The aim of this experiment was to quantify the milk production capacity of cows undergoing extended lactations while fed a pasture-based diet typical of those used in the seasonal-calving dairying systems of Victoria, Australia. One hundred twenty-five Holstein cows were randomly assigned to 1 of 5 groups. Breeding was progressively delayed after calving to enable management of the groups for lactation lengths of 10, 13, 16, 19, and 22 mo (equivalent to calving intervals of 12 to 24 mo). Cows were provided with a daily energy intake of at least 180 MJ of metabolizable energy/cow. This was supplied primarily by grazed pasture with supplementary cereal grain, pasture silage, and hay. Cows were dried off when milk volume fell below 30 kg/wk or when they reached 56 d before their expected calving date. Most cows (>96%) could lactate above this threshold for 16 mo, >80% for 19 mo, and >40% for 22 mo. There were negative relationships between lactation length and annual production of milk and milk solids (milk fat + protein), but losses were small until 16 mo. Annualized yields of milk solids were 497, 498, 495, 474, and 463 kg/cow for the 10, 13, 16, 19, and 22 mo groups, respectively. This reduction in annual production of milk solids with increasing lactation length was relatively less than for milk volume because during extended lactation, cows produced milk with higher concentrations of protein. Cows undergoing extended lactations also finished their lactations having gained more body weight and body condition than cows lactating for only 10 mo. The data showed that many cows on pasture-based diets were capable of lactating longer than the 10 mo that is standard for Victorian herds with seasonally concentrated calving patterns. Further, such extended lactations could be achieved with little penalty in terms of annual milk solids production.

The aim of this study was to measure the effect of type of diet and level of energy intake on the performance of cows undergoing extended lactations. Ninety-six Holstein-Friesian cows that calved in July and August 2004 were assigned randomly to 1 of 8 groups each of 12 cows (including 4 primiparous cows). Two of the 8 groups were assigned to each of 4 treatments that varied in lactation length (300 or 670 d) and diet (3 diets: control, high, or full total mixed ration (TMR). The 4 treatments were 1) control 300: cows were managed for a 300-d lactation and grazed pasture supplemented with grain and forage to provide a minimum daily dietary intake of 160 MJ of ME/cow; 2) control 670: as for control 300 except that cows were managed for a 670-d lactation; 3) high 670: cows were managed for a 670-d lactation and pasture was supplemented with grain and forage to provide a minimum daily dietary intake of 180 MJ of ME/cow; 4) full TMR 670: cows were managed for a TMR system that included a high body condition score at calving with cows offered a TMR during a 670-d lactation. The TMR was initially offered ad libitum indoors until about 440 DIM when the amount of TMR offered was reduced by about 2 kg of DM/d to prevent excessive weight gain. The proportions of cows still milking at the end of a 670-d lactation were similar for the control and high dietary groups. The full TMR group had fewer cows milking at 600 DIM: 17 cows milking compared with 24 cows in the control 670 group and 22 cows in the high 670 group. For the period 1 to 670 DIM, increasing the energy level in the diet (control 670 vs. high 670) resulted in a similar yield of milk and a similar fat concentration in the milk, but greater yields of milk fat and protein and greater milk protein percentage of the milk. The full TMR 670 group produced greater yields of milk and milk components (fat, protein, and lactose) and also protein percentage in the milk than the other groups. The milk solids (fat + protein) ratio for the 3 extended-lactation groups, defined as production achieved during the 24-mo calving interval divided by 2 yr (annualized production) expressed as a ratio of that produced in the normal 12-mo calving interval, was not affected by increasing the level of grain in the pasture-based diets (0.93 vs. 0.90 for control and high diets, respectively), but decreased with the TMR diet (0.79). The control 670 group produced 7.1% less milk, but only 2.4% less milk solids than the control 300 group over the 2-yr period of the study. Combining our data with that from 2 earlier studies of extended lactation demonstrated that Holstein cows with a high proportion of Northern Hemisphere genes offered pasture-based diets could achieve a high milk solids ratio, a greater proportion of cows milking at drying-off, and lower body weight gain over the lactation.

Pasture-based dairy farms are a complex system involving interactions between soils, pastures, forage crops, and livestock as well as the economic and social aspects of the business. Consequently, biophysical and farm systems models are becoming important tools to study pasture-based dairy systems. However, there is currently a paucity of modelling tools available for the simulation of one key component of the system—forage crops. This study evaluated the accuracy of the Agricultural Production Systems Simulator (APSIM) in simulating dry matter (DM) yield, phenology, and herbage nutritive characteristics of forage crops grown in the dairy regions of south-eastern Australia. Simulation results were compared with data for forage wheat (Triticum aestivum L.), oats (Avena sativa L.), forage rape (Brassica napus L.), forage sorghum (Sorghum bicolor (L.) Moench), and maize (Zea mays L.) collated from previous field research and demonstration activities undertaken across the dairy regions of south-eastern Australia. This study showed that APSIM adequately predicted the DM yield of forage crops, as evidenced by the range of values for the coefficient of determination (0.58–0.95), correlation coefficient (0.76–0.94), and bias correction factor (0.97–1.00). Crop phenology for maize, forage wheat, and oats was predicted with similar accuracy to forage crop DM yield, whereas the phenology of forage rape and forage sorghum was poorly predicted (R2 values 0.38 and 0.80, correlation coefficient 0.62 and –0.90, and bias correction factors 0.67 and 0.28, respectively). Herbage nutritive characteristics for all crop species were poorly predicted. While the selection of a model to explore an aspect of agricultural production will depend on the specific problem being addressed, the performance of APSIM in simulating forage crop DM yield and, in many cases, crop phenology, coupled with its ease of use, open access, and science-based mechanistic methods of simulating agricultural and crop processes, makes it an ideal model for exploring the influence of management and environment on forage crops grown on dairy farms in south-eastern Australia. Potential future model developments and improvements are discussed in the context of the results of this validation analysis.