PREDICTION MODELS FOR GROSS TEED EFFICIENCY FOR TMR SYSTEMS

Feed efficiency (FE) has major implications for dairy production profitability and indirectly for environmental sustainability. It thus has become a priority for monitoring the economic viability of milk production and the environmental footprint. As a result, extensive efforts are being made worldwide to include FE in dairy cattle breeding objectives, which include genome-wide association analyses with large data sets of international dairy FE research consortium data. This will then aim e.g. to implement marker-assisted selection. 

All FE traits such as gross feed efficiency and residual feed intake require the measurement of dry matter intake (DMI). Dry matter intake is a key variable in the calculation of an individual cow’s feed efficiency. However, direct measurement of individual animal DMI is expensive and rarely possible, and therefore has been a major factor hindering the inclusion of FE traits into selection objectives. Indirectly though. milk production traits and live weight (LW), which are cheap and easy to measure, could potentially be used as reliable predictors of DMI, since they can satisfactorily account for the amount of feed required for production and maintenance. It can be achieved through models to predict DMI and gross feed efficiency (GFE, kg energy-corrected milk/kg DMI). Limited efforts have, however, been made to explore the possibility of developing such prediction models in dairy cattle in South Africa, despite indications that DMI can be predicted reliably from these easy-to-measure traits, and if this is achieved, models for predicting GFE can be developed through stepwise regression, by identifying traits that significantly account for variation in GFE.  

The study cited was therefore conducted to investigate the correlations of milk production traits and LW with DMI and GFE, and to subsequently develop the most suitable prediction models for daily DMI and GFE using milk production traits and live weight in both first-parity and multi-parity Holstein cows fed appropriate total mixed ration (TMR) diets.  

The data consisted of 30 daily measurements of DMI, milk yield (MY), energy-corrected milk (ECM), butterfat yield (BFY), protein yield (PROY), lactose yield (LACY), butterfat percent (BFP), protein percent (PROP), lactose percent (LACP), and 25 daily LW records of a group of 100 first-parity Holstein cows, fed a total mixed ration. Similar measurements were collected from a group of 110 multi-parity Holstein cows, in lactations 2 to 6. Gross feed efficiency was calculated as kg ECM divided by kg DMI. Forward stepwise regression analyses were performed to develop the models, using the PROC REG procedure of the Statistical Analysis System (SAS) software.

The dataset for first-parity cows averaged 572 ± 15.6 kg LW, 21.9 ± 2.58 kg/day DMI, 34.3 ± 4.29 kg/day MY; 28.5 ± 3.91 kg/day ECM, 1.32 ± 0.22 kg/kg GFE, 2.81 ± 0.35 % BFP, 2.96 ± 0.23 % PROP, 4.91 ± 0.37 % LACP, 0.95 ± 0.14 kg/day BFY, 1.01 ± 0.16 kg/day PROY, and 1.68 ± 0.24 kg/day LACY, and the dataset for multi-parity cows  681 ± 20.6 kg LW, 26.2 ± 1.80 kg/day DMI, 44.8 ± 4.01 kg/day MY; 40.6 ± 3.82 kg/day ECM, 1.55 ± 0.16 kg/kg GFE, 3.32 ± 0.24 % BFP, 3.05 ± 0.10 % PROP, 4.94 ± 0.05 % LACP, 1.49 ± 0.16 kg/day BFY, 1.37 ± 0.14 kg/day PROY, and 2.21 ± 0.19 kg/day LACY. Noteworthy was the observation that although LW was highly significantly correlated with DMI, it was not significantly correlated with GFE. The highest correlations with GFE were ECM (respectively 0.70 for first-parity cows and 0.71 for multi-parity cows) and BFY (respectively 0.83 for first-parity cows and 0.82 for multi-parity cows). The best prediction models for DMI and GFE as developed by stepwise regression in first-parity cows were:

DMI (kg/day) = -54.2 -0.192 ECM (kg/day) + 0.146 LW (kg)        R² = 0.79; RMSE = 1.05 kg/day       

GFE (kg/kg) = 1.88 + 1.34 BFY (kg/day) – 0.003 LW (kg)       R² = 0.91; RMSE = 0.05 kg/kg

In the model developed for GFE for multi-parity cows, only BFY was significant. None of the other predictor traits made significant contributions. The model was:

GFE (kg/kg) = 0.41 + 0.76 BFY (kg/day)        R² = 0.80; RMSE = 0.09 kg/kg

Conclusions and Recommendations: Dry matter intake can be predicted satisfactorily from milk component analyses and live weight for Holstein cows on a TMR production system. Live weight is however not correlated with gross feed efficiency, in fact may be slightly negatively correlated which suggests that the high maintenance requirements of the large cow take away energy from milk synthesis. Although milk yield is a primary determinant of gross feed efficiency, the overwhelming effect of butterfat yield shows that in selection for gross feed efficiency, both milk yield and butter fat percentage should be in the selection index.

 The results of this study partially corroborates results of a large international study based on genomic data which strongly suggest that selecting smaller sized Holstein cows with lower dry matter intake, but higher milk, fat, and protein production, tends to increase gross feed efficiency and profitability. Apart from an individual farmer perspective, from a breed point of view, national indexes should therefore prioritize reducing body size of cows for improved feed efficiency and profitability. This should also benefit the quest to reduce greenhouse gas emissions and mitigate the effects of a warmer climate.