTS: Models developed based on a combined RC and SF 6 dataset predicted the expected pattern in CH 4 values (in g d −1 ) during a lactation cycle, namely an increase during the first weeks after calving followed by a gradual decrease until the end of lactation. The model including MY, P and B information provided the best prediction results (cross-validation statistics: R 2 = 0.68 and standard error = 57 g CH 4 d −1 ). CONCLUSIONS: The models developed accounted for more of the observed variability in CH 4 emissions than previously developed models and thus were considered more robust. This approach is suitable for large-scale studies (e.g. animal genetic evaluation) where robustness is paramount for accurate predictions across a range of animal conditions.

Multiple Country Approach to Improve the Test-Day Prediction of Dairy Cows’ Dry Matter Intake

“…The predictive model’s accuracy and generalization are directly related to the training dataset’s quality and representativeness, with greater variability bringing robustness [ 17 ]. Combining data involving different breeds, diets, and coming from different countries, which already demonstrated scientific achievement while predicting methane [ 18 ], fatty acids [ 19 ], or lactoferrin contents [ 20 ], for instance, was in line with this generalized perspective. Using various datasets from different geolocated farms, we aimed to increase the calibration set variability comparatively to studies presented in Table 1 .…”

Quantification of methane emitted by ruminants: a review of methods

“…Moreover, milk MIR spectra can be obtained routinely at a reasonable cost (already collected for milk payment and/or milk recording). This proxy represents significant interest for large-scale studies (compare animals, herds, periods, geographical regions, and genetic studies) ( Vanlierde et al, 2020 ), but information about the limitation and applicability of milk MIR is lacking.…”

27. Genome-wide association study for mid-infrared methane predictions in Walloon dairy cows

“…All milk samples were analysed using Milkoscan FT4000, FT6000, and FT+ (Foss-Electric A/S, Hillerød, Denmark) by the milk laboratory Comité du Lait (Battice, Belgium) to generate the MIR spectral data. Methane emissions (PME, g/d) were predicted from the recorded spectra using the equations developed by Vanlierde et al (2021). To eliminate potential abnormal records, the PME values below the 0.1 percentile and above the 99.9 percentile were deleted (Kandel et al, 2017).…”