- Farms where cows eat more pasture and crop are consistently more profitable.
- DairyBase data suggests an increase of about $300 of operating profit per hectare is likely per extra tonne of dry matter per hectare eaten.
- A new prototype tool can help farmers determine a locally-relevant estimate of their herd’s potential to eat more pasture and crop.
New Zealand dairy farms vary widely in financial performance. Pasture management, soil fertility and drainage can all be improved1. However, a substantial proportion of any farm’s financial performance is determined by the amount of pasture grown and eaten on-farm2, 3.
Opportunities to improve profitability by increasing this factor do exist. For example, grazing management may not follow best practice. McCarthy et al.4 found around 25 percent of farmers were overgrazing and 25 percent were undergrazing, relative to recommended post-grazing residuals. Clark et al.5 found that, even on research farms with consistent management and measurement, the best paddocks had twice the yield of the poorest paddocks.
In real terms, increasing the amount of pasture and crop eaten on-farm by one tonne of dry matter per hectare (t DM/ha) could equate to around $300 of extra operating profit/ha. Neal et al.3 estimated that doing this could be worth approximately $200 million per year to the dairy sector overall.
However, looking at increasing pasture and crop yields is only part of the picture. We need a tool that can give farmers a more accurate estimation of potential pasture and crop eaten, so they can see what closing the gap might mean.
Mind the (yield) gap
The difference between the potential harvest and the actual harvest is the ‘yield gap’. Yield gaps have become an increasingly popular measurement for assessing the scope for improvement in farm practice and subsequent yield6. For example, for Australian wheat yields, Hochman et al.7 estimated the difference between district yields and modelled maximum yields, when moisture was the only limiting factor, to highlight the opportunity for improvement.
Recognising the difficulty in achieving the moisture-limited maximum, Hochman et al. defined the ’exploitable yield gap’ as the difference between actual yield and 80 percent of the moisture-limited maximum. This illustrates that a yield gap should also be considered in economic terms. Not only may it be difficult to achieve the maximum possible yield, there may be diminishing marginal returns, so achieving maximum yield may not be the most profitable target.
Measuring pasture vs crop eaten
In DairyBase (DairyNZ's online database) pasture and crop eaten is estimated from farm performance data. These estimates are based on the energy demand for animal maintenance and milk production, followed by subtracting the energy supplied by imported supplement.
Approximately 700 DairyBase farms per year have the data necessary to calculate pasture and crop eaten, although this information takes several months to be collected and made available for analysis. While areas of crop are recorded, reliable measures of crop yields are not generally available. This means it’s not possible to separate the contribution of pastures versus crops.
Regional benchmarks for pasture and crop eaten can be generated from this data, but other factors such as rainfall, soil, altitude, terrain and fertiliser use vary widely within most regions. This means a regional benchmark is likely to be of limited relevance to any given farm.
Pasture Potential Tool
Now for the good news: the improved availability of data and spatial estimation tools does allow many of the factors noted to be rolled into an interactive tool, which we call the ‘Pasture Potential Tool’. This tool defines pasture potential for a specified location as the ‘90th percentile of pasture and crop eaten on nearby farm’ (i.e. the level that only one out of ten farmers beat).
The prototype Pasture Potential Tool is available at dairynz.co.nz/pasture-potential
Using the tool
The tool allows farmers to select their region interactively or by entering an address (Figure 1). Coloured areas show the availability of data, with green representing the locations with the most data. The year of interest can be selected from a dropdown menu. This gives an indication of how many farms with pasture and crop eaten data are within a 60 kilometre (km) radius of the selected point (red circle). For reference, a 20km and 40km radius are also shown, in blue and green respectively.
The farmer can then filter the data further by selecting the most relevant characteristics. For example, in Figure 1, farms with mid to higher altitudes are selected, with the ’pumice’ soil order. All observations can be adjusted to a particular level of nitrogen (N) fertiliser assuming a response rate of 10 kilograms (kg) of dry matter (DM) per kg N fertiliser applied.
An example of tool outputs is shown in Figure 2. This chart shows the distribution of DairyBase pasture and crop eaten within a 20km, 40km and 60km distance from the chosen location. The potential achievable (actually the 90th percentile) is shown as a dark blue line with a numerical value. The uncertainty band of this estimate is shown as a shaded area around this level.
Charts do not appear unless there are at least four farms in that group. In Figure 2’s example, a farmer can choose to consider the small number of farms close by (within 20km) as the relevant peer group, or the larger number of farms in the 60km radius. Regardless, the indicative potential is around 12 to 13t DM/ha for that year.
When using the tool, take these points into consideration:
- Data is less available in some regions.
- Terrain (apart from elevation) is currently not taken into account.
- The farmer may be aware of factors that are not accounted for in the tool that could make a substantial difference.
- Environmental or other regulations may mean it is not possible to create an appropriate peer group.
Nonetheless, feedback from groups of farmers who have piloted the tool has been encouraging, reporting it to be a useful first step towards change and improvement.
Capturing the pasture and crop eaten gap
The gap between a farm’s potential for and its actual pasture and crop eaten can be determined using the Pasture Potential Tool in conjunction with the estimated pasture and crop eaten calculated by DairyBase (or from DairyNZ’s online tool for pasture and crop eaten assessment – dairynz.co.nz/pasture-eaten). DairyBase data presented on DairyNZ’s website shows a one DM/ha increase in pasture and crop eaten has corresponded to an average increase in operating profit of approximately $300/ ha in 2014/15.
For example, the possible value of meeting a 12t DM potential for a 100ha farm with a current pasture and crop eaten of 10.5t DM, would have been 1.5t DM x $300 x 100ha = $45,000.
Capturing the opportunity to improve
Once the gap between potential and actual pasture and crop eaten has been identified, the pathway to capture the opportunity still needs to be determined. This is likely to require holistic consideration of the farm. Even though substituting higher-yielding crops for pasture (e.g. for harvest) should improve the overall amount of pasture and crop eaten and increase profitability, this will not necessarily lead to higher profit.
For example, using 12 years of DairyBase data for the Waikato, a regression analysis showed that farms with 10 percent of area in harvested crop would be expected to have approximately 0.6t DM/ha more pasture and crop eaten (p<0.05). However, as there was no significant improvement in operating profit/ha from the increase in area allocated to high-yielding harvested crop, it is likely there was also an increase in costs related to the use of these crops.
Where to from here?
The future of the pasture potential concept may lie in more customised reporting (for example, in DairyBase), or via dashboard tools. New developments could see options for realtime comparisons across relevant peers for their pasture harvest in the year to date.
DairyNZ will also be improving the ‘pasture journey’ for farmers who are looking to improve their level of home-grown feed.
DairyNZ has a number of resources that can help farmers to capture their pasture potential opportunity. Go online to find out about:
- dairynz.co.nz/pasture-potential – view the prototype of the Pasture Potential Tool.
- dairynz.co.nz/pasture-eaten – an online tool for assessing pasture and crop eaten.
- dairynz.co.nz/feed – feed management information (including more specific tools such as DairyNZ’s Spring Rotation Planner or SRP).
- dairynz.co.nz/farm-gauge – Farm Gauge (specifically, its pasture component).
- Chapman, D., R. Rawnsley, B. Cullen, and D. Clark. 2013. Inter-annual variability in pasture herbage accumulation in temperate dairy regions: causes, consequences, and management tools. Proceedings of the 22nd International Grassland Congress, Sydney.
- Silva-Villacorta D., C. W. Holmes, N. M. Shadbolt, N. Lopez-Villalobos, W. Prewer, and C. B. Glassey. 2005. The productivity of pasture-based dairy farms in New Zealand with different levels of extra feed input. Proceedings of the New Zealand Society of Animal Production 65: 63–67.
- Neal, M., B. Dela Rue, and C. Eastwood. 2017. Defining the value proposition for using technology to improve pasture management and harvest more pasture. Proceedings of the 1st Asian-Australasian Conference on Precision Pasture and Livestock Farming (1ACPLF), Hamilton.
- McCarthy, S., C. Hirst, D. Donaghy, D. Gray, and B. Wood. 2014. Opportunities to improve grazing management. Proceedings of the New Zealand Grassland Association 76:75-80.
- Clark, C., A. Romera, K. Macdonald, and D. Clark. 2010. Interpaddock annual dry matter yield variability for dairy farms in the Waikato region of New Zealand. New Zealand. Journal of Agricultural Research 53: 187-191.
- Van Ittersum M. K., K. G. Cassman, P. Grassini, J. Wolf, P. Tittonell, and Z. Hochman. 2013. Yield gap analysis with local to global relevance – a review. Field Crops Research 143: 4-17.
- Hochman, Z., D. Gobbett, H. Horan, and J. Navarro Garcia. 2016. Data rich yield gap analysis of wheat in Australia. Field Crops Research 197:97-106.
This article was originally published in Technical Series December 2018