Hyperspectral imaging could do that and Massey University research, funded by Ravensdown and MPI through a Primary Growth Partnership, is testing its use on pasture at farm-scale.
By taking aerial images using hundreds of wavebands beyond the visual range of the electro-magnetic spectrum they’ve been able to identify “big differences in ME, crude protein and digestibility,” Massey University’s Professor Ian Yule told the New Zealand Grassland Association conference in Masterton last month.
“Everything has an electro-magnetic (reflectance) signature. If we can recognise that electro-magnetic signature we can work out the nutrient concentration in the pasture.
“This technology has got me the most excited I’ve been in 25 years.”
Samuel Dennis with the C-Dax pasture meter.
However, he also highlighted how far there was to go before the potential benefits were realised onfarm.
“We are seven or 15 minutes into the game. We’ve scored a couple of tries and Mother Nature is coming back at us but we’re pretty sure this is going to make a big difference to farming.”
Hill country is the focus of the Massey work because of the extent of such land in NZ, and how hard it is to sample soil or pasture over such rugged terrain and vast areas using conventional means. When image analysis techniques to determine nitrogen, phosphate, potash, sulphur and other characteristics of pasture are sufficiently proven then it could probably be deployed faster in lowland situations. Terrestrial spreader technology to variably apply nutrients is already available and affordable, Ravensdown’s general manager innovation and strategy, Mike Manning, said.
“Once they’ve developed the algorithms and technology to analyse the images for hill country I can’t see any reason why it couldn’t be applied to dairy and finishing country…. I imagine it would be very quick,” he said.
Improved hill country fertility and pasture could also hold a benefit for dairy because it could open up areas previously too infertile, and-or improve rearing of young stock in such areas, Manning said. He was almost as enthusiastic about the potential of the technology as Yule.
“Of all the things I’ve come across in 34 years working in agriculture this is one of the most exciting.”
Another NZGA paper, presented by Armin Werner of Lincoln Agritech, described adapting nitrogen-sensing technology using optical wavelengths – a technique well-proven in cropping – for pasture.
“Our arable colleagues have been using this technology for 15 years to see the nitrogen content of crops,” Werner said. “Five companies offer sensors for crops but for pastures they’ve not been developed yet.”
Werner and his co-authors (the paper lead author is Jessica Roberts) showed an ATV-mounted sensor working in the visual wavelengths could explain 76% of the variability in pasture nitrogen content. Werner said scientists might say that percentage wasn’t high enough but in practical terms it was getting close to what was required to start basic variable rate application onfarm.
“We want to be able to drive a very simple tool that can identify gate areas, troughs, stock camps, and apply less nitrogen or no nitrogen accordingly…. We’re not trying to go down to the level of the urine patch,” he said.
Even if individual urine patches were identified, it would be a challenge to avoid them with current spreader technology, hence the focus on a broader area. Besides, others, notably Agri-Optics from Methven, were working on developing technology to deal with individual urine patches, he said.
Werner said he was hopeful they could have a sufficiently robust sensing system to drive variable rate-nitrogen applications for pasture in six months to a year. It’s envisaged it could be integrated with existing spatial information systems such as TracMap or AgHub.
Pasture mapping to inform decisions
While the first two papers of the NZGA conference’s Technologies session dealt with future developments from a farm perspective, Samuel Dennis’ concluding paper showed how we could better use an existing tool: the C-Dax pasture meter.
With MPI, DairyNZ and Ravensdown funding, he and his co-researchers from AgResearch mapped covers at Lincoln University Dairy Farm for two seasons using the C-Dax and found mapping a month or so after peak growth the most reliable predictor of annual pasture production.
Yield-mapping pastures at that time could be used to identify under-performing areas and better inform remedial action.
“Yield mapping in itself doesn’t tell us what the problems with a pasture are but it does tell us where there are problems and from that we should be able to work out what, if anything, we can do about them,” Dennis said.
Despite LUDF being a relatively uniform farm the variation they found was considerable, with some areas within a paddock producing double the yield of other areas (see figure).
Whether mapping would be commercially viable would depend on the return the pasture generates and the cost of the service, Dennis said.
“It’s most likely to work for dairy, bull beef – anywhere where you’ve got intensive rotationally grazed pasture.”
Mapping at any point in the last third of the regrowth period – ie shortly before grazing – produced reliable results, so a rotationally grazed farm could be mapped in three visits as a contract service.
Parallel transects at up to 50m spacing were sufficient to identify good and bad areas of pasture, with no need to follow the same line each time.
However, narrow features causing variability could mean closer spacing was required on some farms, Dennis said.
At LUDF, factors affecting plant available water had the strongest link to pasture production, notably irrigation, topography and soil type.
As with nitrogen-sensing work, Dennis said little had been done on pasture yield mapping globally but all that was needed was a GPS-linked meter and software to interpret the data. Such meters were available off the shelf in NZ for less than $10,000 and the software could be downloaded for free.
“We’ve already got the tools: we just need to apply them.”
Pasture map protocol
Map approximately one month after peak pasture growth.
Measure pre-grazing in last third of regrowth period.
Use up to 50m run spacing – less if necessary to include features of interest.
Applies to rotational grazing only.
More detail: contact Dennis at www.grounded.co.nz or email Robyn Dynes at robyn.dynes@agresearch.co.nz
