Yield potential rises in the high-rainfall zone

GroundCover Live and online, stay up to date with daily grains industry news online, click here to read more

Newly developed tools are helping to better predict the yield and economic benefits of applying inputs in the high-rainfall zone

Photo of Dr Penny Riffkin chairing a  2017 meeting of researchers

Dr Penny Riffkin chairs a 2017 meeting of researchers collaborating to improve wheat and canola yields in the high-rainfall zone.

PHOTO: Gio Braidotti

Grain yields in the high-rainfall zone (HRZ) have always posed a paradox, remaining far lower than modelling had predicted. However, that is now changing as a result of a comprehensive new research program covering farm-management practices, soil nutrition and plant biology.

Growers are beginning to find the modelling was right – yields of 10 tonnes of wheat per hectare (up from 3 to 4t/ha) and 6t/ha canola (up from 3t/ha) are a realistic possibility.

Heading the collaborative GRDC program is Dr Penny Riffkin, of Agriculture Victoria, in Hamilton. Also taking part are researchers from CSIRO, the University of Tasmania, the South Australian Research and Development Institute (SARDI), Southern Farming Systems, the MacKillop Farm Management Group, the International Plant Nutrition Institute and Yeruga Crop Research Pty Ltd.

“When we started the project and talked about a higher yield potential there was some concern that our targets were not realistic, but now a lot of growers are achieving those higher yields,” Dr Riffkin says. “The HRZ program is making it possible to understand where, when and how such yields are possible.”

Such productivity breakthroughs are important. The HRZ nationally is an area of continued grain-production expansion, with wheat production increasing nearly twofold over the past 20 years, from an average of 1.7 million tonnes during the years 1990 to 1995 to 3.2 million tonnes in 2007 to 2011. In this period, canola production jumped tenfold.

The GRDC has invested long term to help realise the greater potential and true yield potential of the HRZ. Now, findings from the crop-nutrition aspects of the HRZ research program have reached a critical point, culminating in proposed new fertiliser recommendations specifically for the HRZ, and the development of decision-support tools to help manage soil nutrient levels with regards to optimum, maximum and economic yields.

It is more than just nitrogen

Dr Riffkin says that at the start of the project, she heard from many growers who said they were not seeing the expected yield response from the application of nitrogen fertiliser. “We suspect we now understand why,” she says.

Levels of other soil nutrients – particularly phosphorus (P), potassium (K) and sulfur (S) – applied at the start of the season have been found to have profound impacts on the crop’s response to nitrogen (Figure 1 and Figure 2).

Bar chart showing response to applied nitrogen when limited by inadequate supply of other nutrients

Figure 1 Response to applied nitrogen is limited by an inadequate supply of other nutrients.

In this trial of wheat at Bool Lagoon, South Australia, in 2016, the 'Nil' treatment received no other nutrients at sowing, whereas the 'All' treatment received adequate phosphorus, sulfur, potassium, copper and zinc. Nitrogen was applied at 30, 68, and 187kh/ha, mainly as urea. Other treatments (not shown) indicated that the response was to P and S.

“Previously available crop nutrition information, including from the Better Fertiliser Decisions for Cropping Systems in Australia database, was found to lack relevance to the HRZ,” Dr Riffkin says. “So the work we are doing with a multidisciplinary team – including economists, crop system modellers, agronomists and soil scientists – involves understanding constraints to the yield potential and the consequences to the economic yield. Ultimately, we want to learn how to feed crops in the HRZ to achieve optimum yields.”

One of the messages from this data is that established critical soil values for P, K and S may be appropriate for 3t/ha crops, but are possibly not high enough to achieve the higher yield potential in the HRZ.

“We are not only seeing reduced yields, but also a reduced response to nitrogen due to insufficient P, K and S in the soil,” Dr Riffkin says. “However, decisions relating to the application of P, K and S must be made before the crop is sown. Nitrogen is the main nutrient for which you can make decisions in-crop.”



Graphic showing the predicted nitrogen response to the in-season application of nitrogen for a wheat crop at Inverleigh, Victoria

Figure 2 The predicted nitrogen response to the in-season application of nitrogen for a wheat crop at Inverleigh, Victoria.

B/C: Benefit/cost ratio.
The crop was grown under very good seasonal conditions (adequate soil moisture and no drought influencers). The expected yield outcome is in the top quartile (25 percent of historical outcomes). Pre-sowing soil mineral N (0-60 centimetres) - 160 kilograms N/hectare; Colwell P (0-10cm) = 10 milligrams phosphorus/kg, Colwell K (0-10cm) = 50kg potassium/kg, and sulfur KCI-40 (010cm) = 3 mg S/Kg. Sufficient P, K and S = 30, 200 and 10 units, respectively. The benefit-cost ratios are for the marginal unit of N applied and were calculated from a urea price ('as spread') of $435/tonne and a farmgate wheat price of $230/t.

Yield response to soil nutrients

A series of nutrient omission trials (in which all but one nutrient is provided at adequate rates) in South Australia and Victoria were used to identify the most limiting nutrients in different soils and their overall impact on yields and nitrogen response. Grain concentrations of the same nutrients were also assessed in random samples from grain-receival silos to provide a baseline for the HRZ.

Among the sites used to assess individual nutrient deficiency is a long-term phosphate experiment in Hamilton. The site was a pasture experiment for 40 years, but in 2017 was sown to canola. Nested within the larger site are small plot trials that received seven different rates of P fertiliser, which were added to six background levels of P ranging from 14 milligrams/kilogram to 143mg/kg Colwell P. The impact of P deficiency on canola was then readily discernible by contrasting growth in adjacent plots (Figure 3).

Similar trials are underway with wheat and canola at different sites, using various rates of applied P, K or S. The trials will also look at two rates of nitrogen: 60 and 100 per cent of yield potential.

Yield response curves to P, K and S are being developed, which highlight where the extra fertiliser will give yield increases and also identify optimum rates for economic yield increases.

Photos (top and bottom) showing impact of phosphorus deficiency on canola plots

Figure 3 Impact of phosphorus deficiency on canola in the high-rainfall zone

Even with a background level of 53 milligrams per kilogram Colwell P and plent of nitrogen (138kg N/hectare), crops in the high-rainfall zone that did not receive additional P at sowing (plot at the left of each image) are not growing as well as plants that are growing in soils that received 100kg of P before sowing (plot at the right of each image)

“With adequate P, K and S in the soil, we showed that nitrogen produces a much stronger yield response than if one or more of these nutrients are limiting,” Dr Riffkin says, referring to data shown in Figure 1.

“The same nitrogen application with limited P and S produced just 2.6t/ha, whereas when these nutrients were adequate the yield was 4.4t/ha. So this is about maximising the value of applied nitrogen.”

Overall, critical soil test values for 90 per cent of maximum yield in the HRZ appear to be higher than from trials conducted in the low and medium-rainfall zone in the Better Fertiliser Decisions for Cropping Systems in Australia database. For example, the critical Colwell P for wheat and canola in the HRZ, based on six trials, is 39mg/kg, whereas for wheat across all Australian trials it is 27mg/kg and 21mg/kg in canola.

“More trials are needed to gain greater confidence in critical values appropriate to the HRZ, and another 11 trials are underway in the 2017 season,” Dr Riffkin says.

The impacts can be dramatic, even with just a little added P, K and S; for example, on soils containing 130mg/kg available K – which is somewhat marginal but represents levels common to many paddocks in the HRZ.

Wheat did not grow as well in comparison to identical plots that received even just an extra 5kg/ha of K at sowing (Figure 4).

Photo showing wheat plots in the high-rainfall zone affected by phosphorus deficiency

Figure 4 Impact of phosphorus deficiency  on wheat in the high-rainfall zone.

In high-rainfall zone soils containing 130 milligrams per kilogram available potassium (K), wheat that did not receive extra K at sowing (central plot) did not fare as well as plants that received 5kg/hectare (right plot) or 150kg/ha (left plot) of K. All plots received 23kg of nitrogen per hectare at sowing, and mineral N at sowing to a depth of 60cm was 135kg/ha.

The new data will be incorporated into the Better Fertiliser Decisions for Cropping Systems in Australia database to help develop new recommendations. The first prototype of a decision-support tool has been developed that draws on the new crop-modelling capability. “Besides drawing awareness to the new soil nutrition recommendations, we want to provide an in-crop decision tool to identify the optimum nutrient application rates, with a view to making the best economic decisions given seasonal conditions and soil nutrient levels,”

Dr Riffkin says. “We also want to provide a post-crop evaluation function to assess and learn from past fertilisation regimes, including the ability to account for any possible carry forward of nutrients into the next season.” Ultimately, the researchers want to create a framework for getting the best return from investment in nutrients, whether that involves targeting the maximum profit or accommodating budget constraints in ways that result in optimum yields. With the work ongoing, more data is due to be rolled into the models, allowing the first generation of tools to evolve and acquire greater sophistication into the future.

GRDC Research Codes DAV00141-BA, DAV00116

More information:

Dr Penny Riffkin 
penny.riffkin@ecodev.vic.gov.au