Evaluation of the net energy for lactation system and estimation of the energy requirements of dairy cows based on a comprehensive analysis of feeding trials.

Respiration experiments with high-yielding dairy cows in Northern Ireland have shown higher energy maintenance requirements than those used in the requirements standards of, e.g. France, UK, USA and Germany. Therefore, the current net energy for lactation (NEL) system of Germany was tested by comparing measured NEL intake with calculated NEL requirements based on a comprehensive dataset from feeding trials conducted at nine research institutions in Germany, Austria and Switzerland. The relationship between NEL requirements and NEL intake is described by the equation: $\begin{array}{cc}& NELrequirements\left(MJ/d\right)=26.6\left(\pm 0.4\right)+0.82\left(\pm 0.004\right)\cdot NELintake\left(MJ/d\right)\\ & withCoefficientofDetermination\left(R^2\right)=0.677,\\ & RootMeanSquareError\left(RMSE\right)=15.9MJNEL.\end{array}$ The equation indicates a systematic over-estimation of NEL requirements in the lower performance range and an under-estimation at higher energy intake levels. A multiple regression analysis was conducted by calculating metabolisable energy (ME) requirements [MJ/d] using metabolic body size (MBS) [kg^0.75], milk energy performance (LE) [MJ/d] and body weight change (BWC) [kg/d]: $\text{}\text{}\text{}\text{}\text{}\text{}\text{}\text{ME}\text{intake}\left(\text{MEI}\right)\left[\text{MJ}\right]\text{=0}.\text{651}\left(\pm \text{0}.\text{004}\right)\cdot \text{MBS+1}.\text{37}\left(\pm \text{0}.\text{006}\right)\cdot \text{LE}\text{+}\text{16}.\text{6}\left(\pm \text{0}.\text{31}\right)\cdot \text{BWC}$ $\text{with}{\text{R}}^{\text{2}}=0.\text{717},\text{RMSE=24}.\text{0}\text{MJ}.$ These results indicate that the energy maintenance requirements are markedly higher than presumed in the feed evaluation systems commonly in use but confirm the results from Northern Ireland (0.600-0.660 MJ ME/kg^0.75 MBS). ME efficiency for lactation is also higher (k_L = 1/1.37 = 0.73) than that used in the systems and is also similar to the results of Northern Ireland with 0.64-0.69. The energy contribution of BWC derived by this equation is 12.1 MJ/kg (16.6 · 0.73) and distinctly lower than that of 21-25 MJ/kg presumed by the feeding standards, e.g. in Germany. Further, maintenance requirements were linked to milk yield (energy corrected milk (ECM) [kg/d]), as is practiced in the standard Australian energy system: $\text{}\text{}\text{}\text{}\text{}\text{}\text{}\left(\text{MEI}\right)\left[\text{MJ}\right]\text{=0}.\text{640 }+\text{0}.\text{0070}\cdot \text{ ECM)}]\cdot \text{MBS+1}.\text{12)}\cdot \text{LE}\text{+}\text{16}.\text{7}\cdot \text{ BWC}$ $\text{with}{\text{R}}^{\text{2}}=0.\text{719},\text{RMSE=24}.\text{0}\text{MJ}.$ These results demonstrate that maintenance energy requirements are partly dependent on milk yield. A differentiated analysis by stage of lactation showed that the regressions coefficients for MBS, LE and BWC change with lactation month; however, these findings apply especially to the first lactation months (i.e. in phases of intensive mobilisation).