Partners Networking to Advance South Island Dairying

LUDF OLD Research

Blood urea levels in high producing dairy cows
Neospora and Reproduction Study
Nitrogen fertiliser and pasture quality
Selenium Fertiliser Field Trial
On-farm indicators of magnesium status in cows
The Substitution of Dairy By-Products for Pasture
Application of Dexcel Whole Farm Model to Lincoln University Dairy Farm
Modelling the effects of zero-inductions on profitability in dairy systems
Pasture analysis in relation to the spring dip in milk production
Monitoring climate stress in Canterbury dairy cows
Pasture species monitoring
Monitoring Pasture Growth, Soil Moisture and Irrigation Performance
Pasture Growth Rates and Quality
Shorter Milking Times
IMBADA Camera Technology
Effects of 'econ-n' on nitrate leaching and pasture production

Blood urea levels in high producing dairy cows [2005]

As an initial scope for this project the following objective was set: to confirm that high blood urea are a feature of cows in an intensive managed pasture dairy farm with known below average embryo survival (with good submission rates). The trial was designed to collect about 1200 blood samples for urea analysis.

About two hundred animals were sampled once a week for six weeks during the mating period giving 1187 samples from a total of 528 cows. Preliminary analysis suggests that reproductive performance of the cows is below expected levels and this may be indicating problems. The whole herd pregnancy rate after 3 rounds of insemination is 73%, and for animals within the subset sampled for urea at insemination (±10 days) is 69%. Within this subset, cows with High urea levels (above a mean of 46.4 mg / dl) had a pregnancy rate of 66.4 versus 71.43% for those cows designated as Low urea group. Sampled pasture had an average protein level of 223g/kg DM and digestibility of 798/kg DM.

(Conducted by Ricky Mandibela [postgraduate student], Alastair Nicol, and Andrew Sykes,Lincoln University. Funded by Lincoln University.)


Neospora and Reproduction Study [2005]

This study has investigated the possible involvement of disease issues in infertility at LUDF by [i] pregnancy testing to define the prevalence and timing of pregnancy loss and [ii] blood sampling and testing for Neospora and BVD. Thirty [30] cows [4.6% of the herd] lost pregnancies; this is very similar to the national average. Most pregnancies were lost from cows mated in the first three weeks and most pregnancy losses occurred soon after the cow conceived. Most farmers would observe these cows as long returns. All 31 cows tested were negative for Neospora; 9 however were positive for BVD. This is surprising since LUDF has consistently had a bulk milk BVD of less than 10%.

Both tests indicate that the LUDF herd does not have a major underlying disease issue affecting herd fertility.

(Conducted by Donald Arthur, Selwyn Rakaia Vets, Dunsandel; Adrian, van Bysterveldt, Dexcel. Funding from Lincoln University Dairy Farm [LUDF]).


Nitrogen fertiliser and pasture quality [2003]

The objective of this trial was to determine and compare the effects of Ammonium Sulphate Nitrate (ASN) and Urea fertilisers on pasture dry matter production and pasture quality of irrigated dairy pastures in Canterbury.

The influence of ASN and urea at two rates of N application on the production and quality of irrigated dairy pastures in Canterbury, New Zealand was investigated. Under the conditions of this trial representing typical frequent centre-pivot irrigation and high soil fertility, pasture production ranged from 15.9 [Control] to 19.4 [ASN 250] t DM ha-1 yr-1 and was strongly influenced by the rate of N fertiliser application. ASN showed a small significant yield advantage [P <0.05] over urea applied at the 150 N rate but not at the 250 N rate.

The clover content of the sward did not vary significantly with the type, or rate of
fertilisers. Seasonal variation in sward clover levels was high, and moderately high clover
contents were maintained on fertiliser treated pastures [from 10-45%]. Clover levels in the pastures were 5-10% more dry matter with virtually no drop in pasture quality.

Further pasture quality analysis revealed that amino acid concentrations in the herbage varied widely over the season. The data indicate that the variation of amino acid levels in the pasture sampled due to fertiliser treatment is however small.

(Conducted by Jim Moir, Keith Cameron, H J Di, Centre for Soil and Environmental Quality, Lincoln University; Ants Roberts, Ravensdown Fertiliser Co-operative Ltd; and Wybe Kuperus, Dexcel Ltd.).

The authors wish to thank Ravensdown Fertiliser Co-op Ltd for funding this study, and
Carole Barlow, Hamish Masters, Trevor Hendry and Koleightne Ford for field and technical assistance.

View the full report [PDF 665 KB]

Back to top

Selenium Fertiliser Field Trial [2004]

The objective was to compare the effectiveness of a range of selenium fertilisers for increasing pasture herbage selenium concentrations

Summary of selenium trial carried out on Lincoln University dairy farm from October 2003 - October 2004

The uptake of selenium (Se) by pasture herbage from four different Se fertiliser materials was determined in a field experiment throughout the course of a whole year.

The products tested were: the commercially available Selenium Chip™; Selcote® Ultra; and two neformulations (Ravensdown A and B), manufactured in a way to slow the release of Se in the form of sodium selenate. All four Se fertilisers increased pasture herbage Se concentrations above the level required to meet the nutritional requirements of grazing livestock; however, the size and duration of the responses varied between products.

The Ravensdown A product was as effective as Selcote Ultra at maintaining pasture Se concentrations above the critical animal health level for the whole year of the trial. The effects of the other two products, particularly the Selenium Chip, were not so long-lasting. Recoveries of Se in the pasture herbage throughout the year ranged between approximately 15-17% of the Se originally applied in the fertilisers.

(Conducted by R G McLaren. Funded by Ravensdown Fertiliser Cooperative Ltd).

 Back to top

On-farm indicators of magnesium status in cows [2004]

This study investigated the suitability of urinary Mg as an on-farm indicator of Mg status in dairy herds. Serum, urinary and milk Mg concentrations, along with estimates of Mg intake in Mg-supplemented and un-supplemented cows were measured from calving until mid-lactation.

Results from this research showed that urinary Mg concentration has potential to be used as an appropriate indicator of herd Mg status for pasture-fed dairy cows.

The use of urinary Mg concentration offered advantages over blood sampling in that at low serum Mg concentrations and excretion of Mg in urine will virtually cease when serum Mg concentrations fall below 0.7 mmol/l. Hence, cows with low concentrations of urinary Mg can be identified without requiring a veterinarian to take blood samples and the delay between sampling and return of results can be avoided. Under typical New Zealand dairy farm conditions urinary Mg concentration could be a practical indicator of herd Mg status without the need for correction for urine volume using urinary creatinine concentration.

This probably reflects the high and relatively constant water content of high quality pastures used in modern dairy production systems. Observations suggest that creatinine concentrations may fluctuate with changes in live weight that occur after calving and during peak milk production when cows are in negative energy balance.

A urinary Mg concentration reference level of 0.5 mmol/l is suggested. Further studies are required to confirm this for a wider range of New Zealand pasture-fed herds given the significant genetic variation in Mg metabolism that is a feature of the international literature.

Magnesium concentrations in the pasture ranged from 1.51 - 2.21 g/kg DM. Dietary antagonists such as potassium concentration in herbage were greater than 25 g/kgDM - the concentration at which significant impairment of Mg absorption can be anticipated. The estimated mean Mg availability (%) was 13 - 16 during the August to October period in the Mg-supplemented herd.

Mean serum Mg concentrations for both herds ranged between 0.72 - 0.96, and 0.77 - 0.92 for the supplemented and un-supplemented herds, respectively. Serum Mg concentrations from 28% of the Mg-supplemented cows were below the accepted critical clinical threshold of 0.6 mmol/l on one or more sampling days.

The supplemented cows were offered Mg at recommended rates via an in-line water dispenser. However, owing to lower than expected levels of trough water intake, the level of Mg supplementation initially, was not adequate, highlighting the need to also monitor water flow to adjust Mg concentrations or to dust additional Mg on pasture or silage.

In conclusion, it is suggested that urinary Mg concentration may be used as an appropriate on-farm indicator of herd Mg status in pasture-fed dairy cows in early lactation. The occurrence of an animal with a urinary Mg concentration of <0.5 mmol/l in a herd with a mean urinary Mg concentration of <4.0 mmol/l suggests the need for supplementation. Results indicate that for animals in a pasture-based system, with access to similar levels of dry matter and water intake, urinary Mg concentration can be used exclusively, without the need to be corrected for creatinine, thus providing a simple on-farm method to assess herd Mg status and evaluate the need for Mg supplementation.

(Conducted by Professor Andrew Sykes and Amanda Phillips (Masters student) with scholarship support from Dexcel).

Back to top

The Substitution of Dairy By-Products for Pasture [2005]

The objective of the trial was to quantify the level of substitution for pasture at three feeding levels of PROLIQÒ, none (control), 1/2 adlib and adlib using three reps of 8 calves for each treatment for 8 weeks. This was quantified using a combination of pasture disappearance, alkanes measurement of grazing intake and live-weight gain of groups of calves. Water and PROLIQÒ intake were measure and a plasma mineral assessment was carried out.

Conclusion: PROLIQÒ can be fed in large quantities to calves without adverse effects as long as care is taken with introducing them to the feed.

In this trial PROLIQÒ was substituted for pasture at the rate of -0.5 to -0.7kgDM of grass per kgDM PROLIQÒ consumed. This did not give a significant increase in consumption of dry matter in calves fed PROLIQÒ which would explain the lack of live-weight response. However, there is a pasture saving effect of PROLIQÒ making more pasture available for milking cows.

(Conducted by Dr Alistair Nicol, Lincoln University, and Karen Turnbull, Fonterra Summer Scholar).

 Back to top

Application of Dexcel Whole Farm Model to Lincoln University Dairy Farm [2005]

Dexcel has developed a whole farm model of dairy farms that allows investigation of the effects of management, climate (including irrigation) and soil type variability on pasture and animal performance and farm profitability. The ability to simulate multiple years allows us to examine risk related to both climate and prices.

In this study we are examining three aspects using the Lincoln University Dairy Farm as a case study. These are the impact of calving pattern, the decision rules around winter grazing and spring rotation planning and irrigation scheduling.

(Researchers: Pierre Beukes and Bruce Thorrold of Dexcel Ltd. Funding from Dairy InSight and FRST).

Back to top


Modelling the effects of zero-inductions on profitability in dairy systems [2004]

Inductions are an animal welfare issue and regulations aim to achieve <2% inductions on a national basis by 2010. There is a need to assess the longer-term impacts of alternative reproduction management strategies on farm economics and their success in solving the problem of poor reproductive performance.

Dexcel's Whole Farm Model was calibrated against observed data from Lincoln University Dairy Farm where 11% of the herd was subjected to inductions in the 2002/03 season, but a zero-induction policy has since been adopted.

Results from one-year simulations over five different climate years showed a significant decrease in farm profit ($105/ha, P<0.05) from the benchmark farm with 11% inductions (following "best practice" guidelines) to a farm where the later-calvers were culled and replaced by bought-in early-calving heifers and where stocking rate was raised moderately (<10%) to compensate for the higher proportion of heifers in the herd.

Multiple-year simulations reflecting carry-over effects showed that a strategy with a 10-week mating period, annual culling of 25% of the herd including all empties and replacement with synchronized heifers mated to start calving one week before planned start of calving can match the farm profitability of the "Induction" strategy if implemented consistently for 5-8 years.

View the full paper [PDF 45 KB].

(Conducted by P C Beukes, B S Thorrold, M E Wastney, C B Glassey, C R Burke and A M van Bysterveldt, Dexcel Ltd) (Funded by Dairy InSight and MAF (SFF)).


 Back to top

Pasture analysis in relation to the spring dip in milk production [2007]

This project will investigate causes of the dip in MS production at LUDF during October and November. The rate of decline in MS production is much greater during this period than later in the summer. We will investigate the extent to which the decline is related to pasture composition.

The project proposes:

  1. More detailed analysis of pasture composition through [i] sampling pasture in front of the cows three times each week, and [ii] sampling from pasture exclosures managed under a constant cutting regime.
  2. To investigate changes in the production of 'green odour' compounds when grass is damaged over the period September to December [may be extended].

Data collection for the project has been completed and analysis is underway on a detailed set of pasture analysis at high frequency over the period September to December 2006. It should be noted that 2006 was an atypical Spring.

Report - May 2007


  • The objective of this project was a more detailed investigation of the rapid decline in milk solids production experienced at LUDF during late October and November in the 2003/04, 2004/05 and 2005/06 seasons.
  • Analysis of a sample of production curves from dairy farms across New Zealand failed to identify consistent patterns of ‘Spring milk dip’. While some farms in each region showed a dip in milk production during the period of interest it was not as large as that observed on the LUDF and was not repeatable across years.
  • 2006 was an atypical year at LUDF, with a wet Winter and cold Spring resulting in high sugar levels, high DM% and low protein content in the early pasture. There was no evidence of differences in pasture composition between cultivars growing at LUDF (Bealey or Tabu).

A ‘Spring Dip’ in MS production was not observed at LUDF during late October/November 2006.

This allowed us to make comparisons of pasture covers and composition with previous ‘Spring Dips’, notably the large dip in 2005. The LUDF ‘Spring Dip’ appears to result from lower pasture covers in the period soon after peak milk yield and is not related to pasture composition.

It may be more of a feature of herds that adopt extremely low grazing residuals and so are more likely to run into low covers at this time, though national database information did not allow us to verify this assertion (see above). Further studies of this aspect are justified.

The project reinforces the importance of maintaining pasture monitoring during this busy (mating) period.

Further work should focus on ensuring a reliable indication of pasture covers during this critical phase of lactation.

View the full report [PDF, 73KB]

(Conducted by Richard Dewhurst, Jim Gibbs, Lincoln University, Dawn Dalley, Dexcel Ltd, and Richard Christie, SIDDC).


Back to top

Monitoring Climate Stress in Canterbury dairy cows [2006]

It is not known whether dairy cows in Canterbury experience thermal stress during hot, dry days. Measurements are being made of the diurnal range of changes in core temperature in dairy cows of differing body condition score (and milk production) on hot days compared with cool days.

Baseline core temperature data from individual cows are also being obtained and variations examined. Climatic data will also be examined to determine if extreme climatic events can be predicted.

The potential for thermal loading on cows waiting to be milked on summer afternoons appears to be minimal in Canterbury. Likewise, there appears to be minimal thermal impact on cows caused by walking distances up to 6 km on summer afternoons in Canterbury.

These conclusions must be tempered by the fact that few very hot days were encountered during these studies and both farms had low endophyte pasture. High endophyte pastures may increase susceptibility to heat stress and this factor needs to be considered where such pastures occur.

However, the fact that much of the Canterbury region has experienced only a very few days per year when THI exceeded a value of 75 in the 5 years from 1998 to 2003 (McLane et al., 2004), and this has continued to apply in the last 3 summers when temperatures have been lower than the average (National Climate summaries, NIWA National Climate Centre, Auckland), indicates that thermal loads experienced by dairy cows on summer afternoons in Canterbury are only occasionally going to impose severe stress on the animals.

It is likely that measuring the thermoregulatory effort required to dissipate thermal stress of cows by recording their breathing rate will provide better insight into the questions under study in this work. This makes development of a compact recording device for recording breathing rate a primary objective for the conduct of further studies.

(Dr Graham Barrell and Mark Bloomberg, Lincoln University, plus Masters student Duncan McLane).


Back to top

Pasture species monitoring

An evaluation of a range of pasture species conducted on part of the farm [paddocks N5 and N10). Information used to help guide decisions on pasture renovation.

(Conducted by Graham Kerr, Agriseeds)


Back to top


Monitoring Pasture Growth, Soil Moisture and Irrigation Performance

The objective of this trial was to provide data for two Crop & Food Research projects: 1) to provide commercial pasture production data to compare with rain shelter production data, and 2) to provide pasture production data to include in the data set to develop an irrigation calculator for use on other dairy farms.

Pasture growth measurements were taken on three paddocks, N7, N11, and S8 by AgResearch staff from September 2005 to May 2008. The paddocks were selected to represent low [N7 on Eyre shallow silt loam over gravel], medium [N11 on Templeton deep silt loam over gravel] and high [S8 on Wakanui deep silt loam over clay] water capacity. N7 and N11 were irrigated by the North Block centre pivot, and S8 by the South Block centre pivot. The farm manager informed us that the two pivots applied the same amount of water each season. Measurements were made at around monthly intervals using a cage technique. Three areas were trimmed with a mower after grazingt, then protected from stock and mown after the next grazing in that paddock.

HydroServices took neutron probe soil moisture measurements to c.1m depth in the same paddocks at 2-3 week intervals. They also collated the irrigation and rainfall data.

View results [PDF 84KB]

(Researchers: Dick Martin and Steve Thomas, Crop & Food Research. Funded by Dairy InSight and FRST).

Pasture Growth Rates and Quality

  1. Pasture growth rates were being monitored across the farm to guide grazing management decisions and provide information on growth rates under centre pivot irrigation.Information was also used to calibrate the platemeter for centre pivot irrigated South Island conditions.
  2. Pasture cages and platemeters were also being used to measure accumulative growth from pre-grazing and post-grazing residuals in paddocks N3 and S8.

(Conducted by Graham Kerr, Agriseeds, and Adrian van Bysterveldt, Dexcel Ltd)


 Back to top


Shorter Milking Times

The expected outcome of this project is new knowledge on the impact of implementing a systematic approach to determining the end-point of milking on overall milking efficiency in large South Island dairy farms. The project is compatible with the SIDDC key objective "to actively seek labour productivity gains through adoption of technologies and practices that reduce labour requirements or make the work environment more satisfying".

The objective is to demonstrate the outcomes of Australian and New Zealand research that has shown a significant reduction in overall milking time can be achieved by setting a maximum milking time based on average herd yield and milking interval.

Shorter Milking Secrets [PDF 93KB] - Presented at SIDE 2009.

(Researchers: Jenny Jago, DairyNZ, John Williamson, DairyNZ. Funded by SIDE and DairyNZ)


Back to top



IMBADA Camera Technology

In 2005 SIDDC allowed Lincoln Ventures Ltd access to the LUDF to conduct preliminary implementations of the IMBADA Image Processing platform to develop several dairy applications. Progress has been very encouraging, if somewhat slower than anticipated.

From November 2005-January 2006 a prototype camera was installed in the LUDF shed This was used to aperture images o cows for a prototype BCS measurement algorithm. The camera was also used to establish operating requirements for a more robust commercial design (heat, water, power supply, etc). The camera has also been used for additional image capture in August in relation to oestrus detection.

The camera has served its purpose in terms of technical performance measurement and a new design is being implemented, this will be much smaller as the entire platform has been revised. Data capture has led to good progress on developing oestrus detection routines as well as some progress on BCS, but this is a more complex problem.

The oestrus detection is now being undertaken in conjunction with LIC, so this will integrate very well into the Protrack System once this has been installed at the LUDF. The BCS development will now move to a different location.

(Researcher: Frank Bollen, Lincoln Ventures Ltd. Funded by LIC).


Back to top


Effects of ‘eco-n’ on nitrate leaching and pasture production

A recent research breakthrough at Lincoln University and Ravensdown Fertiliser Co-operative Ltd has discovered a new method to reduce nitrate leaching from pastoral soils by 60%, reduce nitrous oxide (a greenhouse gas) emissions by 75%; and at the same time increase pasture yield by 10 to 15%.

The concept involves the use of a nitrification inhibitor to reduce the rate of nitrification in soil and thus reduce the amount of nitrate in the soil. This reduces the risk of nitrate leaching and retains more nitrogen in the plant root zone, which in turn increases plant yield. Lincoln is working with Ravensdown Fertiliser Co-operative Ltd to provide this new nitrification inhibitor technology (called ‘eco-n’) to farmers. Ravensdown launched the commercial product in Feb 2004 and uptake by farmers has been rapid. Extensive lysimeter and field trials are being carried out throughout New Zealand as well as on the Lincoln University Dairy Farm.

Application of 'eco-n' nitrification inhibitor technology continues to produce annual pasture yield increases of over 20% on the LUDF pasture plots. The lysimeter and drainage plot trials continue to show that 'eco-n' can reduce nitrate leaching losses by between 50 to 80% on the LUDF.

(Researchers: Keith Cameron, Hong Di and Jim Moir, Lincoln University. Funded by Ravensdown Fertiliser Co-operative Ltd).


Back to top