Browsing IJAFR, Volume 60, 2021 by Subject "grazing"
Now showing items 1-1 of 1
Exploring the potential of ingestive behaviour, body measurements, thermal imaging, heart rate and blood pressure to predict dry matter intake in grazing dairy cowsThe objective of this study was to develop and validate models to predict dry matter intake (DMI) of grazing dairy cows using animal energy sinks and status traits in combination with traits related to grazing behaviour, body measurements, thermal imaging, heart rate and blood pressure. The dataset used to develop the models comprised 33 measurements from 113 Holstein-Friesian dairy cows. Multivariable regression models were constructed incorporating each independent variable into a benchmark model incorporating the energy sinks (milk yield [MY], fat %, protein % and body weight [BW]) and status traits (feeding treatment, parity and calving day of year). Of the 33 variables tested, 10 showed an association with DMI (P < 0.25). These variables were incorporated into a backward linear regression model. Variables were retained in this model if P < 0.05. Grazing bout duration and rumination mastication rate were retained in the final model. The inclusion of these variables in the model increased DMI prediction by 0.01 (coefficient of determination [R2] = 0.85) compared to the benchmark model alone (R2 = 0.84). The models were applied to data recorded on an independent herd of 51 dairy cows. The R2 upon validation was 0.80 for the benchmark model and 0.79 for the model incorporating rumination mastication rate and grazing bout duration in combination with the benchmark variables. The separation of grazing bout duration and rumination mastication rate to predict DMI revealed rumination mastication rate slightly increases predictive accuracy upon external validation (R2 = 0.81), whereas grazing bout duration did not (R2 = 0.78). This suggests that grazing bout duration is not a robust trait for DMI prediction. Results from this study suggest that rumination mastication rate can slightly increase the accuracy of DMI prediction surpassing known energy sinks and status traits.