Combining precision farming tools with Yield Prophet® is allowing Western Australian growers to apply inputs to crops based directly on subsoil moisture
- Western Australian growers are combining precision agriculture (PA) tools with Yield Prophet® to allocate inputs according to plant-available water capacity
- Research over four years on Brad Smith’s property has shown he has increased his gross margin by $49 per hectare, on average, by applying inputs using variable-rate technology
- Mobile app technology is being trialled to enable PA consultants to send prescription input maps wirelessly to growers.
An increasing number of growers in Western Australia’s northern agricultural region are adopting precision agriculture (PA) techniques to apply inputs according to plant-available water capacity (PAWC) for enhancing risk management and profit potential.
Brad Smith, who farms at Eradu, 50 kilometres east of Geraldton, is a case in point. He has worked with precision farming specialist Craig Topham, from Agrarian Management, over the past four years to use this technology to guide his crop management.
Mr Smith has been running trials on his property since 2010 to determine if, by varying inputs according to PAWC, he could reduce risk and achieve higher returns compared with traditional ‘blanket’ input application.
To demonstrate the thinking behind the work, this article focuses on the 2011 results.
Western Australian grain grower Brad Smith (left) has been working with Craig Topham, of Agrarian Management, on his Eradu farm to test the benefits of ‘convergence PA’ – a system that incorporates precision farming tools such as yield maps, EM38 and radiometrics surveys and Yield Prophet® to apply inputs based on plant-available water capacity. Trials over four years have found the system has increased Mr Smith's gross margin by $49 per hectare every year on average.
PHOTO: Nicole Baxter
Mr Topham likes to use ‘convergence PA’ when talking about the work he has done with Mr Smith and his other clients because it combines the knowledge generated from EM38 and radiometrics surveys, deep-soil tests and yield maps to develop large-scale variable-rate farming maps.
After gathering data, Mr Topham and Mr Smith selected a 983-hectare paddock that appeared to have uniform soils.
However, analysis of the data showed the paddock was more variable than it looked. It was subsequently zoned into three land management units representing high, medium and low production according to PAWC.
The crop-forecasting tool Yield Prophet® was applied to each zone. The soil at each site was then characterised (described) and loaded into the Agricultural Production Systems sIMulator (APSIM) to run Yield Prophet® scenarios.
Mr Topham says the soil characterisation process highlighted the variation in PAWC between each site. The low-production zone had a PAWC of 67 millimetres, the average-production zone had a PAWC of 99mm, and the high-production zone had a PAWC of 116mm.
“The high-production zone contains a higher percentage of clay, potassium and organic carbon and holds 57 per cent more PAW than the low zone,” Mr Topham says.
Validation strips (strip trials) were set up to allow the yield and the financial performance of each zone to be analysed across the paddock so that each treatment ran through all three zones to provide data on the treatment response by zone.
After Mr Topham made the prescription input application maps, Mr Smith used variable-rate technology (VRT) on his seeder to vary compound fertiliser and liquid nitrogen by zone.
The strategy applied higher phosphorus rates to the average zone because there were lower soil phosphorus levels and high PAW at sowing. Nitrogen rates were matched to the PAWC of each zone, with the low-PAWC zone receiving the lowest rate. All zones had a blanket application of urea and potash at the three-leaf stage.
When additional rain fell later in the season, Mr Topham says an extra, non-budgeted application of urea ammonium nitrate (UAN) was applied at GS31 to the medium (13 kilograms/ha of nitrogen) and high (19kg/ha of nitrogen) zones.
No additional nitrogen was applied to the low zone as Yield Prophet® modelling indicated that there was unlikely to be a response.
Season 2011 results
The total rainfall at Mr Smith’s property in 2011 was 395mm, of which 292mm fell in the growing season.
Mr Topham’s analysis through Yield Prophet® showed that 34 per cent of growing-season rainfall was lost through leaching on the low-PAWC zone.
The model also showed the conversion of rain into grain varied from 8.29kg of grain per millimetre of rainfall on the low PAWC zone, to 11.6kg/mm on the average zone and 12.3kg/mm on the high-PAWC zone.
Mr Topham says the gross margin per millimetre of growing-season rainfall varied from $1.81/mm on the low zone to $2.49/mm on the average zone and $2.71/mm on the high zone.
“For comparison, blanket rate inputs on the high-production zone produced a gross margin of $2.34/mm,” he says.
“The ability to target inputs according to PAWC and plant-available water resulted in an additional margin of $0.37/mm of rainfall, which equates to a 16 per cent improvement in the conversion of rain into grain.”
According to Mr Topham, 28 per cent of the 983ha trial area comprised the high-production zone, 47 per cent the average zone and 24 per cent the low zone.
For season 2011, the ‘convergence PA’ strategy saved $5.48/ha in input costs and generated an additional gross margin of $63/ha or a 20 per cent improvement in return on funds employed (ROFE).
Further analysis shows that the ability to vary post-emergent urea and potash at the three-leaf stage would have achieved input cost savings of $11.50/ha while maintaining the 20 per cent increase in ROFE.
Mr Topham says this was assuming the same yield differences were achieved as in 2011, with a constant grain price of $220 per tonne on-farm and the same fertiliser prices.
He says the analysis shows an investment in the development of a variable-rate cropping system based on EM38 and radiometrics, combined with the ability to predict the variation in PAWC through Yield Prophet® crop modelling, has demonstrated a payback period of one year.
“If the implementation of a variable-rate liming program is included, significant additional benefits would apply,” he says.
Over four years, Mr Topham says the research on Mr Smith’s property produced an average input saving of $15.49/ha, a 22 per cent increase in ROFE and a $49/ha increase in gross margin.
He says while there are no guarantees, growers considering implementing a ‘convergence PA’ system on their own farm could be reasonably confident they could pay for the cost of implementing a similar system in the first year and make an extra $25/ha.
Although learning to use the ‘convergence PA’ system was daunting in the first few years, Mr Smith says it was made easier by having access to a network of experts, such as Mr Topham and PA specialist Nick Ross, who could help if problems arose.
Four years down the track, Mr Smith says his confidence has grown in handling the prescription maps and loading them into the rate-control console in his tractor.
Now, Mr Smith uses variable-rate technology to differentially apply everything from nitrogen to trace elements and seed. Higher seeding rates are applied on non-wetting soils to improve germination.
Mr Topham says one of the biggest findings of the trials on Mr Smith’s farm has been the importance of adding potassium to lighter country even in dry conditions.
“In drier years we cut back on potassium, which proved to be the wrong thing to do,” he says. “It’s right to cut back on nitrogen in a dry year, but if you cut back too much on the potassium you can go backwards very quickly.”
Role for apps
Another advancement Mr Topham hopes will improve the ease with which growers use ‘convergence PA’ is the development of new apps for use on smartphones and iPads that enable prescription maps to be sent wirelessly to growers.
One such app, being developed by John Deere, is called Mobile Farm Manager, which in time will give Mr Topham the capacity to view and send prescription input maps to growers.
Currently, Mr Topham develops prescription maps and nutrition strategies by using several different software packages and then emails the data files to growers so they can open them in software programs such as APEX.
In many cases, because of the large file size and problems with internet connectivity, Mr Topham currently has to drive to the grower’s property.
The maps are then loaded onto growers’ computers and saved onto a data card before being taken to the tractor for uploading into the rate-control console.
Mr Topham is trialling the app with several growers this season to check its reliability for data transfer.
He says the app makes use of the mobile telephone network so will work wherever a text message can be received.
Going forward, Mr Topham says he would like to have a closer look at crop responses to various nutrients by soil type.
“For the lighter sands that don’t tend to hold water, I’d like to know the best mix of nutrients to apply,” he says.
“So far we’ve found there is a greater response to potassium as opposed to nitrogen on light soils, but we’ve got to go further than that and see if we can push more production out of the soil by better understanding it.”
0428 645 192,
0427 939 282,
A hole lot of knowledge
Western Australian grain grower Rohan Ford is using precision agriculture and Yield Prophet® to allocate inputs according to plant-available moisture as the season progresses.
PHOTO: Nicole Baxter
On the northern fringe of Western Australia’s wheatbelt at Binnu, 100 kilometres north of Geraldton, grain grower Rohan Ford is sitting in a 1.5-metre-deep hole looking at the red soil.
The pit has been dug to look at crop root growth and soil structure on untrafficked and trafficked soil that forms part of a controlled-traffic farming system Mr Ford has used on the farm for the past 13 years.
Sitting in the pit with him, it is easy to see a big difference between the soils. Root growth on the trafficked soil is nil, whereas on the untrafficked soil, root growth is prolific.
Mr Ford says for him it is all about making the soil as friable as possible so it works like a sponge: able to take in and release the maximum amount of moisture for use by crops.
“The more spongy the soil, the more moisture it holds,” Mr Ford says. “If you don’t squash it with machinery, our soil will hold a lot of water.”
Recently, Mr Ford – working with his consultant Chris Pinkney of Agrarian Management – has implemented a ‘convergence PA’ system on his home property to better estimate his soils’ plant-available water capacity (PAWC).
The main soil type on Mr Ford’s home block is yellow sand with variations in clay content, but other blocks comprise red sandy loams and red sands.
In the past, some of Mr Ford’s soils were considered so poor they were judged unviable for cropping and were therefore fenced-off and planted to tagasaste for cattle grazing.
When EM38 and radiometrics surveys were done adjacent to the tagasaste, the PAWC of the soils was discovered to be better than first thought. Mr Ford is now planning to remove the tagasaste and plant the area to crops.
The crop-forecasting tool Yield Prophet® has been used on two different soil types to gauge PAWC.
However, Mr Ford and Mr Pinkney also dig pits with a backhoe to inspect the soil and use a hand-held electronic moisture meter to measure PAWC at other locations.
Each year, Mr Ford and Mr Pinkney assess soil-test results and available moisture and Mr Pinkney develops prescription maps to vary seed and fertiliser at sowing.
Through careful machinery modification and help from precision agriculture specialist Nick Ross, all of Mr Ford’s machinery now has the capacity for variable-rate application of inputs.
Base rates of nutrients are added at seeding. Using Yield Prophet®, a hand-held electronic moisture meter and other visual assessments to check PAWC, Mr Ford now has more confidence to feed the crop as the season progresses.
In time, he looks forward to moving beyond Yield Prophet®, as some growers have done already, to one day making use of electronic moisture probes scattered across the farm to provide real-time information about subsoil moisture.
“Knowing the subsoil moisture content not only helps with applying nitrogen and other inputs, it also helps decide the type of crop and variety we plant,” Mr Ford says.
0429 331 045,
0429 100 363,
Variable soils profit from site-specific inputs
Darren Cobley, Ellendale, Western Australia, estimates using precision agriculture and Yield Prophet® to manage inputs according to plant-available moisture has saved him about $20 per hectare on potash alone.
Ellendale grain grower Darren Cobley farms in a medium-rainfall area 50 kilometres south-east of Geraldton, Western Australia. He says soil types vary considerably across his landholding, from heavy clays to red loams, gravels, yellow sands and deep white sands.
Three years ago, he had his farm surveyed using EM38 and radiometrics technology.
He worked with Craig Topham, from Agrarian Management, to ground-truth the data with deep-soil tests and used the results to categorise the soils based on plant-available water capacity (PAWC). Season 2012 was the first year inputs were applied differentially.
Soil tests are now done every year to refine the fertiliser management strategy.
For example, some of Mr Cobley’s soils do not require potassium, so he uses prescription maps supplied by Mr Topham to apply potash only where necessary. Mr Cobley says the rates can vary from nil to 25 kilograms per hectare in one pass of the seeding rig.
“Variable-rate technology (VRT) has allowed us to cut back our potash outlay from a single blanket application of about 25kg/ha to about 17kg/ha on average,” he says.
“For potash alone, that’s a saving of about $20/ha.”
With so many different soil types, Mr Cobley elected to only install the crop-forecasting tool Yield Prophet® on one soil type to monitor the variation in PAWC throughout the season.
Historical yield data is used to correlate the PAWC at his only Yield Prophet® site with the likely PAWC for the other soil types.
Mr Cobley pays particular attention to Yield Prophet® in late July and uses the model to determine if top-up nitrogen is warranted later in the season based on subsoil moisture.
Aside from fertiliser, Mr Cobley also uses VRT to differentially apply seed.
“On better soils we’ll apply about 100kg/ha of seed wheat as it is more likely to germinate,” he says.
“But on non-wetting soils we might apply 140kg/ha of seed to improve the germination percentage.”
0429 108 594,
Cover crops and drains help win a dry argument
Canola researchers in omega-3 breakthrough
GRDC Project Code