Drone sensor trial measuring nutrients in pasture

By Dairy News

Imagine using a drone to fly over a couple of hundred hectares of pasture and within hours all the information about its biomass and nutrient levels are on your phone or computer.

Sounds a bit too good to be true?

Well, welcome to the future of pasture management.

And it’s only three to five years away, according to an expert.

Researchers recently flew a drone across a small plot of rye-grass – 50m by 50m – at a dairy in Mepunga West, western Victoria as part of a Federal Government-funded project called More Profit from Nitrogen.

University of Melbourne Department of Infrastructure Engineering Associate Professor Dongryeol Ryu oversaw collecting the images of the pasture and analytics retrieval.

He said this technology was still developing, but aerial imaging would help extend the capacity of information about feed on dairy farms.

“We have a pretty good idea how to get information of the biomass, based on the colour you can tell if it’s nutrient deficient or nutrient rich when in field,” Prof Ryu said.

“To get the same level of information over hundreds of hectares is a different story.”

That’s the focus of Prof Ryu’s work at Mepunga West.

He had used this sort of technology in the horticulture industry and others had used satellites and airborne data to do similar pasture measurements in dairy, but – as far as he knew – this was the first time nutrients were measured on a dairy farm via drone.

This specific project started three years ago, but before that Prof Ryu’s had a field site at the University of Melbourne’s Dookie campus which was sown to lucerne with biomass as the primary interest. He said the sensing capacity for nitrogen and other nutrients was developed two years ago.

The technology which takes these measurements is called hyperspectral sensing.

“The multispectral sensor has the capability of recognising by colour, based on three colours – red, green, blue – and it can be extended by using 200 – 300 colours depending on sensors,” he said.

“Some sensors inform 250 or 180 colours, the size of information we can collect is huge.

“Science-level sensors, in terms of accuracy, are the best you can get.”

The target for the accuracy of measuring biomass is more than 80 per cent. For nitrogen it is about 80 per cent, but there hasn’t been enough data collected to determine the accuracy of protein measurements.

Ground data, such as soil samples, were also collected to validate the data collected via the drone.

Accuracy will be the key benefit for farmers as well as the ability to save time. All these benefits mean better on-farm decision-making.

Prof Ryu said it could be feasible in three to five years to “easily” map 200 – 300 ha in a few hours and use that information for a variety of business decisions, such as rotating cattle for grazing and locating lost livestock.

Assoc Prof Ryu said this sort of technology was not necessarily affordable now, but he suspected in time the price would drop. like technologies such as 3-D printing and drones.

He said during his research career, the price of this sensor technology had already dropped more than 10 times.

Looking ahead, he said the biggest challenge for this type of technology was internet network speeds. He said sending “multiple gigabytes” of data back to the office from certain regions in Victoria could take days.

“It’s much quicker to send it by mail,” he said.

Prof Ryu predicted in three to five years dairy farmers would be measuring large areas of land and have the information about biomass and nutrient levels within a couple of hours.

“With a proper interface (design) the information flow could be as smooth as a weather forecast,” he said.

University of Melbourne Senior Lecturer at The School of Agriculture and Food in the Faculty of Veterinary and Agricultural Science Helen Suter was the project leader for this More Profit from Nitrogen field trial.

The focus was to look at how the soil nitrogen could be better used.

She said early findings indicated that a lot of the nitrogen used by pasture during autumn came from the soil through mineralisation, while nitrogen spread went into the soil. This was due to a combination of things including the growing conditions, soil moisture and soil-supplied nitrogen.