A research team from the University of Melbourne and Tasmanian Institute of Agriculture responsible for conducting modelling work to test and validate Fert$mart nitrogen best management practices provide an insight into how these latest guidelines have been prepared and relate to the various dairy regions.
Nitrogen (N) fertiliser use on dairy farms has increased in the past two decades, both in the number of farms and the frequency of application. But just how efficient is the use of N when applied as a fixed recipe after every grazing, as opposed to adjusting our N rates and timing based on growing conditions e.g. soil moisture and rainfall forecasts?
The national More Profit from Nitrogen program aims to improve N use efficiency (NUE or kilograms (kg) of extra pasture per kg N applied) and farm profit while reducing N losses, across the dairy, horticulture, cotton and sugar industries. One of the dairy projects is using the DairyMod computer model to check the current Fert$mart best management practice recommendations for N fertiliser, to ensure they stack up across a range of dairy pasture systems and conditions.
The modelling team, from the University of Melbourne and the Tasmanian Institute of Agriculture, started with challenging the basic 4Rs for N fertiliser, being the ù Right Rate, Right Source, Right Timing and Right Placement ù using test sites at Ellinbank, Vic; Terang, Vic; Elliott, Tas; Mt Gambier, SA; and Taree, NSW.
The project first looked at the optimal rate (Right Rate) of N fertiliser to apply for each month of the year. Figure 1 shows how the optimal N rate was determined, using an example of a spring N application on an irrigated site at Elliott (Tas). N fertiliser was applied from zero up to 150kg N per hectare per application, to define the full shape of the likely pasture growth response. The Ymax identifies the maximum pasture growth response to N (in this case, 100kg dry matter (DM) per ha per day), corresponding to 143kg N/ha for the particular grazing rotation.
Obviously, it would not be recommended to apply this high rate of N, as pasture growth responses are initially steep but taper off at higher N rates (what is called diminishing returns). For example, the first kg of N applied gave a response of more than 30 units of pasture DM per unit N applied, while the last unit of N applied produced almost zero additional pasture and therefore would not make economic sense to apply. In this case, under ideal growth conditions, at Elliott, a perennial ryegrass pasture could achieve 100kg DM/ha per day if 60kg N/ha was applied, assuming no other limitations. Obviously, this is almost the ideal growth conditions for perennial ryegrass.
Taking this further, Figure 2 shows some of the resulting N responses for all the sites in the research project, both irrigated and dryland, for each season separately. These are presented as box-plots to show the year-to-year variability (spread) over the past 20 years. Winter and spring show the most reliable pasture N responses at Elliott, Mt Gambier and Taree, while Ellinbank and Terang had an occasional failed spring. In contrast, summer and autumn periods show far poorer pasture N responses and greater variability between years, particularly at Ellinbank and Terang (the non-irrigated sites).
So how is this research better informing the Fert$mart Best Management Practice recommendations? To date, the study supports the following recommendations:
Apply N fertiliser at rates of 25-50kg N/ha per rotation (or 1-1.75kg N/ha per day) depending on the season of the year.
This study suggests that the optimum rate of N changes with growing conditions ù higher N rates can be applied when growth conditions are good (e.g. spring), but conversely, lower N rates should be applied when growth potential is lower (e.g. winter). Applying high N rates during times of low growth will lead to higher N losses.
Apply N fertiliser strategically, rather than by fixed recipe.
First compare the likely pasture N response with other purchased feed options, as the response efficiency changes through the season. Grain may be a cheaper option than additional N fertiliser at times of the year (e.g. autumn at Terang).
Spring and winter pasture N responses are more reliable than summer and autumn due to soil moisture limiting pasture growth (note how tighter the spread is between years in winter and spring in Figure 2, compared with the much broader spread in autumn and summer for the rainfed sites).
Soil moisture is far more important than rainfall in determining the N response.
Summer and autumn N responses are questionable unless the pasture is irrigated or the soil profile is already wet at the time of N application, with follow up rain predicted during the regrowth period.
In this modelling, 18 out of the last 20 years showed a flat response to N in at the autumn break at the dryland Terang site. This is due to a combination of a dry soil at the end of summer, together with plenty of N left in the soil from microbes breaking down soil organic matter in summer, but no plant growth using this N. D
For a copy of the research project's midpoint Fert$mart N BMP Guidelines, see More Profit from Nitrogen program webpage at: www.dairyingfortomorrow.com.au/tackling-specific-issues/soils and locate the document in the "Quantifying the whole farm systems impact nitrogen best practice on dairy farms" section.
This project is supported by funding from the Australian Government Department of Agriculture and Water Resources as part of its Rural R&D for Profit program, The University of Melbourne, Tasmanian Institute of Agriculture and Dairy Australia.