The objective of this study was to identify genomic regions associated with milk fat percentage (FP), crude protein percentage (CPP), urea concentration (MU) and efficiency of crude protein utilization (ECPU: ratio between crude protein yield in milk and dietary crude protein intake) using grazing, mixed-breed, dairy cows in New Zealand. Phenotypes from 634 Holstein Friesian, Jersey or crossbred cows were obtained from two herds at Massey University. A subset of 490 of these cows was genotyped using Bovine Illumina 50K SNP-chips. Two genome-wise association approaches were used, a single-locus model fitted to data from 490 cows and a single-step Bayes C model fitted to data from all 634 cows. The single-locus analysis was performed with the Efficient Mixed-Model Association eXpedited model as implemented in the SVS package. Single nucleotide polymorphisms (SNPs) with genome-wide association p-values ≤ 1.11 × 10⁻⁶ were considered as putative quantitative trait loci (QTL). The Bayes C analysis was performed with the JWAS package and 1-Mb genomic windows containing SNPs that explained > 0.37% of the genetic variance were considered as putative QTL. Candidate genes within 100 kb from the identified SNPs in single-locus GWAS or the 1-Mb windows were identified using gene ontology, as implemented in the Ensembl Genome Browser. The genes detected in association with FP (MGST1, DGAT1, CEBPD, SLC52A2, GPAT4, and ACOX3) and CPP (DGAT1, CSN1S1, GOSR2, HERC6, and IGF1R) were identified as candidates. Gene ontology revealed six novel candidate genes (GMDS, E2F7, SIAH1, SLC24A4, LGMN, and ASS1) significantly associated with MU whose functions were in protein catabolism, urea cycle, ion transportation and N excretion. One novel candidate gene was identified in association with ECPU (MAP3K1) that is involved in post-transcriptional modification of proteins. The findings should be validated using a larger population of New Zealand grazing dairy cows.

In this study, we modelled and compared lactation curves of efficiency of crude protein utilisation (ECPU) and the nitrogen (N) excreta partitioning of milking cows of two contrasting spring-calving pasture-based herds to test some aspects of farming intensification practices on cow performance and N partition. In the low-intensity production system (LIPS), 257 cows were milked once-daily and fed diets comprised of pasture with low supplementary feed inclusion during lactation (304 kg pasture silage/cow). In the high-intensity production system (HIPS), 207 cows were milked twice-daily and fed pasture with higher supplementary feed inclusion (429 kg pasture silage and 1695 kg concentrate/cow). The dietary crude protein (CP) utilisation was calculated for each cow at every herd test date as the ECPU as a proportion of protein yield (PY) from the CP intake (CPI) derived from intake assessments based on metabolisable energy requirements, and the CP balance (CPB) calculated as the difference between CPI and PY. Total N excreta partitioned to faeces (FN) and urine (UN) was estimated by back-calculating UN from FN, considering dietary N, and from N retained in body tissues, taking into account live weight change during the lactation. The higher CPI (2.7 vs. 2.5 kg CP/day), along with the reduced milk yield (1100 kg milk/cow less), of the LIPS cows led to a lower ECPU (23% vs. 31%) and to a higher CPB (2.1 vs. 1.8 kg CP/day) when compared to the HIPS cows. Mean N excreta, and particularly UN, was significantly higher in LIPS cows, and this was explained by higher dietary CP and by the reduced PY when compared to the HIPS cows. Reducing the low-CP supplementation in the “de-intensified” herd lessened the ECPU, resulting in higher UN, which is sensitive in terms of body water eutrophication.

The objectives of this study were two-fold. Firstly, to estimate the likely correlated responses in milk urea nitrogen (MUN) concentration, lactation yields of milk (MY), fat (FY) and crude protein (CPY) and mature cow liveweight (LWT) under three selection scenarios which varied in relative emphasis for MUN; 0% relative emphasis (MUN_0%: equivalent to current New Zealand breeding worth index), and sign of the economic value; 20% relative emphasis positive selection (MUN_+20%), and 20% relative emphasis negative selection (MUN_−20%). Secondly, to estimate for these three scenarios the likely change in urinary nitrogen (UN) excretion under pasture based grazing conditions. The predicted genetic responses per cow per year for the current index were 16.4 kg MY, 2.0 kg FY, 1.4 kg CPY, −0.4 kg LWT and −0.05 mg/dL MUN. Positive selection on MUN in the index resulted in annual responses of 23.7 kg MY, 2.0 kg FY, 1.4 kg CPY, 0.6 kg LWT and 0.10 mg/dL MUN, while negative selection on MUN in the index resulted in annual responses of 5.4 kg MY, 1.6 kg FY, 1.0 kg CPY, −1.1 kg LWT and −0.17 mg/dL MUN. The MUN_−20% reduced both MUN and cow productivity, whereas the MUN_+20% increased MUN, milk production and LWT per cow. Per cow dry matter intake (DMI) was increased in all three scenarios as milk production increased compared to base year, therefore stocking rate (SR) was adjusted to control pasture cover. Paradoxically, ten years of selection with SR adjusted to maintain annual feed demand under the MUN_+20% actually reduced per ha UN excretion by 3.54 kg, along with increases of 63 kg MY, 26 kg FY and 16 kg CPY compared to the base year. Ten years of selection on the MUN_0% index generated a greater reductions of 10.45 kg UN and 30 kg MY, and increases of 32 kg FY and 21 kg CPY per ha, whereas the MUN_−20% index reduced 14.06 kg UN and 136 kg MY with increases of 32 kg FY and 18 kg CPY compared to base year. All three scenarios increased partitioning of nitrogen excreted as feces. The selection index that excluded MUN was economically beneficial in the current economic circumstances over selection indices including MUN regardless of whether selection was either for or against MUN. There was no substantial benefit from an environmental point of view from including MUN in the Breeding Worth index, because N leaching is more a function of SR rather than of individual cow UN excretion. This study demonstrates that attention needs to be paid to the whole system consequences of selection for environmental outcomes in pastoral grazing circumstances.

Precision in grazing management is highly dependent on accurate pasture monitoring. Typically, this is often overlooked because existing approaches are labour-intensive, need calibration, and are commonly perceived as inaccurate. Machine-learning processes harnessing big data, including remote sensing, can offer a new era of decision-support tools (DST) for pasture monitoring. Its application on-farm remains poor because of a lack of evidence about its accuracy. This study aimed at evaluating and quantifying the minimum data required to train a machine-learning satellite-based DST focusing on accurate pasture biomass prediction using this approach. Management data from 14 farms in New South Wales, Australia and measured pasture biomass throughout 12 consecutive months using a calibrated rising plate meter (RPM) as well as pasture biomass estimated using a DST based on high temporal/spatial resolution satellite images were available. Data were balanced according to farm and week of each month and randomly allocated for model evaluation (20%) and for progressive training (80%) as follows: 25% training subset (1W: week 1 in each month); 50% (2W: week 1 and 3); 75% (3W: week 1, 3, and 4); and 100% (4W: week 1 to 4). Pasture biomass estimates using the DST across all training datasets were evaluated against a calibrated rising plate meter (RPM) using mean-absolute error (MAE, kg DM/ha) among other statistics. Tukey’s HSD test was used to determine the differences between MAE across all training datasets. Relative to the control (no training, MAE: 498 kg DM ha⁻¹) 1W did not improve the prediction accuracy of the DST (P > 0.05). With the 2W training dataset, the MAE decreased to 342 kg DM ha⁻¹ (P < 0.001), while for the other training datasets, MAE decreased marginally (P > 0.05). This study accounts for minimal training data for a machine-learning DST to monitor pastures from satellites with comparable accuracy to a calibrated RPM which is considered the ‘gold standard’ for pasture biomass monitoring.

Milk urea nitrogen content is moderately heritable and is phenotypically related to urine nitrogen (UN). Based on this relationship, it has been suggested that genetic selection for lower milk urea nitrogen in grazing dairy cows could decrease UN concentration thereby reducing nitrogen excretions into the ground. The objective of this study was to compare the nitrogen use efficiency (NUE) and excretion in grazing cows with high and low milk urea nitrogen breeding values (MUNBV) in two farms of contrasting farming intensity. On the high-intensity farm (HIF) 68 and 70 cows with low and high MUNBV, respectively, were fed higher levels of supplementation and milked twice-daily, while on the low-intensity farm (LIF) 82 and 86 cows with low and high MUNBV, respectively, were fed lower levels of supplementation milked once-daily. Nitrogen use efficiency (g/g) was calculated as the ratio of daily milk N to daily N intake. Daily N intake (g/day) was derived from feed intake estimates based on energy requirements. The UN (g/day) was estimated by back-calculation from dietary N and subtracting milk N, faecal N, and N retained in body tissues. Irrespective of farm, cows with low MUNBV had significantly lower MY and milk urea nitrogen (p < 0.001) but this was not linked to significantly less UN. In the LIF, cows with low MUNBV had lower NUE (p < 0.001) than cows with high MUNBV, and this was explained by the reduced protein yield (p < 0.001). Selecting cows for low MUNBV was not an effective tool to reduce N losses and to increase the NUE in two dairy farms of contrasting farming intensity.

The objective of the study was to estimate genetic parameters of efficiency of crude protein utilisation (ECPU) and its genetic correlation ( r _G ) with other production traits across lactation in grazing dairy cows. Data were originated from 467 (2016–17 season) and 460 cows (2017–18 season) representing Holstein‐Friesian (F), Jersey (J) and their crosses (F×J), from two New Zealand farms. The ratio between crude protein yield in milk and crude protein intake is defined as ECPU. (Co)variance components between ECPU and other traits were estimated using bivariate random regression models (RRM) that included fixed effects of herd‐test‐date, parity, calving date, proportion of F, and F×J heterosis, with random effects of animal additive genetic, and within‐lactation permanent environment. Estimates of heritability ( h ² ) of daily ECPU ranged from 0.02 to 0.41. Estimates of r _G between daily ECPU and milk yield traits were strong and positive throughout lactation suggesting that selection for milk production traits improves ECPU of dairy cows. Daily r _G between ECPU and mid‐infrared predictions of milk urea (MU) concentration were mostly negative during lactation confirming that cows with high ECPU had genetically lower MU. Estimates of genetic relationships between ECPU and fertility traits are required to assess implications of including ECPU in breeding objective.