Predicting the likelihood of conception to first insemination of dairy cows using milk mid-infrared spectroscopy

The objective of this study was to examine the ability of milk mid-infrared (MIR) spectroscopy and other on-farm data, such as milk yield, milk composition, stage of lactation, calving age, days in milk at insemination, and somatic cell count, to identify cows that were most or least likely to conceive to first insemination. A total of 16,628 spectral and milk production records of 7,040 cows from 29 commercial dairy herds across 3 Australian states were used. Three models, comprising different explanatory variables, were tested. Model 1 included features that are readily available on farms participating in milk recording, such as milk yield, milk composition, somatic cell count, days from calving to insemination, and calving season. Days in milk and age at calving were incorporated into model 1 to form model 2. In model 3, MIR was added to model 2, but to avoid double counting, milk composition traits of model 2 were removed. The models were first trained on extreme data [i.e., including cows that (1) conceived to first insemination and (2) cows with no conception event recorded and with only 1 insemination]. Then, the models were validated in a fresh data set with all cows regardless of conception outcomes present to test for their ability to identify cows that conceived or did not conceive to first insemination. To do this, we ranked the predicted probability of all cows in the validation set and then selected the top and bottom records in varying proportions from 5 to 40% (i.e., where the model predicted the highest versus lowest likelihood of conception to first insemination, respectively) and compared with the actual values. The model's performance was evaluated through herd-year by herd-year external validation and measured as the proportion of selected records being correct. The results show that when more cows are selected (i.e., descending confidence), the accuracy of the models was reduced, and selecting the 10% of cows with the highest confidence of predictions produces optimal accuracy. Irrespective of the proportions, none of the models could predict cows that conceived to first insemination, with an accuracy around 0.48. When attempting to predict the bottom 10% of cows, which had the least likelihood of conception to first insemination, model 1 had prediction accuracy around 0.64. Compared with model 1, the addition of days in milk and calving age (model 2) resulted in a negligible improvement in prediction accuracy (0.01 to 0.03). Model 3 had the highest prediction accuracy (0.76), which implies that in the models tested, MIR is of primary importance in the prediction of fertility of dairy cows. In conclusion, this study indicates that MIR and other milk recording data could be used to identify cows with potential difficulty in getting pregnant to first insemination with promising accuracy.

Classifying the fertility of dairy cows using milk mid-infrared spectroscopy

The objective of this study was to investigate the potential of milk mid-infrared (MIR) spectroscopy, MIR-derived traits including milk composition, milk fatty acids, and blood metabolic profiles (fatty acids, β-hydroxybutyrate, and urea), and other on-farm data for discriminating cows of good versus poor likelihood of conception to first insemination (i.e., pregnant vs. open). A total of 6,488 spectral and milk production records of 2,987 cows from 19 commercial dairy herds across 3 Australian states were used. Seven models, comprising different explanatory variables, were examined. Model 1 included milk production; concentrations of fat, protein, and lactose; somatic cell count; age at calving; days in milk at herd test; and days from calving to insemination. Model 2 included, in addition to the variables in model 1, milk fatty acids and blood metabolic profiles. The MIR spectrum collected before first insemination was added to model 2 to form model 3. Fat, protein, and lactose percentages, milk fatty acids, and blood metabolic profiles were removed from model 3 to create model 4. Model 5 and model 6 comprised model 4 and either fertility genomic estimated breeding value or principal components obtained from a genomic relationship matrix derived using animal genotypes, respectively. In model 7, all previously described sources of information, but not MIR-derived traits, were used. The models were developed using partial least squares discriminant analysis. The performance of each model was evaluated in 2 ways: 10-fold random cross-validation and herd-by-herd external validation. The accuracy measures were sensitivity (i.e., the proportion of pregnant cows that were correctly classified), specificity (i.e., the proportion of open cows that were correctly classified), and area under the curve (AUC) for the receiver operating curve. The results showed that in all models, prediction accuracy obtained through 10-fold random cross-validation was higher than that of herd-by-herd external validation, with the difference in AUC ranging between 0.01 and 0.09. In the herd-by-herd external validation, using basic on-farm information (model 1) was not sufficient to classify good- and poor-fertility cows; the sensitivity, specificity, and AUC were around 0.66. Compared with model 1, adding milk fatty acids and blood metabolic profiles (model 2) increased the sensitivity, specificity, and AUC by 0.01, 0.02, and 0.02 unit, respectively (i.e., 0.65, 0.63, and 0.678). Incorporating MIR spectra into model 2 resulted in sensitivity, specificity, and AUC values of 0.73, 0.63, and 0.72, respectively (model 3). The comparable prediction accuracies observed for models 3 and 4 mean that useful information from MIR-derived traits is already included in the spectra. Adding the fertility genomic estimated breeding value and animal genotypes (model 7) produced the highest prediction accuracy, with sensitivity, specificity, and AUC values of 0.75, 0.66, and 0.75, respectively. However, removing either the fertility estimated breeding value or animal genotype from model 7 resulted in a reduction of the prediction accuracy of only 0.01 and 0.02, respectively. In conclusion, this study indicates that MIR and other on-farm data could be used to classify cows of good and poor likelihood of conception with promising accuracy.

Effect of timing of corn silage supplementation to Holstein dairy cows given limited daily access to pasture: intake and performance

The timing in which supplements are provided in grazing systems can affect dry matter (DM) intake and productive performance. The objective of this study was to evaluate the effect of timing of corn silage supplementation on ingestive behaviour, DM intake, milk yield and composition in grazing dairy cows. In total, 33 Holstein dairy cows in a randomized block design grazed on a second-year mixed grass-legume pasture from 0900 to 1500 h and received 2.7 kg of a commercial supplement at each milking. Paddock sizes were adjusted to provide a daily herbage allowance of 15 kg DM/cow determined at ground level. The three treatments imposed each provided 3.8 kg DM/day of corn silage offered in a single meal at 0800 h (Treatment AM), equally distributed in two meals 0800 and 1700 h (Treatment AM-PM) or a single meal at 1700 h (Treatment PM). The experiment was carried out during the late autumn and early winter period, with 1 week of adaptation and 6 weeks of measurements. There were no differences between treatments in milk yield, but 4% fat-corrected milk yield tended to be greater in AM-PM than in AM cows, which did not differ from PM (23.7, 25.3 and 24.6±0.84 kg/day for AM, AM-PM and PM, respectively). Fat percentage and yield were greater for AM-PM than for AM cows and intermediate for PM cows (3.89 v. 3.66±0.072% and 1.00 v. 0.92±0.035 kg/day, respectively). Offering corn silage in two meals had an effect on herbage DM intake which was greater for AM-PM than AM cows and was intermediate in PM cows (8.5, 11.0 and 10.3±0.68 kg/day for AM, AM-PM and PM, respectively). During the 6-h period at pasture, the overall proportion of observations on which cows were grazing tended to be different between treatments and a clear grazing pattern along the grazing session (1-h observation period) was identified. During the time at pasture, the proportion of observations during which cows ruminated was positively correlated with the DM intake of corn silage immediately before turn out to pasture. The treatment effects on herbage DM intake did not sufficiently explain differences in productive performance. This suggests that the timing of the corn silage supplementation affected rumen kinetics and likewise the appearance of hunger and satiety signals as indicated by observed changes in temporal patterns of grazing and ruminating activities.

Benefits and costs of grazing various proportions of perennial ryegrass and chicory for dairy production

Understanding the economic trade-off between changes in the supplementary feed required and the cost of pasture renovation is important when considering investing in alternative forages. Perennial ryegrass (Lolium perenne L.) is the main pasture species used for dairy production in temperate Australia. Alternatives to perennial ryegrass are grown to complement the seasonal growth pattern of perennial ryegrass, and to potentially increase annual dry matter (DM) yield. A case study analysis of a dairy farm in Gippsland was used to explore the benefits and costs over 15 years when either 0%, 20% or 40% of the milking area was sown to chicory (Cichorium intybus L.), with the balance sown to perennial ryegrass. Chicory was part of a 3-year pasture renovation cycle; in the year of establishment, annual ryegrass was sown in the autumn, with chicory sown in spring, followed by 27 months of production. This was compared with a 5-year renovation cycle of perennial ryegrass. Stocking rates of 3.3 and 2.5 cows/ha were modelled. A whole farm budget approach with stochastic simulation was used to quantify the potential effect on profit and risk. The profitability of growing chicory depended on the balance among (1) savings in supplementary feed costs during summer and autumn, and (2) possible reductions in the overall supply of DM during winter and early spring, and (3) increased pasture renovation costs. Stocking rate influenced the most profitable percentage of land sown to chicory. When stocking rate was 3.3 cows/ha, sowing 20% of the milking area to chicory returned a net present value (NPV) over 15 years AU$31 000 greater, on average, than did sowing 0% chicory, and AU$46 000 greater than sowing 40% chicory. With 2.5 cows/ha, sowing 40% of the milking area to chicory returned an NPV AU$39 000–AU$102 000 greater, on average, than did sowing either 20% or 0% chicory, respectively. The ratio of perennial ryegrass to chicory had little effect on the variability of NPV. For an individual farm, the most profitable percentage will fluctuate over time with variations in prices, seasonal conditions and management choices.

Development of a system to rank perennial ryegrass cultivars according to their economic value to dairy farm businesses in south-eastern Australia

Dairy production systems in south-eastern Australia are based primarily on grazed pasture. Perennial ryegrass (Lolium perenne L.) is the major grass species used in this region and farmers are faced with the challenge of choosing from more than 60 commercially available cultivars. This paper describes the development of a system termed as a forage value index that ranks the overall performance of perennial ryegrass cultivars relative to cultivar Victorian according to the summation of the estimated difference in the value of seasonal dry-matter (DM) yield of the cultivars. Average predicted seasonal DM yields were calculated by analysing the results of eight available perennial ryegrass plot trials across south-eastern Australia, using a multi-environment, multi-harvest linear mixed model. The differences in the model-predicted DM yield of each cultivar was compared with cultivar Victorian in each of five seasonal periods (autumn, winter, early spring, late spring, summer) to generate a series of performance values (1 per period) for each cultivar. Each performance value was then multiplied by an economic value (AU$/kg extra pasture grown) relating to each of four regions (Gippsland, northern Victoria, south-western Victoria, Tasmania) and seasonal period and aggregated to generate an overall forage value index rating for each cultivar. Economic values ranged from AU$0.11 to AU$0.39 per extra kilogram of DM grown, depending on the season and region, which translated into estimated benefits on dairy farms of up to AU$183 per ha per year for farmers that use high-yielding cultivars in place of cultivar Victorian perennial ryegrass.

Assessing the profitability of native pasture grazing systems: a stochastic whole-farm modelling approach

Grazing enterprises on the Central Tablelands of New South Wales employ a range of different strategies to manage temperate native grassland pastures common in the high rainfall zone of southern Australia. This paper uses a stochastic whole-farm simulation modelling approach to assess the impact of grazing system and stocking rate (SR) on the long-term profitability of a representative case-study enterprise. In particular, the impact of infrastructure costs, debt and downside risk, on whole-farm performance are examined over a 10-year planning horizon. In total, 12 different strategies were modelled under both price and climate risk, with a matrix of three paddock systems (1-paddock, 4-paddock and 20-paddock rotations) and four stocking rates (SR of 3, 4.2, 5.3 and 7 ewes/ha). Profitability was primarily driven by SR. In general, higher SR increased total farm output and annual profits under favourable conditions, although they were also associated with higher costs and greater downside risk in poor seasons, which in turn was magnified by the compounding effect of accumulating debt over time. When SR increased above 4.2 ewes/ha, it had a negative impact on lamb sale weights, resulting in lower prices due to lambs not meeting the ≥40-kg liveweight specification. Although this was offset by increased whole-farm production volumes at 5.3 ewes/ha, declines in profitability occurred at 7 ewes/ha as a result of significant increases in supplement feeding costs, and lambs not meeting sale weight specifications. The analytical scale of the analysis also had an impact on the relative profitability between alternative treatments. When assessed using a partial measure of economic analysis (gross margin per ha), there was little difference between paddock system treatments at the same SR. When the cost of additional fencing and water infrastructure were accounted for at the whole-farm analytical scale, the 20 paddock system was markedly less profitable than the 1- and 4-paddock rotations. This highlights the need for assessing production systems at an appropriate analytical and temporal scale to better understand the relationship between the key drivers of long-term profitability and risk. Overall there were relatively small differences in whole-farm performance between the four best performing strategies in this study. Given the trade-offs between profitability, downside risk, ground cover and feedbase sustainability, the lower risk 1- and 4-paddock systems with a SR of 4.2 ewes/ha are proposed as being optimal.

Increasing home-grown forage consumption and profit in non-irrigated dairy systems. 2. Forage harvested

A dairy farmlet experiment was conducted at Terang in south-west Victoria, Australia, over 4 years to test the hypothesis that a 30% increase in forage harvested per ha could be achieved in a production system that incorporated a range of Complementary Forages with perennial ryegrass (CF) compared with a well managed perennial ryegrass-only farmlet (‘Ryegrass Max’, RM). The CF farmlet included perennial ryegrass pasture (44% of the farmlet area on average over 4 years), but also incorporated oversowing perennial ryegrass with short-term ryegrasses (average 16% of farmlet area) to increase winter growth, tall fescue-based pasture (average 20% of farmlet area) to increase production in the late spring–summer period, a double cropping rotation (15% of farmlet area) based on winter cereal for silage production followed by summer brassica crops for grazing, and summer crops used in the pasture renovation process (average 5% of farmlet area). The RM and CF farmlets were stocked at 2.2 and 2.82 June-calving cows/ha, respectively and average annual nitrogen (N) fertiliser application rates (pasture only) were 141 and 153 kg N/ha, respectively. The total amount of forage harvested per year was generally less than predicted from pre-experimental modelling of both farmlets. However, the proposed target of a 30% increase in home-grown forage harvest per ha in the CF system compared with RM was exceeded in 2005–06 (+33%), with 21, 16 and 11% higher forage harvest achieved in CF in 2006–07, 2007–08 and 2008–09, respectively (average for all 4 years = 20%). Annual forage harvested in RM ranged between 6.5 and 8.9 t DM/ha compared with 7.9–10.3 t DM/ha in CF. Approximately two-thirds of the increased forage harvest in CF came from higher rates of pasture consumption per ha and one-third from the double cropping component of the system, although the performance of the double crop (mean annual production of 11.5 t DM/ha) was well below the expected 20 t DM/ha based on pre-experimental modelling. The higher per-hectare pasture harvest rates in CF were primarily due to increased perennial ryegrass pasture consumption achieved through higher stocking rates and efficient responses to higher N inputs from both higher fertiliser rates and additional supplementary feeding. In CF, the DM harvested from pastures oversown with short-term ryegrasses was lower than perennial ryegrass, while tall fescue-based pastures were similar to perennial ryegrass. Poor spring rainfall in 2006–07 and 2008–09 likely contributed to the lower than expected DM yields of tall fescue-based pasture and the summer crops within the double cropping component. Home-grown forage harvest rates can be increased by 11–33% above what is currently achieved by best industry practice with perennial ryegrass-only pastures using complementary forages but perennial ryegrass will remain a key component of the forage base for dairy production in southern Australia.

Increasing home-grown forage consumption and profit in non-irrigated dairy systems. 4. Economic performance

The profitability of dairy farm systems in southern Australia is closely related to the amount of pasture grown and consumed on-farm by dairy cows. However, there are doubts regarding the extent to which gains in feed supply from perennial ryegrass pasture can continue to support productivity growth in the industry. A farmlet experiment was conducted in south-western Victoria for 4 years (June 2005–May 2009), comparing a production system based on the use of forage species that complement perennial ryegrass in their seasonal growth pattern (‘Complementary Forages’, or CF) with a well managed system solely based on perennial ryegrass pasture (‘Ryegrass Max’, or RM). The forage base in CF included perennial ryegrass with a double-cropping rotation of winter cereal grown for whole-crop silage, followed by a summer brassica for grazing on 15% of farmlet area, a summer-active pasture based on tall fescue (on average 20% of farmlet area), perennial ryegrass oversown with short-rotation ryegrasses (average 16% of farmlet area) and summer brassica crops used in the process of pasture renovation (average 5% of farmlet area). The stocking rate was 2.2 and 2.8 cows/ha on RM and CF, respectively. Both systems were profitable over the 4 years of the experiment, with the modified internal rate of return over 4 years being 14.4% and 14.7% for the RM and CF farmlets, respectively. The coefficient of variation (%) of annual operating profit over 4 years was higher for the CF farmlet (56% and 63% for RM and CF, respectively). A severe drought in one of the 4 years exposed the more highly stocked CF system to greater supplementary feed costs and business risk. By comparison, the RM system performed consistently well across different seasons and in the face of a range of milk prices. The very small gain in profit from CF, plus the associated higher risk, makes it difficult to endorse a substantial change away from the traditional RM feed supply to greater reliance on summer-grown forages on non-irrigated dairy farms in southern Australia, as implemented in this experiment.

Effects of different strategies for feeding supplements on milk production responses in cows grazing a restricted pasture allowance

Milk production responses of grazing cows offered supplements in different ways were measured. Holstein-Friesian cows, averaging 227 d in milk, were allocated into 6 groups of 36, with 2 groups randomly assigned to each of 3 feeding strategies: (1) cows grazed perennial ryegrass pasture supplemented with milled barley grain fed in the milking parlor and pasture silage offered in the paddock (control); (2) same pasture and allotment supplemented with the same amounts of milled barley grain and pasture silage, but presented as a mixed ration after each milking (PMR 1); and (3) same pasture and allotment, supplemented with a mixed ration of milled barley grain, alfalfa hay, corn silage, and crushed corn grain (PMR 2). For all strategies, supplements provided the same metabolizable energy and grain:forage ratio. [75:25, dry matter (DM) basis]. Each group of 36 cows was further allocated into 4 groups of 9, which were assigned to receive 6, 8, 10, or 12 kg of supplement DM/cow per day. Thus, there were 2 replicated groups per supplement amount per dietary strategy. The experiment had a 14-d adaptation period and an 11-d measurement period. Pasture allotment was approximately 14 kg of DM/d for all cows and was offered in addition to the supplement. Positive quadratic responses to increasing amounts of supplement were observed for yield of milk, energy-corrected milk (ECM), and fat and protein, and positive linear responses for concentrations of fat and protein for cows on all 3 supplement feeding strategies. No difference existed between feeding strategy groups in yield of milk, ECM, or protein at any amount of supplement offered, but yield and concentration of fat was higher in PMR 2 cows compared with control and PMR 1 cows at the highest amounts of supplementation. Responses in marginal ECM production per additional kilogram of supplement were also greater for PMR 2 than control and PMR 1 cows when large amounts of supplement were consumed. For all diets, marked daily variation occurred in ruminal fluid volatile fatty acids and pH, especially in cows fed the largest amounts of supplement. It was concluded that when supplements are fed to grazing dairy cows, a simple mix of grain and pasture silage has no benefit over traditional strategies of feeding grain in the parlor and forage in the paddock. However, yield of milk fat and marginal milk production responses can be greater if the strategy uses an isoenergetic ration that also contains alfalfa hay, corn silage, and corn grain.