United States

Technology Helps GPJ Reporters But Won’t Replace Them

A monthly column focusing on research, data and context in journalism.

Technology Helps GPJ Reporters But Won’t Replace Them

GLOBAL PRESS HQ — It’s tempting to assume that technology can replace local experience or knowledge. Drone data and satellite imagery inform major military decisions, so why wouldn’t they also tell stories? They nearly always look impressive, and the data is information that, in large part, wouldn’t be possible to collect otherwise.

We agree at Global Press Journal that these types of information are increasingly useful tools in our storytelling toolbox, but we differ from other news organizations in one key way: We’ll never use drone data or satellite imagery without pairing it with in-person reporting done by one of our journalists.

This policy has been confirmed many times over, as reporters collect details that help make sense of the information collected by these modern tools. Most recently, it was confirmed by a Global Press Journal reporter based in rural Democratic Republic of Congo. On a recent trip to visit reporters there, Global Press editors showed a video published by The New York Times. That video used satellite imagery to detect fires in DRC’s Ituri province, which is the home region for one of our reporters. The Times’ video matched the fires with populated areas. Then, while noting that the Times’ team wasn’t entirely certain because they had no reporters or sources on the ground, the video’s narrator deduced that the violence was linked to politics.

“Historically, these distant conflicts have been difficult to analyze,” a New York Times staffer wrote. “But new technologies allow us to investigate them in close to real time.”

As the GPJ teams in DRC and Rwanda watched the video, our reporter from Ituri province shook her head. Yes, there is violence in that part of the province, but it has little to do with politics, our reporter said. (Her village, which is not an immediate neighbor to the area highlighted by The New York Times, isn’t experiencing violence at all now, she told us.)

The people in Ituri province live many days’ drive from the country’s urban centers, she said. They’re aware of the country’s political situation, but it doesn’t inform their daily lives.

In reality, our reporter said, the fires were part of an ongoing dispute between two communities that began as a disagreement over cattle-grazing land.

What struck our reporters about the video was that the Times team repeatedly acknowledged that it hadn’t been possible for them to get a reporter to Ituri province.

Just before we showed our reporters that video, Global Press published our own project that used satellite imagery. Our Kashmiri reporter, Raihana Maqbool, traveled to the Marwah Valley, a remote part of her region, to write about a dam that the Indian government plans to build. The government says that about 1,600 people will lose their homes when the dam begins operation. But local people told Raihana that about 36,000 people would be forced to move.

With Raihana’s reporting as our foundation, we used satellite imagery to identify communities in the proposed dam’s submersion zone. Then, using to-scale topographic data, we simulated the dam, using the structure’s dimensions as published by the Indian government. When the dam fills, our simulation shows, as many as a dozen communities will go underwater. We can’t know for sure how many people live in each community, but our estimates align with what local people – not the Indian government – say will happen.

With each project, we’re building protocols for how we can interweave satellite data with our reporters’ invaluable, on-the-ground reporting. Now, we’re examining how we can use that sort of data to confirm post-segregation policies in Zimbabwe. But we’ll never publish a story based entirely on what we think we know from what is literally a bird’s-eye view of a global location.

As always, human knowledge carries far greater value in our stories than does an isolated piece of data.