SOCIAL MEDIA HAS given modern war reporting a slight dilemma.

Can it help verify war stories through a screen? Or does that leave publishers wide open to falling for hoaxes and propaganda?

On one hand, coverage is no longer restricted to where a correspondent and camera crew can get to in a hostile environment.

Images and pictures from civilians on the ground can stream directly to an app on any foreign desk editor’s smartphone. Real-time content from witnesses can now be turned around into a TV package or a digital front page in minutes. No flights, security crews or flak jackets needed.

But it comes with a catch.

Photographs and videos are often the things journalists rely on as proof that an event happened in the way they reported it.

The sole source of proof used to be journalists with boots on the ground; now it’s potentially anyone, anywhere with a social media account.

The challenge for news teams is to find ways to check what they are harvesting from hashtags and social media streams is legitimate.

As a result, verification journalism has become an industry: journalists will cross check visual evidence against satellite images, weather reports, geo-located Instagram posts – anything they can access that will confirm or cast doubt on whether a video or image is of what it claims to be.

Major networks now employ verifiers to work in tandem with journalists in conflict zones.

CNN were able to capture early close range footage of Russian soldiers near Kyiv within hours of Russia announcing a ‘military operation’ in Ukraine thanks to the work of their verification team.

By analysing footage of military helicopters from Ukrainian users on social media, they were able to find the location using Google Earth after cross-referencing a distinct yellow roof seen in the video. 

❗️The man filming this video says 20-30 Russian helicopters are "bombing the airport." @nexta_tv says it's in Hostomel right outside Kyiv. Antonov airport there is used for cargo & troop transport pic.twitter.com/jzApkN71Az

— Alec Luhn (@ASLuhn) February 24, 2022

They used weather reports to help establish that the conditions seen in the video matched the date it claimed to have been shot on 24 February.

After creating a panorama of the video scene and consulting NASA’s publicly-available map of live fire information, the team found Hostomel Airport to be the likely location.

CNN then scrambled correspondent Matthew Chance to the area, where he captured some of the first proof of Russian forces landing close to Kyiv.

"Frankly, we didn't even know the Russian forces were going to be here. We assumed this was the Ukrainian forces .... We came to speak with them ... but only emerged during the conversation that they're all Russians and there are no Ukrainian military forces in sight..." pic.twitter.com/kgDkgR4O9c

— Oliver Darcy (@oliverdarcy) February 24, 2022

The dual approach of verifiers using screens in an office to report alongside correspondents on the ground has been seen across the world with the BBC in the UK and the ABC in Australia being just two examples. 

However at the New York Times, verification has moved beyond a reporting accessory to leading investigations in its own right – and they have the trophies to prove it works.

The NYT Visual Investigations Team has picked up a Pulitzer and four Emmys among other awards in the five years since its creation.

‘You can reveal indisputable facts around war crimes’ 

Irish journalist Malachy Browne led the the Pulitzer-winning team behind the story headlined ‘Russia Bombed Four Syrian Hospitals. We Have Proof.’

That particular investigation which took ‘thousands of hours’, according to Browne, used recordings of Russian radio traffic and cross-checked them with social media videos of the attack, footage from local journalists, voice messages from doctors, flight logs from plane spotters, incident reports and expert weapon analysis, alongside other sources.

The synthesised evidence was able to prove something suspected but never confirmed – that Russian forces bombed Syrian hospitals repeatedly.

By using open source intelligence – in other words, publicly available information – rather than solely relying on reporters on the ground “you can reveal indisputable facts around war crimes and other atrocities denied by governments,” Browne told The Journal. 

“There’s much more evidence and content immediately available and it’s growing in a way that can identify in the moment stories we should be pursuing.”

Earlier this month the team similar digital sleuthing methods to cut across Russian denials of civilian murders in Bucha. Satellite imagery of the Kyiv suburb, made infamous by the reports of mass civilian murders and rape, showed bodies lying in the streets when Russian soldiers were still present.

The New York Times Visual Investigation teams analysis of satellite images showing bodies on the streets of Bucha The New York Times The New York Times

Russian officials previously claimed the bodies were placed there after they had withdrawn from the area, dismissing the claims as a ‘hoax’.

The New York Times says their analysis proves this isn’t true. Satellite images prove to be particularly useful in this kind of analysis which is dependent on establishing a time frame. 

Aside from the fact they can be gathered without putting journalists in harm’s way, they can put a solid date on an event. 

“If a satellite image is taken on a particular day and then on another day the thing you’re looking at isn’t there anymore, then you have a time frame for when something happened – that’s a hard fact,” said Browne.

Satellite imagery can help establish both the when and where, which Browne describes as ‘the key journalist questions.’

He gives the team’s investigation into the 6 January Capitol Hill riot as another example of building up a timeline using open source digging. 

“The TV cameras were at a distance but some rioters were live streaming on gaming platforms. By finding those, they carried you through and you had a much more immediate sense of what was happening,” said Browne.

The footage allowed journalists to trace the paths of the rioters and to build an inside picture of events.

This, Browne says, is “really helpful from a nerdy journalist point of view” because the footage carries an timestamp on the livestream which enables him to cross reference exact moments against other footage.

“There’s an adage in journalism that chronology is your friend – that helps you build what happened on this day,” Browne said.

The challenges

Despite their ability to draw on a large net of sources to build a case, visual investigations do have limitations and can’t supplant traditional reporting entirely, according to Browne.

“If you take bodies on the street we can tell when they appeared but without autopsies we can’t exactly say what happened to them,” he explained.

“In short I think you’re best equipped if you can have access on the ground combined with forensic analysis.”

While journalists investigate alleged war crimes and conflict zones in real time using satellites and civilian footage as a paid profession, there are also those who do it in their spare time, usually for free.

People who are part of the open source intelligence community – known as OSINT – comb through hours of footage and images, pausing and stopping to painstakingly identify things like postboxes, street signs or a clothing brand to identify where it might have been taken.

There are 83,000 users in one OSINT subreddit alone, which is aimed at investigating online child abuse by locating objects such as child seats, drink bottles and baby clothes by their country of origin.

Current images of objects Europol are asking people to indentify to futher investigations into child abuse

Both Europol and the FBI have harnessed the online community through campaigns asking people to #TRACEANOBJECT seen in background of sexually explicit material of minors.

“The most innocent clues can sometimes crack a case,” reads the EUROPOL website. 

Bellingcat, one of the most famous open source investigation organisations, routinely uses volunteers to aid its investigations and is currently gathering evidence of war crimes in Ukraine.

Bellingcat's map of Civilian Harm in Ukraine Bellingcat Bellingcat

But why do people pick a hobby that exposes them to traumatic footage and involves often repetitive, boring yet complex analysis of internet flotsam and jetsam?

For Paddy Kerley it was finding a community who were “a bunch of nerds like me trying to build a picture.”

What started as being curious about North Korean weapon testing while studying engineering led to Kerley ‘learning this kind of stuff and using Google Earth to build a mental picture of what was going on.’

He works with others on various projects both in Ireland and internationally that investigates plane crashes, cyber warfare and lately the invasion of Ukraine.

Kerley examined reports from Reuters of Russian soldiers in Chernobyl disturbing radioactive dust and set about testing the claims some had been sent to hospital with acute radiation poisoning.

Using satellite footage, dating the Russian positions, expert opinion and radiation readings from people on the ground, Kerley says he’s sceptical soldiers would have been exposed long enough to cause poisoning.

“It came down to scouring for the right data and doing maths,” he told The Journal. 

“I just like to know the answer to questions, it’s that simple. I go to bed tired but I go to bed satisfied.”

He says the days leading up to the invasion “ were the most stressful” of his life, describing seeing the reports from the OSINT community showing Russian positions near the border “an existential moment of dread.”

“If I called this and I’m a moron in front of a computer in the middle of nowhere seeing what is about to happen, then what is the rest of the world doing?”

Getting things right

Kerley acknowledges some OSINT content on social media needs to be taken with a grain of salt and encourages audiences ‘to weigh evidence’ when wading into the field.

A recent scouring of Russian military officials on Instagram revealed to him the potential darkside to OSINT after realising certain tools could be used by the wrong people.

“These sites could be horrendous for stalkers but they’re useful for tracking Russian Generals,” he said.

The techniques have been used by the online community for years but when they’re used by publications like the New York Times it can be riskier considering you have the masthead’s reputation on the line rather than an anonymous user name.

Browne describes it as a “fear of getting it wrong when you have every opportunity to get it things right.”

“It’s a pressure we put on ourselves, getting the right evidence and if we don’t know something we say it,” he said.

The team draws from the publication’s depth of experience on its variety of desk as well as digital experts to analyse sound and vision to determine if it’s been manipulated and with what software.

In the humanitarian space, Amnesty International’s Evidence Lab is building a dossier of war crimes using similar tools. 

So they have verified 11 attacks on civilians, some using cluster bomb munition, using the OSINT tools of satellite imagery, videos and images of weapons remnants. 

The list is updated regularly and is expected to grow as the conflict continues. 

You can test your verification skills with our test of Irish satellite images here – see if you can identify what these buildings are from these birds-eye views. 

This work is co-funded by Journal Media and a grant programme from the European Parliament. Any opinions or conclusions expressed in this work are the author’s own. The European Parliament has no involvement in nor responsibility for the editorial content published by the project. For more information, see here.