Team effort to solve the variability puzzle

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

The Ground Cover matchmaker series looks at specific researcher–grower collaborations. In this instalment we look at the significant role played by researcher Brett Whelan and grower James Hassall in developing precision agriculture in Australia

Image of growers and researcher.

James Hassall (left) and Brett Whelan (centre) collaborate on precision agriculture.

The researcher

Brett Whelan, "We can work on the science behind it but people just need to know what to do"

Perhaps more than any other field of agricultural research, the advancement of precision agriculture (PA) is dependent on scientists and growers working together over time.

You simply need the scale and natural variation of a working farm in order to test ever-evolving PA tools, first their ability to identify factors in crop production variability and then, their effectiveness in Australian conditions to apply inputs according to that variability.

This is certainly the perspective of Australian PA pioneer Associate Professor Brett Whelan, who heads Australia’s pre-eminent and longest-running PA research laboratory at the University of Sydney and has been involved in many advances in the field for more than 20 years.

Testing on small trial plots simply will not cut it, as the very nature of the technology aims to address variability at a farm scale. It is “crucial” to collaborate with working growers over successive seasons and in different regions, Brett says.

“You use the same paddock for consistency, to show the importance of different layers of information and the impact of PA practices across time, so we’re not just making decisions off one yield year, because conditions vary,” he says. “The real ideal is to understand variation across the growing areas of Australia, so we work with growers in a range of places.”

Brett attributes much of his success to his collaboration with the many farming families who have shared their paddocks with him – and among the most important and enduring has been his collaboration with James Hassall.

James and Brett have been collaborating for almost two decades, building essential information on PA tools for growers to use.

“We have been building a protocol – essentially a method – highlighting factors that are important in yield variability and how to go about implementing PA,” Brett says. “We can work on the science behind it but people just need to know what to do.”

The two met at a PA course Brett held in the 1990s soon after he brought the first yield monitor to Australia from the US – a PA tool that reads crop yield by the second, radically improving the accuracy by which growers measure variability in crop production across their properties.

James, who Brett describes as a true innovator “building and adapting his own technology”, was keen to learn more about his own crops’ variability and volunteered his Gilgandra farm in central New South Wales as an early study site.

This proved to be just the beginning and Brett has since conducted many trials there, helping to inform the development of PA and educating the best young minds in the industry through James hosting both undergraduate and PhD candidates on-farm.

Image of an early model protein sensor.

An early model protein sensor that Associate Professor Brett Whelan and James Hassall used to research the variation in grain protein at harvest.

For students to be on the property – and to have continuity through access to the same paddocks over time – is integral, Brett says. In fact, he nominates the hands-on education it affords as being essential to PA advancement. “Being able to take my students to the farm for days is invaluable to the industry because they get a much better education than sitting in the classroom,” he says.

Brett has played a leading role in many of Australia’s PA advances, which now includes GPS vehicle navigation on about 90 per cent of cropping enterprises, yield mapping on one-third and growers increasingly using a map-based “management class” approach in their operations, treating specific parts of a paddock differently according to yield potential.

Much of Brett’s work has been in collaboration with James on-farm, so much so that James is listed as a co-author on several research publications including one breaking new PA ground – on in-harvest protein sensors.

“Wheat is marketed based on protein levels so if you start changing your input levels of nitrogen (through PA practice) you need to understand how that might impact on protein because it has a direct link,” Brett says. “So we have been helping a company develop an on-harvester protein sensor, although the idea may still be a bit before its time.”

Brett says many advances would not have been possible without James’s collaboration over years of surveying, testing and understanding the underlying agronomy dictating crop variability according to Australian conditions.

Brett’s research has underpinned practices that form the backbone of PA practice today, such as the use of soil sensors, elevation data and yield maps to create statistically valid management class maps and a protocol to ground-truth them. These can then be used to direct variable-rate application of inputs to different parts of a paddock.  

It has also formed the basis of a guide to PA for industry and students, Precision Agriculture for Grain Production Systems, as well as two recent GRDC publications drawing substantially from James’s field data, including one on crop reflectance sensors.

Reflectance sensors, Brett says, represent the next phase of PA, adding a ‘real-time’ layer of information to the ‘historical’ soil, elevation and yield information used to vary input rates. Commercial products are already on the market, but research is continuing on how to best use them in Australia.

“We have done a lot of work on reflectance sensors, which measure the health and vigour of the plants in-season using red and near-infrared light,” he says. “Through work on James’s farms, we showed it does improve management, but there’s still work to do on how they can be of most benefit.”

As well as providing the resources with which to work, collaborating with growers such as James also helps direct researchers to advances that will be most beneficial to, and best received by, industry.

“It all comes down to business in the end,” Brett says. “We want to help growers delineate where yields really do change, what causes that change and what you need to do differently to optimise business by wisely using inputs and maximising production.”

Brett says there is value to be gained up and down the grain supply chain by incorporating PA from a marketing angle at the point of sale, back down to breeding as growers look to use different crops for different parts of the farm, and ultimately enhance both their business and the environment.

He sees a day when ‘PA’ will be back to just ‘A’ – when agriculture will implicitly employ PA tools and it will become the operational norm. “It’s clear that the benefits of PA will be widely incorporated into agriculture and new advances in management will continue to evolve,” he says.

More information:

Brett Whelan,
02 8627 1132,
brett.whelan@sydney.edu.au

Image of grower James Hassall with a GreenSeeker device.

James Hassall with a GreenSeeker® mounted on the front of a tractor for a mapping trial at his Gilgandra property.

The grower

James Hassall, "Discovering all those little things that characterise the paddock meant we could design management strategies."

It was the sound of grains hitting the cab of his header during harvest that first got James Hassall thinking there had to be more affecting his yields than met the eye.

The wheat crop looked uniform enough, but in high-yielding areas there was a lot of noise as clean grain was thrown against the back of the cab, whereas things went quiet in other parts of the paddock not yielding so well.

“There was a fair bit of variability but we had applied the same amount of fertiliser across the paddock and there didn’t seem to be any relationship between yield and straw volume,” he says. “I knew there was something else going on, but what was it?”

Precision agriculture (PA) had first sparked James’s interest as a university student, but it was at a week-long course hosted by Associate Professor Brett Whelan, introducing PA and detailing yield mapping, where he joined the dots.

Image of a harvester cab with laptops c.2004

In the early days of PA on James Hassall’s Gilgandra farm, more than a decade ago, computers were starting to play their part, processing inputs from sensors to vary input rates.

“Brett spoke about how GPS works and about using yield mapping to identify high, medium and low-yielding areas of the paddock and then how to combine those maps with other maps from soil and biomass sensors,” he says. “The whole approach just made a lot of sense.”

James’s father was an early adopter of direct drilling and no-till and, with an innate interest in technology, it is no surprise that James jumped at the chance when, over a beer after the course, Brett asked him to be involved in furthering his research.

Before long James was hosting groups of Brett’s students at his Gilgandra farm, six hours west of the University of Sydney, for up to a week at a time.

Turning his farm over as a living laboratory for students not only helped them learn and progress the advancement of PA, it also provided tangible business benefits to James by establishing farm characteristics and revealing some valuable information about his crops’ variable yields.

Soil conductivity testing helped determine the variability in his soil profiles, with a previously hidden gravelly soil layer in some parts providing important answers as to why some crops would fall over through the season.

This prompted James to lessen nitrogen application in those areas to reduce early vigour. “Discovering all those little things that characterise the paddock meant we could design management strategies so as not to push the crop too hard in those areas,” he says.

In one memorable early project measuring nitrogen variability, Brett split one of James’s 120-hectare paddocks into different zones based on yield maps and soil characterisation and had students “running around the paddock” collecting hundreds of samples for analysis.

“I was amazed at how variable the nitrogen was across the paddock,” James says. “It ranged from enough to grow a five-tonne-per-hectare crop in some parts to others where you would be lucky to produce 1t/ha. So it was massive.”

Traditionally, James and his family had based their soil nitrogen estimates on three core samples taken from across the paddock and had applied nitrogen uniformly, based on a 3t/ha return, so these results told him he should radically change his nitrogen use.

Today, a base nitrogen amount is applied along with two nitrogen-rich strips through the paddock. Throughout the season the crop is monitored against the nitrogen-rich strips for biomass and vigour; crops falling behind are top-dressed. For this purpose, James uses a commercial reflectance sensor – a GreenSeeker® – to detect live green vegetation.

PA tools are now an integral and integrated part of his farming system: a 2200ha enterprise cropping wheat, canola, durum, chickpeas, lupins and triticale on a full three-metre tramline based around a nine-metre header and seeder, and a 27m boomspray.

“I’ve tried to simplify the system a fair bit and concentrate on nitrogen at the moment because that seems to be the main driving factor in how my crops are performing,” he says. “I think I’m getting a lot better nitrogen-use efficiency these days and PA has given me the confidence to use more targeted nitrogen in my cropping system.”

James estimates his adoption of PA technologies has increased yields by 10 to 20 per cent. He says his involvement in the research advancing PA has also benefited his professional and personal development. 

For example, it helped inspire him to undertake a GRDC-sponsored Nuffield Scholarship in 2009 on PA, through which he found that Australia was at the forefront of both technological and agronomic advances in the field, and through which he was also able to investigate the future of farming equipment.

Of particular interest to him were real-time nitrogen sensors but, as with many innovations, the challenge, he says, is in finding a commercial partner willing to develop technology in advance of an established market.

His own work with Brett investigating real-time protein monitors also has significant potential for further research and commercial application. “We’ve got a pretty good handle now on what drives yield and yield variability across the paddock, so it would be great as the next step to work out what’s driving protein variability,” he says. “There’s plenty more to learn yet.”

Being named as a co-author on this, and other scientific papers, has, as well as being able to “wave it around at the kids”, also given him credibility when he speaks on PA. “It gives growers the confidence that what I’m saying is based on more than just my experience – it’s peer-reviewed research so it’s not just my opinion.”

Like Brett, James believes PA will be adopted more broadly as the cost of technology drops and more PA specialists emerge. “There’s no reason that once growers are shown the benefits of PA they won’t employ a PA consultant, just as they do an agronomist or accountant, to crunch all the numbers and look after that side of things.”

For James, who has been heavily involved in developing not only the on-farm evidence but also developing the PA hardware – “soldering wires and getting early technologies to talk to each other” – crunching computer data has not been so much of a stretch. “Because I’ve got an interest in it I do it myself, but I can see why others find it difficult,” he says.

Over time, James believes PA will become more familiar, much like the experience of direct drilling, which took a while to become established after it was introduced in the 1980s, but which is today commonplace. “It just takes a certain amount of time for it to step from being something new and wonderful and different to just being something standard,” he says.

More information:

James Hassall,
0428 488 214,

james.hassall@bigpond.com

Next:

SA trial assesses different weed strategies

Previous:

Post-claying nutrition explored

Region South, North