Inmarsat satellite failure: what happened, and what if it happens again?

Described by industry leaders as "horrendous" timing, it disrupted one of the busiest times for the industry and left many puzzled.

It also affected the maritime and aviation sectors and growers in other countries.

While lasting just 45 hours, the outage had a profound effect given that April signals the start of winter crop sowing for plenty of farmers.

Many others were also trying to finish harvesting summer crop, spray weeds or spread fertiliser.

Growers in Queensland, NSW and Victoria were staring at more than nine million hectares of bare dirt destined for wheat, barley, canola and chickpea, but they could no longer use their machines normally.

That's because most modern tractors and combine harvesters are equipped with satellite correction and are guided to an accuracy within two centimetres, reducing wastage of chemical, fertiliser, seed and ultimately time and money.

Plenty jumped straight on the phone to their machinery dealer to find out what was going on, while others flocked to social media to ask questions of fellow farmers.

WA and SA growers' reports of being unaffected were countered by confused and disgruntled reports from easterners.

Online advice ranged from 'do you have a base station to switch over to', to 'back in my day we drove machines - take the steering wheel and be a real operator'.

Unfortunately, it's not that simple, as some planters and sprayers are solely reliant on satellites to be able to operate.

It all started when British communications satellite, Inmarsat I-4 F1, went offline at 7.14am AEST on Monday, April 17.

This satellite covers all of Australia and large parts of Asia.

Meanwhile the I-4 F2, which cuts through the middle of Australia and encompasses much of Asia, Europe and Africa, was unaffected.

Inmarsat's official explanation was the satellite suffered "a partial loss of power" and did not provide further details.

Interestingly, Russian operator of Inmarsat technology Morsviazsputnik stated in a report the power loss was "believed to be due to a fault in one of its solar arrays".

To the relief of farmers, it returned to normal operations on Wednesday morning.

But given the Australian cotton and grains industry is worth more than $27 billion, the outage raised questions about farming's reliance on technology, contingency plans, food security, and what can be done in the event of a future glitch.

Earlier this month, GRDC hosted a webinar to answer the many questions people had, inviting representatives from companies caught up in the outage - John Deere, CASE IH and Trimble - to discuss.

Also invited as an independent source, DataFarming managing director Tim Neale made the point that it wasn't a 'GPS outage', but rather a communications satellite outage.

When it comes to satellites, there are three main types - positioning (GNSS - includes GPS), communications, and imaging.

The GNSS types send signals to ag, mining and surveying equipment with the help of fixed public (government) and private (manufacturer) base stations around the country.

These signals are then sent to a processing centre to provide high accuracy positioning.

Growers can connect to that centre via the internet, but usually via a ground station connected to a comms satellite like the Inmarsat.

Another way is using real-time kinematic positioning (RTK) hardware, which receives a signal from a GNSS satellite and talks to the machine.

Growers with RTK were unaffected by the outage, but the hardware is expensive and not as easy to update.

Mr Neale said the outage not only stopped high accuracy steering, it stopped rate controllers as well.

"It caused the high accuracy signal to stop auto steering in a lot of machines," Mr Neale said.

"The other important thing here is that rate controllers on the machines were also impacted, so you couldn't just grab hold of the steering wheel, in all cases.

"You could have driven straight, but no seed or fertiliser was coming out."

CASE IH AFS product manager ANZ Sean McColley said most manufacturers allowed their systems to be switched back to 'autonomous' (an uncorrected source) in cases like this, but that wasn't always helpful.

"This will let you continue to do steering and product control, but unfortunately, some of the farming practices we're using in modern farming actually require a high accuracy GNSS positioning," Mr McColley said.

Trimble continental sales manager Darren Walford said while the outage caused plenty of headaches, it was a rare phenomenon.

"It was a very disruptive outage and it couldn't have happened at a worse time, but it's a very rare occurrence," Mr Walford said.

"Can anyone remember this happening at another time?

"Usually for the satellite based corrections, any issue lasts for minutes or hours.

"If your base station goes down [however], that can be days or weeks."

Mr Neale said there were some alternatives to communications satellites, and some growers used them during the outage.

They included connecting to a base station or network where it existed and was compatible; switching back to autonomous low accuracy, grabbing hold of the steering wheel; using controlled traffic permanent wheel tracks; and reverting controllers back to low accuracy.

All four guests agreed communications satellites were still the lowest risk and most reliable, foolproof option for delivering machinery corrections.

John Deere precision ag product manager ANZ Thomas Dewhurst also cautioned growers against spending tens of thousands of dollars on backup hardware now, especially with satellite redundancy on the way.

Mr Dewhurst said between the 1990s and 2020, Australia had three satellites covering the country, providing the east coast with redundancy and the west coast with double redundancy.

But in 2020, Inmarsat decommissioned its satellite at 178E longitude, which left just 143.5E, he said.

He said originally, Inmarsat had planned to have another satellite - 83.5E - up and functioning by the fourth quarter of 2022, but with COVID they had supply issues getting the parts required to build the satellite and that got pushed back 12 months.

Mr Dewhurst said in Q4 2022, Inmarsat launched 83.5E, which was in place during the outage but was still in the process of being commissioned.

"That should be fully functional in the next couple of months and that will bring redundancy back over the east coast of Australia, because we'll have 143.5E and we'll have 83.5E as well," he said.

"On top of that, the 64E satellite is going to be relocated back to the 178E position.

"This is going to mean that Australia is completely covered. If two satellites went down, you would still have access to correction."

Mr McColley added that Inmarsat also recently announced three new satellites for 2026.

While not suggesting the Inmarsat failure was caused by a solar storm, Mr Neale said they were behind some outages.

"It might sound really wacky that the sun impacts our communications and networks, but it does," Mr Neale said.

"Solar activity will cause solar flares, putting a lot more electromagnetic energy into the atmosphere, and that will disrupt the signals coming down to earth.

"It can certainly have a massive impact and can shut down whole communication networks.

"The good thing is that we have dramatically increased the number of satellites that are buzzing around now, so the full GNSS constellation is way more than what it was even 10 years ago. That adds a little bit more redundancy into the system."

Touching on concerns that the satellite was too old - it is 18 but has an operational life of 13 years - Mr Neale said it was not unusual for satellites to last a lot longer than what they were designed for.

Asked what would happen in the event a satellite or satellites were shot down or sabotaged, Mr Neale again reiterated there would be redundancies in place.

"We're working with a wide range of countries - there's Europeans, Japanese, Russians - there's a wide range of satellites up there now, more than just what we had in the past."

IN THE NEWS: