The feeding of cereal-based supplements is common in the Australian dairy industry, as it allows cows to increase intakes of total dry matter (DM) and metabolisable energy (ME), while achieving greater stocking rates, greater pasture utilisation and greater milk production per hectare than occurs when cows are fed pasture-only diets. However, for this practice to be profitable, it is important to know how much extra milk, milk protein and milk fat are produced for each kilogram DM consumed. This is difficult to determine in such a complex biological system. We combined information from 24 concentrate-feeding experiments using meta-analysis techniques, so as to develop improved prediction models of the milk, milk protein and milk fat produced when cereal-based concentrates are fed to grazing, lactating dairy cows. Model terms, consistent with biological processes, linear, quadratic and factorial, were selected according to statistical significance. The models were then tested in two ways, namely, their goodness of fit to the data, and their ability to predict novel production data from a further six, unrelated, experiments. A sensitivity analysis was also undertaken to determine how sensitive these predictions are to changes in key inputs. The predictive model for milk yield was shown to very closely reflect milk yield (kg/cow.day) measured under the experimental conditions in unrelated experiments (r = 0.96), with very little bias (Lin's bias correction factor = 0.98) and high concordance (Lin's concordance coefficient = 0.95). Predictions generated by multiplying predicted milk protein concentration by predicted milk yield closely matched observed milk protein yield (kg/cow.day) (r = 0.96, Lin's bias correction factor = 0.98, Lin's concordance coefficient = 0.95), and predictions found by multiplying predicted milk fat concentration by predicted milk yield closely matched observed milk fat yield (kg/cow.day) (r = 0.94, Lin's bias correction factor = 0.99, Lin's concordance coefficient = 0.93). Factors included in the new models for milk, milk protein and milk fat yield reported here have been identified previously as elements that can influence milk production. The value to the dairy industry from being able to predict profitable amounts of concentrates to feed at various stages throughout lactation is considerable. For farmers and their advisers, being able to apply these models to estimate the immediate marginal milk protein and milk fat responses to supplementary feeds should lead to more robust, efficient and profitable milk production systems.

Purchased concentrates are a significant variable cost of a dairy business. Farm economic theory states that feeding supplements will enable a dairy farmer to improve profit as long as the marginal revenue received from the milk produced exceeds the marginal cost of the supplement. To do this, the quantities of milk, milk protein and milk fat produced from a unit of concentrate added to the diet are needed. Recent research has compiled results from short-term concentrate feeding experiments conducted in Victoria over a 30-year period. Using these data, relationships for the response of milk production to cereal grain supplements in dairy cows grazing temperate pastures have been developed and shown to be a better predictor than previous relationships. These response functions were used in the present study to investigate the economics of tactical (short-term; weekly, monthly or seasonally) and strategic (medium- to longer-term) supplementary feeding decisions in a pasture-based system, including, specifically, how much concentrate should be fed in a particular farm situation, given a certain feed cost and milk price. In the present paper, the relevant production economics method is explained and applied to determine the amount of supplement to feed that will maximise the margin of total extra milk income minus the total cost of supplement, thereby adding the most to farm profit. Currently, when dairy farmers make decisions about how much more supplement to feed their herd, they are making implicit judgements about the extra milk, and other potential benefits, that they expect to result as well as what the milk will be worth. More finely tuned decisions about feeding supplements based on comparing marginal cost and marginal revenue would add more to farm profit than decisions based on other common criteria, such as feeding supplement for maximum milk production. While some farmers may already be feeding supplements close to the point where marginal cost equals marginal revenue, the formal method of marginal analysis reported here makes explicit what is done implicitly at present and tests farmers’ intuitive decision-making. More detailed information about the responses to supplements and the costs and benefits of feeding supplements under particular circumstances at different times through the lactation has the potential to enable better, more profitable decisions to be made about feeding cows and managing the whole farm.

The anticipated effects of climate change, competing demands from the environment, industry and urban users, and changes in water policy are likely to reduce the amount and increase the variability of water allocations to dairy farmers in northern Victoria. The way two irrigated dairy farms that differed in feedbase characteristics, herd size and farm area, would operate and perform with reduced and more variable water allocations was examined over 10 years. Strategies to manage the impact of changed water availability were tested; namely, increasing milk production by feeding more supplementary feed, changing the feed system to present supplements in a partial mixed ration (PMR), and increasing milk production by using a PMR. Neither farm was profitable under medium climate change, or if the conditions that generated the low inflows of water into irrigation supply dams between 1996–97 and 2006–07 prevailed, unless changes were made to the farm system. Feeding supplements in a well formulated mixed ration have the potential to increase the efficiency of metabolisable energy use and offers the opportunity to increase feed intake and milk production. A PMR system enabled one of the farms to maintain and increase profit under medium climate change conditions; however, risk, measured as variability in profit, also increased. Under more severe reductions in water availability, neither of the farms examined was profitable over the run of years. Changes to the farm system other than feeding additional supplementary feed to increase milk production and/or using a PMR system, would be needed to counteract the effects of reduced and more variable water availability and maintain profit.

In this experiment 12 rumen-fistulated cows in late lactation, grazing Persian clover (Trifolium resupinatum) at four pasture allowances [9, 16, 32 and 53 kg dry matter (DM)/cow. day], had their entire rumen contents removed and sampled six times over 6 days, with the time of sampling varied between days. These rumen pool size measurements occurred at 0645 hours, 0915 hours, 1045 hours, 1345 hours, 1530 hours and 1900 hours. The cows were offered new allocations of pasture twice daily at 0715 hours and 1530 hours. Herbage intake increased (P < 0.001) from 5.6 to 20.4 kgDMas pasture allowance increased. Pasture allowance had no effects on rumen pool sizes of wet matter (WM) (P = 0.95), DM (P = 0.914), neutral detergent fibre (NDF, P = 0.499) or the DM% of the digesta (P = 0.078). Rumen pool sizes of WM, DM and NDF varied throughout the day. Pool sizes of WM and DM were lowest at 0645 hours and highest at 1900 hours for all allowances (P < 0.001) and along with pool sizes of NDF, were also lower at 0645 hours compared with 0915 hours and at 1530 hours compared with 1900 hours. The lack of increase in rumen pool sizes as pasture allowance increased when cows grazed Persian clover suggests that outflow rates increased with the increasing intakes. The low, and similar, rumination times between allowances also suggests that there is little restriction to the ability of outflow rates to increase when cows graze Persian clover at increasing allowances. The changes in rumen pool sizes over the day were consistent with the twice daily allocation of fresh pasture to the cows. The diurnal rumen pool size information obtained from this experiment will assist with the development of mechanistic rumen models that are capable of simulating the discontinuous intake and varying rumen pool sizes of cows in grazing systems.

Effect of herd nutrition, time of year and season of calving on milk fat composition and physical properties were examined on irrigated commercial dairy farms in northern Victoria that made use of split-calving and a diverse range of feeding systems. Twenty-four farms were included in the study, and from each farm, morning and evening milk samples were collected from 16 cows that calved in autumn and 16 cows that calved in spring. There were no significant effects of season of calving on the concentration of fatty acids or phospholipids in milk fat, but there were interactions between season of calving and time of year (P<0.001). These differences could be attributed to changes in energy balance and body condition with stage of lactation. The phospholipids comprising mainly phosphotidylcholine (PC), phosphtidylethanolamine (PE) and sphingomyelin (SP) also varied, with PC and PE being highest in late lactation and SP lowest during peak and mid lactation for both calving groups. Milk fat colour and the concentrations of free fatty acids were more influenced by factors associated with time of year rather than stage of lactation. Milk fat colour in particular showed strong seasonal variation being distinctly lighter in summer-early autumn when compared with rest of the year. Increasing the amount of concentrates fed was associated with decreases in short-chain fatty acid concentration and increases in the solid fat content of milk fat. Variations in nutritional management practices had only small (non-significant) effects on fat composition.

A case study and whole-farm modelling approach was used to examine the potential impacts of negative associative effects on milk production and economic performance of two dairy farms in northern Victoria. The two case studies differed in herd and farm size, calving pattern, forages grown and use of labour, but both had production systems based on grazed pasture, grain fed in the dairy at milking and conserved hay fed out in the paddock. The feeding system of each farm was altered by implementing a partial mixed ration (PMR), where cows grazed once a day and received supplements in a well formulated mix once a day. Negative associative effects between feeds were included in the biophysical modelling by deriving a relationship from published studies between declining neutral detergent fibre digestibility and increasing grain intake. Before applying a PMR system, both farms were profitable and earning competitive rates of return after tax, with mean real internal rate of return higher than 5%, and positive mean annual operating profit and mean net present value, at a discount rate of 5%. Feeding a PMR enabled both farms to increase profitability and internal rate of return, particularly if milk production was increased as well, but only when associative effects were less than those in the feeding system based on grain fed in the dairy and hay in the paddock. Increased profitability was also associated with higher standard deviation in annual operating profit, internal rate of return and net present value, in other words risk increased under the PMR feeding system, as the businesses would be more vulnerable to fluctuating supplementary feed prices.

Dairy farm systems are complex and diverse. Since the late 1990s, dairy farm businesses in Victoria, Australia have operated in particularly volatile natural, economic and policy circumstances, as well as dealing with the long-term challenge of persistently rising real costs and real prices fluctuating around a declining trend. Maintaining profitability in this environment requires sound decision making. Sound decision making in farming involves assembling, analysing and assimilating information about the whole system in a robust way. To assist this end, a decade ago, a dairy research and development project, Dairy Directions, was developed. In this program, the whole farm approach of farm management economics was used to investigate questions about options dairy farmers had to maintain and improve profitability, to achieve their goals. Commencing in one region of Victoria, with an initial focus on farm decisions, the scope of the work evolved to identifying gaps in scientific knowledge, contributing information to public policy formation, and expanded to other regions. A steering committee of farmers, related industry representatives, scientists and farm economists proved the key to success in this research process. In this paper, the whole farm economics approach to farm systems research is described and explained; in particular the role of using information about response functions, risk, time and case studies in answering questions about alternative farm futures. The application and results of the whole farm approach to a range of research questions about dairy farming in Victoria is presented. As well as confirming the known, findings have also identified unrecognized dimensions, and challenged theory.

Feed costs are the major component of the variable costs and a significant component of the total costs of milk production on Australian dairy farms. To improve farm productivity, farmers need to understand how much feed is being consumed and the nutritive characteristics of the diet. This paper reviews an existing simple approach, the ‘Target 10’ approach, which is commonly used by the dairy industry in Victoria to estimate annual forage consumption. An alternative approach – the ‘Feeding Systems’ approach – is then introduced. The ‘Feeding Systems’ approach is compared with estimated forage consumption measured under experimental conditions. An analysis of the sensitivity of both approaches to incremental changes in key variables is presented. The ‘Feeding Standards’ approach was concordant with estimated forage consumption measured under experimental conditions. Sensitivity analysis has highlighted key variables which may have considerable influence over simulated forage consumption using this approach. Given that none of the key variables tested in this analysis can be varied in the ‘Target 10’ approach, we feel confident that the ‘Feeding Standards’ approach provides an improved method of back-calculating annual on-farm forage consumption. Using a robust approach to calculate forage consumption which fully accounts for metabolisable energy requirements is important where farmers are using home-grown forage consumption as an indicator of farm feeding system performance. It is also important to understand the assumptions involved in estimating metabolisable energy supply from either supplements or forage.

Seventy multiparous Holstein-Friesian cows were fed different amounts of pasture and concentrates, or a total mixed ration (TMR), for 42 d in mid-lactation to test the hypothesis that the concentration of Se in milk would depend on the amount of Se consumed, when the Se is primarily organic in nature, regardless of the diet of the cows. Of the 70 cows, 60 grazed irrigated perennial pasture at daily allowances of either 20 or 40 kg of dry matter (DM)/cow. These cows received 1 of 3 amounts of concentrates, either 1, 3, or 6 kg of DM/cow per day of pellets, and at each level of concentrate feeding, the pellets were formulated to provide 1 of 2 quantities of Se from Se yeast, either about 16 or 32 mg of Se/d. The other 10 cows were included in 2 additional treatments where a TMR diet was supplemented with 1 kg of DM/d of pellets formulated to include 1 of the 2 quantities of supplemental Se. Total Se intakes ranged from 14.5 to 35.9 mg/d, and of this, the Se-enriched pellets provided 93, 91, and 72% of the Se for cows allocated 20 and 40 kg of pasture DM/d or the TMR, respectively. No effects of the amount of Se consumed on any milk production variable, or on somatic cell count, body weight, and body condition score, for either the pasture-fed or TMR-fed cows were found. Milk Se concentrations responded quickly to the commencement of Se supplementation, reaching 89% of steady state levels at d 5. When milk Se concentrations were at steady state (d 12 to 40), each 1 mg of Se eaten increased the Se concentration of milk by 5.0 μg/kg (R2 = 0.97), and this response did not seem to be affected by the diet of the cows or their milk production. The concentration of Se in whole blood was more variable than that in milk, and took much longer to respond to change in Se status, but it was not affected by diet at any time either. For the on-farm production of Se-enriched milk, it is important to be able to predict milk Se concentration from Se input. In our study, type of diet did not affect this relationship.

Six herds on five commercial dairy farms were involved in the production of high selenium (Se) milk. The farms had a range of herd sizes, herd structures, feeding systems and milk production per cow. On all farms, pelleted concentrate supplements containing Se yeast were fed twice daily in the dairy for 16 days. The objectives were to: (1) produce milk with Se concentrations exceeding 225 μg/kg on the five farms for pilot-scale production of a high protein milk powder; (2) validate a predictive relationship between Se intake and milk Se concentration developed in research; and (3) examine the time taken from the introduction of Se yeast to steady-state concentrations of Se in milk under a range of commercial farming conditions. We hypothesised that the relationship between Se intake and its concentration in milk found in research would apply on commercial farms. Daily Se intake, which was primarily from Se yeast in the pelleted concentrates, varied from 35 to 51 mg Se/cow. Grazed pasture and conserved forage contributed less than 1 mg Se/cow on all farms. The time taken from the introduction of pellets containing Se yeast to steady-state milk Se concentrations was 4–7 days. The steady-state Se concentrations in milk varied from 166 to 247 µg/kg, but these concentrations were only 55–72% of predicted values. All the milk produced from the five farms on the last 2 days of feeding of Se-enriched pellets was used to produce a milk protein concentrate with a Se concentration of 5.4 mg/kg. Factors that might have affected Se incorporation into milk and the implications of these results for commercial production of high Se milk or milk products are discussed.

Fifteen rumen fistulated Holstein cows in late lactation and fed a total mixed ration offered ad libitum were supplemented with Se yeast to provide 0, 11, 20, 30, or 42 mg of supplemental Se/day to test the hypothesis that amounts of Se secreted in milk, excreted in urine and feces, and apparently retained in tissues would increase in direct proportion to Se intake. One-half of the yeast supplement was placed directly into the rumen through the fistula of each cow just before milking in the morning and again in the evening, and estimates of average daily excretion of Se were made using total collections of urine and feces from 25 to 31 d after treatments commenced. Amounts of Se secreted daily in milk and apparently retained in tissues increased linearly with average daily intake of Se. The amount of Se excreted in feces and total excretion of Se in urine plus feces increased curvilinearly with Se intake, such that proportionately less Se was excreted as the amount of Se fed increased. On average, total Se excretion accounted for 66%, Se secretion in milk accounted for 17%, and Se apparently retained in tissues accounted for 17% of total Se intake by cows. Thus, in herds fed large amounts of Se yeast, most of the Se will be excreted and retained on-farm. High concentrations of Se will be found where urine and feces accumulate (e.g., yards and effluent ponds), and effluent management practices must be tailored to avoid environmental issues.

Sixteen cows in early lactation were individually fed diets consisting of fresh Persian clover (Trifolium resupinatum)-dominant pasture, offered to all cows at 3.7 kg DM/100 kg liveweight (LW); either alone or supplemented with amounts of crushed wheat ranging from ~0.3 to 0.9 kg DM/100 kg LW (four treatments with four cows per treatment). Cows fed Persian clover alone consumed 19 kg DM/day and total DM intake increased (P < 0.001) in a linear manner as the amount of wheat consumed increased, with no significant effects on clover intake. As the proportion of wheat in the diet increased, dietary neutral detergent fibre (NDF) concentrations declined from 28 to 24%, and in vivo NDF (P = 0.055) and acid detergent fibre (ADF; P = 0.015) digestibilities also declined. There were no significant effects of proportion of wheat in the diet on apparent digestibility of DM, organic matter or gross energy. The extent to which negative associative effects on NDF digestion was associated with the clover could not be determined as it was not possible to distinguish between the NDF derived from clover or wheat, but the decline in ADF digestibility suggested that most of the response lay with the clover since the wheat only contained relatively small amounts of ADF. Ruminal fluid pH was below 6.0 for more than 18 h/day in all cows. There was no effect of wheat in the diet on average ruminal fluid pH, but lowest values during the day were negatively related (P < 0.05) to the proportion of wheat in the diet. As the proportion of wheat in the diet increased, ruminal fluid ammonia-N concentration (P < 0.001) and the acetate + butyrate to propionate ratio (P < 0.001) decreased. The proportion of wheat in the diet did not affect nylon bag estimates of NDF degradation rates for grain or forage. Although most data indicated that effects of proportion of wheat in the diet on the utilisation of consumed nutrients were small, the marginal milk response to additional wheat averaged only 0.9 kg energy-corrected milk/kg DM wheat.

Sixteen cows in mid-lactation (milk yield of 23.8 +/- 2.3 kg/d) were individually fed diets consisting of chopped perennial ryegrass hay, offered at 3 kg of dry matter (DM)/100 kg of body weight (BW), fed either alone or supplemented with amounts of crushed wheat ranging from 0.4 to 1.6 kg of DM/100 kg of BW (increasing at nominal intervals of 0.4 kg of DM/100 kg of BW; 5 nominal treatments in total). Three cows were allocated to each treatment except the mid-range wheat treatment, which had 4 cows. Results were analyzed by regression because the intake of the wheat by cows within treatments varied. The hay was used to reflect the characteristics of summer pastures in southeastern Australia. Feed intake and fecal output were measured to determine digestion coefficients, feeds were incubated in nylon bags in the rumen, and rumen variables were monitored. Estimates of metabolizable energy (ME) of the hay from in vivo or in vitro digestibility were also compared. The digestibility of neutral detergent fiber (NDF) was depressed linearly as the amount of crushed wheat consumed increased to 36% of DM intake. The extent to which negative associative effects on NDF digestion were associated with the hay could not be determined, as it was not possible to distinguish between the NDF from hay and that from wheat. However, acid detergent fiber (ADF) digestion also declined, suggesting that most of the response lay with the hay because ADF was negligible in the wheat. Most data indicated that effects of proportion of wheat in the diet on the utilization of consumed nutrients were small. Despite substitution of wheat for hay reducing the forage intake of cows, there was a positive linear effect on marginal milk responses (1.3 kg of energy-corrected milk/kg of DM wheat). Mean rumen fluid pH declined as the proportion of wheat in the diet increased. The lowest pH for any individual cow during a 24-h period was 5.4, and the amount of time that rumen fluid pH was <6.0 ranged from 0 to 14 h depending on the amount of wheat consumed. It was concluded that these perturbations of the rumen environment were probably sufficient to result in negative associative effects. In addition, estimates of the ME content of the hay were higher when calculated from in vitro compared with in vivo digestibility, which has implications when estimating the amount of feed required for production.

Three experiments were conducted in Vietnam to measure responses in forage intake and diet digestibility to concentrate feeding in growing cattle. The basal diet in all experiments was grass (elephant grass, Pennisetum purpureum, in experiments 1 and 2; native grass in experiment 3) fed at 1.25% liveweight (DM basis) during the day and rice straw fed ad libitum at night. The concentrate in experiment 1 was cassava powder plus urea. In experiments 2 and 3, a formulated concentrate comprised of maize, rice bran, fish meal, urea and salt was used. The concentrates were offered at ~0.3, 0.7, 1.3 and 2.0% liveweight in each experiment, and a zero concentrate treatment was also included. Effects of the amount of concentrate on substitution rate of concentrate for forage and on diet neutral detergent fibre (NDF) digestibility were measured to assess interactions between feeds. Substitution rate was high (0.5–0.7 kg DM reduction in forage intake per kg DM supplement consumed) and not affected by amount of supplement fed in experiment 1. In experiments 2 and 3, there was no substitution at the lowest intake of supplement, but it subsequently increased with amount of supplement consumed, to a high of 0.3–0.5 kg DM reduction in forage intake per kg DM supplement consumed. The cassava powder supplement markedly decreased dietary NDF digestibility (from 62 to 41%), whereas the formulated concentrate only reduced NDF digestibility by 11 and 8% units in experiments 2 and 3, respectively. It was calculated that depressions in NDF digestibility would reduce the estimated metabolisable energy content of the basal forage by 1–3.6 MJ/kg DM at the higher supplement intakes. The importance of these results in making decisions on tactical feeding systems to fatten cattle in central Vietnam is discussed.

A study of irrigated pasture-based dairy farms that used split calving (autumn and spring) was undertaken in northern Victoria, Australia, to examine associations between nutrition, time of year and season of calving on the concentrations of isomers of trans 18 : 1 fatty acids and conjugated linoleic acids (CLA) in milk fat. Factors associated with time of year explained most of the variation, with the highest concentrations observed in spring and summer when pasture intake by herds was high. However, there was substantial variation observed between herds and time of year. The mean total CLA concentration was 9.1 mg/g milk fatty acids (range 1.1–35.4 mg/g) with the cis,trans-9,11 accounting for ~84% of the total CLA. The mean total trans 18 : 1 concentration was 60.5 mg/g milk fatty acids (range 13.6–267 mg/g) with vaccenic acid (trans-11 18 : 1) accounting for ~53% of total trans 18 : 1 fatty acids. Total CLA and vaccenic acid were highest in August–September (southern hemisphere spring) (15.1 and 76.3 mg/g milk fat) and lowest in November–March (5.6 mg/g milk fat) and May–July (9.53 mg/g milk fat), respectively. There was no association between season of calving and milk CLA or trans 18 : 1 fatty acid concentrations. Trans-10 and -11 18 : 1 fatty acids and trans/trans-CLA were negatively correlated with milk fat concentrations. Management strategies designed to increase the concentration of CLA and trans 18 : 1 fatty acids in milk fat would not need to consider the effects of season of calving or stage of lactation, but should focus on pasture availability and quality.